Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees, 38515-38612 [2015-15464]

Download as PDF Vol. 80 Monday, No. 128 July 6, 2015 Part II Department of Labor tkelley on DSK3SPTVN1PROD with PROPOSALS2 Wage and Hour Division 29 CFR Part 541 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees; Proposed Rule VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00001 Fmt 4717 Sfmt 4717 E:\FR\FM\06JYP2.SGM 06JYP2 38516 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules DEPARTMENT OF LABOR Wage and Hour Division 29 CFR Part 541 RIN 1235–AA11 Defining and Delimiting the Exemptions for Executive, Administrative, Professional, Outside Sales and Computer Employees Wage and Hour Division, Department of Labor. ACTION: Proposed rule and request for comments. AGENCY: The Fair Labor Standards Act (FLSA or Act) guarantees a minimum wage and overtime pay at a rate of not less than one and one-half times the employee’s regular rate for hours worked over 40 in a workweek. While these protections extend to most workers, the FLSA does provide a number of exemptions. The Department of Labor (Department) proposes to update and revise the regulations issued under the FLSA implementing the exemption from minimum wage and overtime pay for executive, administrative, professional, outside sales, and computer employees. This exemption is referred to as the FLSA’s ‘‘EAP’’ or ‘‘white collar’’ exemption. To be considered exempt, employees must meet certain minimum tests related to their primary job duties and be paid on a salary basis at not less than a specified minimum amount. The standard salary level required for exemption is currently $455 a week ($23,660 for a full-year worker) and was last updated in 2004. By way of this rulemaking, the Department seeks to update the salary level to ensure that the FLSA’s intended overtime protections are fully implemented, and to simplify the identification of nonexempt employees, thus making the EAP exemption easier for employers and workers to understand. The Department also proposes automatically updating the salary level to prevent the level from becoming outdated with the often lengthy passage of time between rulemakings. Lastly, the Department is considering whether revisions to the duties tests are necessary in order to ensure that these tests fully reflect the purpose of the exemption. DATES: Submit written comments on or before September 4, 2015. ADDRESSES: You may submit comments, identified by Regulatory Information Number (RIN) 1235–AA11, by either of the following methods: Electronic Comments: Submit comments through tkelley on DSK3SPTVN1PROD with PROPOSALS2 SUMMARY: VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 the Federal eRulemaking Portal https:// www.regulations.gov. Follow the instructions for submitting comments. Mail: Address written submissions to Mary Ziegler, Director of the Division of Regulations, Legislation, and Interpretation, Wage and Hour Division, U.S. Department of Labor, Room S–3502, 200 Constitution Avenue NW., Washington, DC 20210. Instructions: Please submit only one copy of your comments by only one method. All submissions must include the agency name and RIN, identified above, for this rulemaking. Please be advised that comments received will become a matter of public record and will be posted without change to https:// www.regulations.gov, including any personal information provided. All comments must be received by 11:59 p.m. on the date indicated for consideration in this rulemaking. Commenters should transmit comments early to ensure timely receipt prior to the close of the comment period as the Department continues to experience delays in the receipt of mail in our area. For additional information on submitting comments and the rulemaking process, see the ‘‘Public Participation’’ section of this document. For questions concerning the interpretation and enforcement of labor standards related to the FLSA, individuals may contact the Wage and Hour Division (WHD) local district offices (see contact information below). Docket: For access to the docket to read background documents or comments, go to the Federal eRulemaking Portal at https://www.regulations.gov. FOR FURTHER INFORMATION CONTACT: Mary Ziegler, Director of the Division of Regulations, Legislation, and Interpretation, Wage and Hour Division, U.S. Department of Labor, Room S–3502, 200 Constitution Avenue NW., Washington, DC 20210; telephone: (202) 693–0406 (this is not a toll-free number). Copies of this proposed rule may be obtained in alternative formats (Large Print, Braille, Audio Tape or Disc), upon request, by calling (202) 693–0675 (this is not a toll-free number). TTY/TDD callers may dial toll-free 1–877–889–5627 to obtain information or request materials in alternative formats. Questions of interpretation and/or enforcement of the agency’s regulations may be directed to the nearest WHD district office. Locate the nearest office by calling WHD’s toll-free help line at (866) 4US–WAGE ((866) 487–9243) between 8 a.m. and 5 p.m. in your local time zone, or log onto WHD’s Web site at https://www.dol.gov/whd/ PO 00000 Frm 00002 Fmt 4701 Sfmt 4702 america2.htm for a nationwide listing of WHD district and area offices. Electronic Access and Filing Comments Public Participation: This proposed rule is available through the Federal Register and the https:// www.regulations.gov Web site. You may also access this document via WHD’s Web site at https://www.dol.gov/whd/. To comment electronically on Federal rulemakings, go to the Federal eRulemaking Portal at https:// www.regulations.gov, which will allow you to find, review, and submit comments on Federal documents that are open for comment and published in the Federal Register. You must identify all comments submitted by including ‘‘RIN 1235–AA11’’ in your submission. Commenters should transmit comments early to ensure timely receipt prior to the close of the comment period (11:59 p.m. on the date identified above in the DATES section); comments received after the comment period closes will not be considered. Submit only one copy of your comments by only one method. Please be advised that all comments received will be posted without change to https://www.regulations.gov, including any personal information provided. SUPPLEMENTARY INFORMATION: Table of Contents I. Executive Summary II. Background A. What the FLSA Provides B. Legislative History C. Regulatory History D. Overview of Existing Regulatory Requirements III. Presidential Memorandum IV. Need for Rulemaking V. Proposed Regulatory Revisions A. Setting the Standard Salary Level B. Special Salary Tests C. Inclusion of Nondiscretionary Bonuses in the Salary Level Requirement D. Highly Compensated Employees E. Automatically Updating the Salary Levels F. Duties Requirements for Exemption VI. Paperwork Reduction Act VII. Analysis Conducted In Accordance With Executive Order 12866, Regulatory Planning and Review, and Executive Order 13563, Improving Regulation and Regulatory Review A. Introduction B. Methodology To Determine the Number of Potentially Affected EAP Workers C. Determining the Revised Salary Level Test Values D. Impacts of Revised Salary and Compensation Level Test Values E. Automatic Updates F. Duties Test Appendix A: Methodology for Estimating Exemption Status Appendix B: Additional Tables VIII. Initial Regulatory Flexibility Analysis (IRFA) E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 A. Reasons Why Action by the Agency Is Being Considered B. Statement of Objectives and Legal Basis for the Proposed Rule C. Description of the Number of Small Entities to Which the Proposed Rule Will Apply D. Projected Reporting, Recordkeeping, and Other Compliance Requirements of the Proposed Rule E. Identification to the Extent Practicable, of All Relevant Federal Rules That May Duplicate, Overlap, or Conflict With the Proposed Rule IX. Unfunded Mandates Reform Act Analysis A. Authorizing Legislation B. Assessment of Costs and Benefits C. Summary of State, Local, and Tribal Government Input D. Least Burdensome Option or Explanation Required X. Executive Order 13132, Federalism XI. Executive Order 13175, Indian Tribal Governments XII. Effects on Families XIII. Executive Order 13045, Protection of Children XIV. Environmental Impact Assessment XV. Executive Order 13211, Energy Supply XVI. Executive Order 12630, Constitutionally Protected Property Rights XVII. Executive Order 12988, Civil Justice Reform Analysis Proposed Amendments to Regulatory Text I. Executive Summary The FLSA was passed to both guarantee a minimum wage and to limit the number of hours an employee could work without additional compensation. Section 13(a)(1), which excludes certain white collar employees from minimum wage and overtime pay protections, was included in the original Act in 1938. The exemption was premised on the belief that the exempted workers earned salaries well above the minimum wage and enjoyed other privileges, including above-average fringe benefits, greater job security, and better opportunities for advancement, setting them apart from workers entitled to overtime pay. The statute delegates to the Secretary of Labor the authority to define and delimit the terms of the exemption. On March 13, 2014, President Obama signed a Presidential Memorandum directing the Department to update the regulations defining which white collar workers are protected by the FLSA’s minimum wage and overtime standards. 79 FR 18737 (Apr. 3, 2014). Consistent with the President’s goal of ensuring workers are paid a fair day’s pay for a fair day’s work, the memorandum instructed the Department to look for ways to modernize and simplify the regulations while ensuring that the FLSA’s intended overtime protections are fully implemented. Since 1940, the regulations implementing the white collar VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 exemption have generally required each of three tests to be met for the exemption to apply: (1) The employee must be paid a predetermined and fixed salary that is not subject to reduction because of variations in the quality or quantity of work performed (the ‘‘salary basis test’’); (2) the amount of salary paid must meet a minimum specified amount (the ‘‘salary level test’’); and (3) the employee’s job duties must primarily involve executive, administrative, or professional duties as defined by the regulations (the ‘‘duties test’’). One of the Department’s primary goals in this rulemaking is updating the section 13(a)(1) exemption’s salary requirements. The Department has updated the salary level requirements seven times since 1938, most recently in 2004. Under the current regulations, an executive, administrative, or professional employee must be paid at least $455 per week ($23,660 per year for a full-year worker) in order to come within the standard exemption; in order to come within the exemption for highly compensated employees (HCE), such an employee must earn at least $100,000 in total annual compensation. The Department has long recognized the salary level test as ‘‘the best single test’’ of exempt status. If left at the same amount over time, however, the effectiveness of the salary level test as a means of determining exempt status diminishes as the wages of employees entitled to overtime increase and the real value of the salary threshold falls. In order to maintain the effectiveness of the salary level test, the Department proposes to set the standard salary level equal to the 40th percentile of earnings for full-time salaried workers ($921 per week, or $47,892 annually for a full-year worker, in 2013).1 The Department is also proposing to set the highly 1 The BLS data set used to set the salary level for this rulemaking consists of earnings for full-time (defined as at least 35 hours per week) non-hourly paid employees. For the purpose of this rulemaking, the Department considers data representing compensation paid to non-hourly workers to be an appropriate proxy for compensation paid to salaried workers. The Department relied upon 2013 data in the development of the NPRM. The Department will update the data used in the Final Rule resulting from this proposal, which will change the dollar figures. If, after consideration of comments received, the Final Rule were to adopt the proposed salary level of the 40th percentile of weekly earnings, the Department would likely rely on data from the first quarter of 2016. The latest data currently available are for the first quarter of 2015, in which the 40th percentile of weekly earnings is $951, which translates into $49,452 for a full-year worker. Assuming two percent growth between the first quarter of 2015 and the first quarter of 2016, the Department projects that the 40th percentile weekly wage in the final rule would likely be $970, or $50,440 for a full-year worker. PO 00000 Frm 00003 Fmt 4701 Sfmt 4702 38517 compensated employee annual compensation level equal to the 90th percentile of earnings for full-time salaried workers ($122,148 annually). Furthermore, in order to prevent the levels from becoming outdated, the Department is proposing to include in the regulations a mechanism to automatically update the salary and compensation thresholds on an annual basis using either a fixed percentile of wages or the CPI–U. The Department is proposing to update the salary and compensation levels to ensure that the FLSA’s intended overtime protections are fully implemented and to simplify the identification of overtime-protected and exempt employees, thus making the exemptions easier for employers and workers to understand. The proposed increase to the standard salary level is also intended to address the Department’s conclusion that the salary level set in 2004 was too low to efficiently screen out from the exemption overtime-protected white collar employees when paired with the standard duties test. The Department believes that a standard salary level at the 40th percentile of all full-time salaried employees ($921 per week, or $47,892 for a full-year worker, in 2013) will accomplish the goal of setting a salary threshold that adequately distinguishes between employees who may meet the duties requirements of the EAP exemption and those who likely do not, without necessitating a return to the more detailed long duties test.2 The Department believes that the proposed salary compensates for the absence of a long test, which would have allowed employers to claim the exemption at a lower salary level, but only if they could satisfy a more restrictive duties test; moreover, it does so without setting the salary at a level that excludes from exemption an unacceptably high number of employees who meet the duties test. The Department also believes that, by reducing the number of workers for whom employers must apply the duties test to determine exempt status, this proposal is responsive to the President’s directive to simplify the exemption. Similarly, the Department believes that the proposal to set the HCE total annual compensation level at the annualized value of the 90th percentile of weekly wages of all fulltime salaried employees ($122,148 per year) will ensure that the HCE 2 From 1949 until 2004 the regulations contained two different tests for exemption—a long duties test for employees paid a lower salary, and a short duties test for employees paid at a higher salary level. E:\FR\FM\06JYP2.SGM 06JYP2 38518 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules exemption continues to cover only employees who almost invariably meet all the other requirements for exemption. Finally, the Department proposes to automatically update the standard salary and compensation levels annually to ensure that they maintain their effectiveness going forward, either by maintaining the levels at a fixed percentile of earnings or by updating the amounts based on changes in the CPI– U. The Department believes that regularly updating the salary and compensation levels is the best method to ensure that these tests continue to provide an effective means of distinguishing between overtimeeligible white collar employees and those who may be bona fide EAP employees. The Department is not making specific proposals to modify the standard duties tests but is seeking comments on whether the tests are working as intended to screen out employees who are not bona fide EAP employees; in particular, the Department is concerned that in some instances the current tests may allow exemption of employees who are performing such a disproportionate amount of nonexempt work that they are not EAP employees in any meaningful sense. In 2013, there were an estimated 144.2 million wage and salary workers in the United States, of whom the Department estimates that 43.0 million are white collar salaried employees who may be impacted by a change to the Department’s part 541 regulations. Of these workers, the Department estimates that 21.4 million are currently exempt EAP workers who are subject to the salary level requirement and may be potentially affected by the proposed rule.3 In Year 1 the Department estimates 4.6 million currently exempt workers who earn at least the current weekly salary level of $455 but less than the 40th earnings percentile ($921) would, without some intervening action by their employers, become entitled to minimum wage and overtime protection under the FLSA (Table ES1). Similarly, an estimated 36,000 currently exempt workers who earn at least $100,000 but less than the 90th earnings percentile ($122,148) per year and who meet the HCE duties test but not the standard duties test may also become eligible for minimum wage and overtime protection. In Year 10, with automatic updating of the salary levels, the Department projects that between 5.1 and 5.6 million workers will be affected by the change in the standard salary level test and between 33,000 and 42,000 workers will be affected by the change in the HCE total annual compensation test, depending on the updating methodology used (CPI–U or fixed percentile of wage earnings, respectively). Additionally, the Department estimates that an additional 6.3 million white collar workers who are currently overtime eligible because they do not satisfy the EAP duties tests and who currently earn at least $455 per week but less than the proposed salary level would have their overtime protection strengthened in Year 1 because their exemption status would be clear based on the salary test alone without the need to examine their duties. Three direct costs to employers are quantified in this analysis: (1) Regulatory familiarization costs; (2) adjustment costs; and (3) managerial costs. Assuming a 7 percent discount rate, the Department estimates that average annualized direct employer costs will total between $239.6 and $255.3 million per year, depending on the updating methodology used as shown in (Table ES1). In addition to the direct costs, this proposed rulemaking will also transfer income from employers to employees in the form of higher earnings. Average annualized transfers are estimated to be between $1,178.0 and $1,271.4 million, depending on which of the two updating methodologies analyzed in this proposal is used. The Department also projects average annualized deadweight loss of between $9.5 and $10.5 million, and notes that the projected deadweight loss is small in comparison to the amount of estimated costs. Impacts of the proposed rule extend beyond those quantitatively estimated. For example, a potential impact of the rule’s proposed increase in the salary threshold is a reduction in litigation costs. Other unquantified transfers, costs, and benefits are discussed in section VII.D.vii. TABLE ES1—SUMMARY OF REGULATORY COSTS AND TRANSFERS, STANDARD AND HCE SALARY LEVELS WITH AUTOMATIC UPDATING [Millions 2013$] Future years c Automatic updating method b Cost/Transfer a Average annualized value Year 1 Year 2 Year 10 3% Real rate 7% Real rate Affected Workers (1,000s) Standard ............................................ HCE ................................................... Percentile ............. CPI–U .................. Percentile ............. CPI–U .................. 4,646 4,646 36 36 4,747 4,634 36 35 5,568 5,062 42 33 — — — — — — — — 225.3 198.6 1,339.6 1,191.4 11.2 9.7 248.8 232.3 1,271.9 1,173.7 10.5 9.6 255.3 239.6 1,271.4 1,178.0 10.5 9.5 Costs and Transfers (Millions 2013$) Direct employer costs ........................ tkelley on DSK3SPTVN1PROD with PROPOSALS2 Transfers d .......................................... DWL ................................................... Percentile ............. CPI–U .................. Percentile ............. CPI–U .................. Percentile ............. CPI–U .................. 592.7 592.7 1,482.5 1,482.5 7.4 7.4 188.8 181.1 1,160.2 1,126.4 10.8 10.3 a Costs and transfers for affected workers passing the standard and HCE tests are combined. percentile method sets the standard salary level at the 40th percentile of weekly earnings for full-time salaried workers and the HCE compensation level at the 90th percentile. The CPI–U method adjusts both levels based on the annual percent change in the CPI–U. c These costs/transfers represent a range over the nine-year span. b The 3 White collar salaried workers not subject to the EAP salary level test include teachers, academic VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 administrative personnel, physicians, lawyers, judges, and outside sales workers. PO 00000 Frm 00004 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38519 d This is the net transfer from employers to workers. There may also be transfers of hours and income from some workers to other workers. Unquantified transfers, costs and benefits are addressed in Section VII. The Department believes that the proposed increase in the standard salary level to the 40th percentile of weekly earnings for full-time salaried workers and increasing the HCE compensation level to the 90th percentile of full-time salaried workers’ earnings, combined with annual updating, is the simplest method for securing the effectiveness of the salary level as a bright-line for ensuring that employees entitled to the Act’s overtime provisions are not exempted. The Department recognizes that the proposed standard salary threshold is lower than the historical average salary for the short duties test (the basis for the standard duties test) but believes that it will appropriately distinguish between overtime-eligible white collar salaried employees and those who may meet the EAP duties test without necessitating a return to the more rigorous long duties test. A standard salary threshold significantly below the 40th percentile, or the absence of a mechanism for automatically updating the salary level, however, would require a more rigorous duties test than the current standard duties test in order to effectively distinguish between white collar employees who are overtime protected and those who may be bona fide EAP employees. The Department believes that this proposal is the least burdensome but still cost-effective mechanism for updating the salary and compensation levels, and indexing future levels, and is consistent with the Department’s statutory obligations. tkelley on DSK3SPTVN1PROD with PROPOSALS2 II. Background A. What the FLSA Provides The FLSA generally requires covered employers to pay their employees at least the federal minimum wage (currently $7.25 an hour) for all hours worked, and overtime premium pay of one and one-half times the employee’s regular rate of pay for all hours worked over 40 in a workweek.4 However, there are a number of exemptions from the FLSA’s minimum wage and overtime requirements. Section 13(a)(1) of the FLSA, codified at 29 U.S.C. 213(a)(1), exempts from both minimum wage and overtime protection ‘‘any employee employed in a bona fide executive, administrative, or professional capacity . . . or in the capacity of outside 4 As discussed infra, the Department estimates that 128.5 million workers are subject to the FLSA and the Department’s regulations. Most of these workers are covered by the Act’s minimum wage and overtime pay protections. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 salesman (as such terms are defined and delimited from time to time by regulations of the Secretary, subject to the provisions of [the Administrative Procedure Act] . . .).’’ The FLSA does not define the terms ‘‘executive,’’ ‘‘administrative,’’ ‘‘professional,’’ or ‘‘outside salesman.’’ Pursuant to Congress’ grant of rulemaking authority, the Department in 1938 issued the first regulations at 29 CFR part 541, defining the scope of the section 13(a)(1) exemptions. Because Congress explicitly delegated to the Secretary of Labor the power to define and delimit the specific terms of the exemptions through notice and comment rulemaking, the regulations so issued have the binding effect of law. See Batterton v. Francis, 432 U.S. 416, 425 n.9 (1977). The Department has consistently used its rulemaking authority to define and clarify the section 13(a)(1) exemptions. Since 1940, the implementing regulations have generally required each of three tests to be met for the exemptions to apply: (1) The employee must be paid a predetermined and fixed salary that is not subject to reduction because of variations in the quality or quantity of work performed (the ‘‘salary basis test’’); (2) the amount of salary paid must meet a minimum specified amount (the ‘‘salary level test’’); and (3) the employee’s job duties must primarily involve executive, administrative, or professional duties as defined by the regulations (the ‘‘duties test’’). B. Legislative History Although section 13(a)(1) exempts covered employees from both the FLSA’s minimum wage and overtime requirements, its most significant impact is its removal of these employees from the Act’s overtime protections. It is widely recognized that the general requirement that employers pay a premium rate of pay for all hours worked over 40 in a workweek is a cornerstone of the Act, grounded in two policy objectives. The first is to spread employment by incentivizing employers to hire more employees rather than requiring existing employees to work longer hours, thereby reducing involuntary unemployment. See, e.g., Davis v. J.P. Morgan Chase, 587 F.3d 529, 535 (2d Cir. 2009) (‘‘The overtime requirements of the FLSA were meant to apply financial pressure to spread employment to avoid the extra wage and to assure workers additional pay to compensate them for the burden of a PO 00000 Frm 00005 Fmt 4701 Sfmt 4702 workweek beyond the hours fixed in the act.’’) (internal quotation marks omitted). The second policy objective is to reduce overwork and its detrimental effect on the health and well-being of workers. See, e.g., Barrentine v. Arkansas-Best Freight System, Inc., 450 U.S. 728, 739 (1981) (‘‘The FLSA was designed to give specific minimum protections to individual workers and to ensure that each employee covered by the Act would receive a fair day’s pay for a fair day’s work and would be protected from the evil of overwork as well as underpay.’’) (internal quotation marks and brackets omitted). Section 13(a)(1) was included in the original Act in 1938 and was based on provisions contained in the earlier National Industrial Recovery Act of 1933 (NIRA) and state law precedents. Specific references in the legislative history to the exemptions contained in section 13(a)(1) are scant. However, the exemptions were premised on the belief that the exempted workers typically earned salaries well above the minimum wage and were presumed to enjoy other privileges to compensate them for their long hours of work, such as aboveaverage fringe benefits, greater job security, and better opportunities for advancement, setting them apart from the nonexempt workers entitled to overtime pay. See Report of the Minimum Wage Study Commission, Volume IV, pp. 236 and 240 (June 1981).5 Further, the type of work exempt employees performed was difficult to standardize to any time frame and could not be easily spread to other workers after 40 hours in a week, making enforcement of the overtime provisions difficult and generally precluding the potential job expansion intended by the FLSA’s time-and-a-half overtime premium. Id. The universe of employees eligible for the exemptions has fluctuated with amendments to the FLSA. Initially, persons employed in a ‘‘local retailing capacity’’ were exempt, but Congress eliminated that language from section 13(a)(1) in 1961 when the FLSA was expanded to cover retail and service enterprises. See Public Law 87–30, 75 Stat. 65 (May 5, 1961). Teachers and 5 Congress created the Minimum Wage Study Commission as part of the Fair Labor Standards Amendments of 1977. See Sec. 2(e)(1), Public Law 95–151, 91 Stat. 1246 (Nov. 1, 1977). This independent commission was tasked with examining many FLSA issues, including the Act’s minimum wage and overtime exemptions, and issuing a report to the President and to Congress with the results of its study. E:\FR\FM\06JYP2.SGM 06JYP2 38520 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules academic administrative personnel were added to the exemption when elementary and secondary schools were made subject to the FLSA in 1966. Sec. 214, Public Law 89–601, 80 Stat. 830 (Sept. 23, 1966). The Education Amendments of 1972 made the Equal Pay provisions, section 6(d) of the FLSA, expressly applicable to employees who were otherwise exempt from the FLSA under section 13(a)(1). Sec. 906(b)(1), Public Law 92–318, 86 Stat. 235 (June 23, 1972). A 1990 enactment expanded the exemptions to include in the regulations defining exempt executive, administrative, and professional employees, computer systems analysts, computer programmers, software engineers, and similarly skilled professional workers, including those paid on an hourly basis if paid at least 61⁄2 times the minimum wage. Sec. 2, Public Law 101–583, 104 Stat. 2871 (Nov. 15, 1990). The compensation test for computer-related occupations was subsequently capped at $27.63 an hour (61⁄2 times the minimum wage in effect at the time) as part of the 1996 FLSA Amendments, when Congress enacted the new section 13(a)(17) exemption for such computer employees. Section 13(a)(17) also incorporated much of the regulatory language that resulted from the 1990 enactment. See 29 U.S.C. 213(a)(17), as added by the 1996 FLSA Amendments (Sec. 2105(a), Public Law 104–188, 110 Stat. 1755 (Aug. 20, 1996)). tkelley on DSK3SPTVN1PROD with PROPOSALS2 C. Regulatory History The FLSA became law on June 25, 1938, and the first version of part 541, setting forth the criteria for exempt status under section 13(a)(1), was issued that October. 3 FR 2518 (Oct. 20, 1938). Following a series of public hearings, which were discussed in a report issued by WHD,6 the Department published revised regulations in 1940, which, among other things, updated and expanded the salary level test. 5 FR 4077 (Oct. 15, 1940). Further hearings were convened in 1947, as discussed in a WHD-issued report,7 and revised regulations, which updated the salary levels required to meet the salary level test for the various exemptions, were issued in 1949. 14 FR 7705 (Dec. 24, 6 Executive, Administrative, Professional . . . Outside Salesman Redefined, Wage and Hour Division, U.S. Department of Labor, Report and Recommendations of the Presiding Officer (Harold Stein) at Hearings Preliminary to Redefinition (Oct. 10, 1940) (‘‘Stein Report’’). 7 Report and Recommendations on Proposed Revisions of Regulations, Part 541, by Harry Weiss, Presiding Officer, Wage and Hour and Public Contracts Divisions, U.S. Department of Labor (June 30, 1949) (‘‘Weiss Report’’). VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 1949). An explanatory bulletin interpreting some of the terms used in the regulations was published as subpart B of part 541 in 1949. 14 FR 7730 (Dec. 28, 1949). In 1954, the Department issued revisions to the regulatory interpretations of the salary basis test. 19 FR 4405 (July 17, 1954). In 1958, based on another WHD-issued report,8 the regulations were revised to update the required salary levels. 23 FR 8962 (Nov. 18, 1958). Additional changes, including periodic salary level updates, were made to the regulations in 1961 (26 FR 8635, Sept. 15, 1961), 1963 (28 FR 9505, Aug. 30, 1963), 1967 (32 FR 7823, May 30, 1967), 1970 (35 FR 883, Jan. 22, 1970), 1973 (38 FR 11390, May 7, 1973), and 1975 (40 FR 7091, Feb. 19, 1975). Revisions to increase the salary levels in 1981 were stayed indefinitely by the Department. 46 FR 11972 (Feb. 12, 1981). In 1985, the Department published an Advance Notice of Proposed Rulemaking that reopened the comment period on the 1981 proposal and broadened the review to all aspects of the regulations, including whether to increase the salary levels, but this rulemaking was never finalized. 50 FR 47696 (Nov. 19, 1985). The Department revised the part 541 regulations twice in 1992. First, the Department created a limited exception from the salary basis test for public employees, permitting public employers to follow public sector pay and leave systems requiring partial-day deductions from pay for absences for personal reasons or due to illness or injury not covered by accrued paid leave, or due to budget-driven furloughs, without defeating the salary basis test required for exemption. 57 FR 37677 (Aug. 19, 1992). The Department also implemented the 1990 law requiring it to promulgate regulations permitting employees in certain computer-related occupations to qualify as exempt under section 13(a)(1) of the FLSA. 57 FR 46744 (Oct. 9, 1992); see Sec. 2, Public Law 101–583, 104 Stat. 2871 (Nov. 15, 1990). On March 31, 2003, the Department published a Notice of Proposed Rulemaking proposing significant changes to the part 541 regulations. 68 FR 15560 (Mar. 31, 2003). On April 23, 2004, the Department issued a Final Rule (2004 Final Rule), which raised the salary level for the first time since 1975, and made other changes, some of which are discussed below. 69 FR 22122 (Apr. 8 Report and Recommendations on Proposed Revision of Regulations, Part 541, Under the Fair Labor Standards Act, by Harry S. Kantor, Presiding Officer, Wage and Hour and Public Contracts Divisions, U.S. Department of Labor (Mar. 3, 1958) (‘‘Kantor Report’’). PO 00000 Frm 00006 Fmt 4701 Sfmt 4702 23, 2004). Current regulations retain the three tests for exempt status that have been in effect since 1940: A salary basis test, a salary level test, and a job duties test. D. Overview of Existing Regulatory Requirements The regulations in part 541 contain specific criteria that define each category of exemption provided by section 13(a)(1) for bona fide executive, administrative, professional, outside sales employees, and teachers and academic administrative personnel. The regulations also define those computer employees who are exempt under section 13(a)(1) and section 13(a)(17). See §§ 541.400–.402. The employer bears the burden of establishing the applicability of any exemption from the FLSA’s pay requirements. Job titles and job descriptions do not determine exempt status, nor does paying a salary rather than an hourly rate. To qualify for the EAP exemption, employees must meet certain tests regarding their job duties and generally must be paid on a salary basis of not less than $455 per week.9 In order for the exemption to apply, an employee’s specific job duties and salary must meet all the requirements of the Department’s regulations. The duties tests differ for each category of exemption. The Department last updated the salary levels in the 2004 Final Rule, setting the standard test threshold at $455 per week for executive, administrative, and professional employees. Since its prior revision in 1975, the salary level tests had grown outdated and were thus no longer effective at distinguishing between exempt and nonexempt employees. Mindful that nearly 30 years had elapsed between salary level increases, and in response to commenter concerns that similar lapses would occur in the future, in the 2004 Final Rule the Department expressed the intent to 9 Alternatively, administrative and professional employees may be paid on a ‘‘fee basis.’’ This occurs where an employee is paid an agreed sum for a single job regardless of the time required for its completion. § 541.605(a). Salary level test compliance for fee basis employees is assessed by determining whether the hourly rate for work performed (i.e., the fee payment divided by the number of hours worked) would total at least $455 per week if the employee worked 40 hours. See § 541.605(b). Some employees, such as doctors and lawyers (§ 541.600(e)), teachers (§ 541.303(d); § 541.600(e)), and outside sales employees (§ 541.500(c)), are not subject to a salary or fee basis test. Some, such as academic administrative personnel, are subject to a special, contingent salary level. See § 541.600(c). There is also a separate salary level in effect for workers in American Samoa (§ 541.600(a)), and a special salary test for motion picture industry employees (§ 541.709). E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules ‘‘update the salary levels on a more regular basis.’’ 69 FR 22171. Under the current part 541 regulations, an exempt executive employee must be compensated on a salary basis at a rate of not less than $455 per week and have a primary duty of managing the enterprise or a department or subdivision of the enterprise. § 541.100(a)(1)–(2). An exempt executive must also customarily and regularly direct the work of at least two employees and have the authority to hire or fire, or the employee’s suggestions and recommendations as to the hiring, firing, or other change of status of employees must be given particular weight. § 541.100(a)(3)–(4). An exempt administrative employee must be compensated on a salary or fee basis at a rate of not less than $455 per week and have a primary duty of the performance of office or non-manual work directly related to the management or general business operations of the employer or the employer’s customers. § 541.200. An exempt administrative employee’s primary duty must include the exercise of discretion and independent judgment with respect to matters of significance. Id. An exempt professional employee must be compensated on a salary or fee basis at a rate of not less than $455 per week and have a primary duty of (1) work requiring knowledge of an advanced type in a field of science or learning customarily acquired by prolonged, specialized, intellectual instruction and study, or (2) work that is original and creative in a recognized field of artistic endeavor, or (3) teaching in a school system or educational institution, or (4) work as a computer systems analyst, computer programmer, software engineer, or other similarlyskilled worker in the computer field. §§ 541.300; 541.303; 541.400. An exempt professional employee must perform work requiring the consistent exercise of discretion and judgment, or requiring invention, imagination, or talent in a recognized field of artistic endeavor. § 541.300(a)(2). The salary requirements do not apply to certain licensed or certified doctors, lawyers, and teachers. §§ 541.303(d); 541.304(d). An exempt outside salesperson must be customarily and regularly engaged away from the employer’s place of business and have a primary duty of making sales, or obtaining orders or contracts for services or for the use of facilities. § 541.500. There are no salary or fee requirements for exempt outside sales employees. Id. The 2004 Final Rule created a new ‘‘highly compensated’’ test for exemption. Under the HCE exemption, VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 employees who are paid total annual compensation of at least $100,000 (which must include at least $455 per week paid on a salary or fee basis) are exempt from the FLSA’s overtime requirements if they customarily and regularly perform at least one of the exempt duties or responsibilities of an executive, administrative, or professional employee identified in the standard tests for exemption. § 541.601. The HCE exemption applies only to employees whose primary duty includes performing office or non-manual work; non-management production line workers and employees who perform work involving repetitive operations with their hands, physical skill, and energy are not exempt under this section no matter how highly paid. Id. Employees who meet the requirements of part 541 are excluded from both the Act’s minimum wage and overtime pay protections. As a result, employees may work any number of hours in the workweek and not be subject to the FLSA’s minimum wage and overtime pay requirements. Some state laws have stricter exemption standards than those described above. The FLSA does not preempt any such stricter state standards. If a State establishes a higher standard than the provisions of the FLSA, the higher standard applies in that State. See 29 U.S.C. 218. III. Presidential Memorandum On March 13, 2014, President Obama signed a Presidential Memorandum directing the Department to update the regulations defining which ‘‘white collar’’ workers are protected by the FLSA’s minimum wage and overtime standards. 79 FR 18737 (Apr. 3, 2014). The memorandum instructed the Department to look for ways to modernize and simplify the regulations while ensuring that the FLSA’s intended overtime protections are fully implemented. As the President noted at the time, the FLSA’s overtime protections are a linchpin of the middle class and the failure to keep the salary level requirement for the white collar exemption up-to-date has left millions of low-paid salaried workers without this basic protection.10 The current salary level threshold for exemption of $455 per week, or $23,660 annually, is below the poverty threshold for a family of four.11 10 https://www.whitehouse.gov/the-press-office/ 2014/03/13/fact-sheet-opportunity-all-rewardinghard-work-strengthening-overtime-pr. 11 See https://www.census.gov/hhes/www/poverty/ data/threshld/. The current salary level is less than the 10th percentile of full-time salaried workers. PO 00000 Frm 00007 Fmt 4701 Sfmt 4702 38521 Following issuance of the memorandum, the Department embarked on an extensive outreach program, conducting listening sessions in Washington, DC, and several other locations, as well as by conference call. The listening sessions were attended by a wide range of stakeholders: Employees, employers, business associations, non-profit organizations, employee advocates, unions, state and local government representatives, tribal representatives, and small businesses. In these sessions the Department asked stakeholders to address, among other issues: (1) What is the appropriate salary level for exemption; (2) what, if any, changes should be made to the duties tests; and (3) how can the regulations be simplified. Stakeholders representing employers expressed a wide variety of views on the appropriate salary level, ranging from a few who said the salary should not be raised, to several who noted their entry level managers already earned salaries far above the current annual salary level of $23,660. A number of representatives of national employers also noted regional variations in the salary levels they pay to EAP employees. Several employers encouraged the Department to consider nondiscretionary bonuses in determining whether the salary level is met, noting that such bonuses are a key part of exempt employees’ compensation in their industries and contribute to an ‘‘ownership mindset.’’ Many employer stakeholders stated that they consider first-line managerial positions to be the gateway to developing their future senior managers and organizational leadership. A number of these employer stakeholders also raised concerns about changing currently exempt employees to nonexempt employees as a result of an increase in the salary requirement, stating that employees are attached to the perceived higher status of being in exempt salaried positions, and value the time flexibility and steady income that comes with such positions. These stakeholders also stressed the need for flexibility under the regulations, in particular emphasizing the value they place on a work culture that encourages managers to lead by example and ‘‘pitch in’’ to assist nonexempt employees. They stressed that changing the duties tests to limit exempt employees’ ability to perform nonexempt work—such as California’s 50 percent primary duty rule—would negatively impact the culture of the workplace, be difficult and costly to implement, and lead to increased litigation. They also noted the significant investment they made in E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38522 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules reviewing employee classifications as a result of the 2004 Final Rule to determine whether employees met the revised duties tests. Finally, several employer representatives suggested that adding to the regulations additional examples of how the exemptions may apply to specific occupations would simplify employers’ determinations of EAP exemption status. Stakeholders representing employees universally endorsed the need to increase the salary level, noting that it has not been updated since 2004. Several employee advocates also stressed the need to index the salary level to ensure that it maintains its effectiveness as a demarcation line between exempt and overtime-eligible employees without having to rely on time consuming future rulemaking. Both individual employees and their representatives shared their concerns that some employers are taking advantage of exempt employees, requiring them to perform large amounts of routine work in order to keep down labor costs, and a few suggested that there needs to be a maximum hours cap for EAP exempt employees. They stressed that employees in ‘‘management’’ positions who are required to spend disproportionate amounts of time performing routine nonexempt tasks (ringing up customers, stocking shelves, bussing tables, cleaning stores and restaurants, etc., alongside or in place of front line workers) are not bona fide executives and do not, in fact, enjoy the flexibility and status traditionally associated with such positions and therefore are entitled to the overtime protections the FLSA was designed to provide. Employee advocates pointed to the California overtime rule as more protective of such workers. While the HCE exemption was not a primary focus of any of the listening sessions, a number of business stakeholders stated that the $100,000 total annual compensation requirement was too high, and a few suggested that the duties test for the HCE exemption should be dropped and the exemption should be based on compensation level alone. In contrast, the employee stakeholders who addressed the issue argued that the HCE duties test was too lax and that the $100,000 total annual compensation requirement was too low, particularly in light of the wage gains at the top end of the earnings spectrum since 2004. Some employee advocates suggested eliminating the HCE exemption. While the outside sales exemption was also not a central focus of the sessions, several stakeholders representing employer interests argued VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 that the distinction between inside and outside sales positions in the application of the EAP exemption does not reflect the realities of the modern workplace.12 The Department’s outreach has made clear that there are also some widespread misconceptions about overtime eligibility under the FLSA. For example, many employers and employees mistakenly believe that payment of a salary automatically disqualifies an employee from entitlement to overtime compensation irrespective of the duties performed. Many employees are also unaware of the duties required to be performed in order for the exemption to apply. Additionally, many employers seem to mistakenly believe that nonexempt white collar employees must be converted to hourly compensation. Similarly, other employers erroneously believe that they are prohibited from paying nondiscretionary bonuses to EAP employees, given that they cannot be used to satisfy the salary requirement. Some employers also mistakenly believe that the EAP regulations limit their ability to permit white collar employees to work part-time or job share.13 The Department believes that many of these misconceptions can be addressed through its education and outreach efforts.14 Lastly, the Department notes that multiple stakeholders on both sides of the issue expressed frustration with the exempt/nonexempt terminology and asked the Department to consider more descriptive terms. The Department recognizes that the terms ‘‘exempt’’ and 12 Section 13(a)(1) expressly includes within the EAP exemption ‘‘any employee employed . . . in the capacity of outside salesman.’’ 29 U.S.C. 213(a)(1). As discussed in the 2004 Final Rule, ‘‘the Administrator does not have statutory authority to exempt inside sales employees from the FLSA minimum wage and overtime requirements under the outside sales exemption.’’ 69 FR 22162. 13 As the Department has previously explained, there is no special salary level for EAP employees working less than full-time. 69 FR 22171. Employers, however, can pay white collar employees working part-time or job sharing a salary of less than the required EAP salary threshold and will not violate the Act so long as the salary equals at least the minimum wage for all hours worked and the employee does not work more than 40 hours a week. FLSA2008–1NA (Feb. 14, 2008). 14 Such misconceptions are not new. In 1940 the Department responded to the related argument that employers would convert overtime-eligible white collar employees to hourly pay instead of more secure salaries, stating: ‘‘Without underestimating the general desirability of weekly or monthly salaries which enable employees to adjust their expenditures on the basis of an assured income (so long as they remain employed), there is little advantage in salaried employment if it serves merely as a cloak for long hours of work. Further, such salaried employment may well conceal excessively low hourly rates of pay.’’ Stein Report at 7. PO 00000 Frm 00008 Fmt 4701 Sfmt 4702 ‘‘nonexempt’’ are not intuitive and can be confusing to both employers and employees. In an attempt to address this concern, the Department uses the terms ‘‘overtime protected’’ and ‘‘overtime eligible’’ at times in this NPRM as synonyms for nonexempt, and ‘‘not overtime protected’’ and ‘‘overtime ineligible’’ as synonyms for exempt. While the Department will continue to use the terms exempt and nonexempt as technical terms to ensure accuracy and continuity, we will, where appropriate, endeavor to use these more descriptive terms to aid the regulated community. The Department also uses the term ‘‘EAP exemption’’ throughout this NPRM to reflect the section 13(a)(1) exemption for executive, administrative, and professional employees. The discussions in the listening sessions have informed not just the development of this NPRM, but also the Department’s understanding of the role of overtime in the modern workplace. Some of the issues raised in the listening sessions are specifically referenced below in the Department’s proposals; some issues that were raised are either beyond the scope of this rulemaking or beyond the Department’s authority under the FLSA. For example, several employers expressed concern that employees who would become newly entitled to overtime under a higher salary level requirement would lose the flexibility they currently enjoy to work remotely on electronic devices because of employer concerns about overtime liability. Because this concern involves compensation for hours worked by overtime-protected employees, it is beyond the scope of this rulemaking. The Department, however, understands the importance of this concern and will publish a Request for Information in the near future seeking information from stakeholders on the use of electronic devices by overtimeprotected employees outside of scheduled work hours. The Department appreciates the views of all the participants in the listening sessions and welcomes further input from the public in response to this NPRM. Finally, consistent with the President’s commitment to a 21stcentury regulatory system, the Department would consider conducting a retrospective review of the Final Rule resulting from this proposal at an appropriate time in the future. IV. Need for Rulemaking One of the Department’s primary goals in this rulemaking is updating the section 13(a)(1) exemption’s salary level requirement. A salary level test has been part of the regulations since 1938 and E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules has been long recognized as ‘‘the best single test’’ of exempt status. Stein Report at 19, 42; see Weiss Report at 8–9; Kantor Report at 2–3. The salary an employer pays an employee provides ‘‘a valuable and easily applied index to the ‘bona fide’ character of the employment for which exemption is claimed’’ and ensures that section 13(a)(1) of the FLSA ‘‘will not invite evasion of section 6 and section 7 for large numbers of workers to whom the wage-and-hour provisions should apply.’’ Stein Report at 19. The 1949 Weiss Report’s statement remains true today: ‘‘The experience of [the Department] since 1940 supports the soundness of the inclusion of the salary criteria in the regulations.’’ Weiss Report at 8. In setting the salary level for the long test (which paired a lower salary with a limitation on the amount of non-exempt work an exempt worker could perform) the Department sought to provide a ready guide to assist employers in identifying employees who were unlikely to meet the duties tests for the exemptions. The salary level’s function in differentiating exempt from nonexempt employees takes on greater importance when there is only one duties test that has no limitation on the amount of nonexempt work that an exempt employee may perform, as has been the case since 2004. The Department set the standard salary level in 2004 equivalent to the former long test salary level, thus not adjusting the salary threshold to account for the absence of the more rigorous long duties test. The long test salary level was designed to operate as a ready guide to assist employers in identifying employees who were unlikely to meet the duties tests for the EAP exemption. The salary level required for exemption under section 13(a)(1) is currently $455 a week and has not been updated in more than 10 years. The annual value of the salary level ($23,660) is now lower than the poverty threshold for a family of four. If left at the same amount, the effectiveness of the salary level test as a means of helping determine exempt status diminishes as the wages of employees entitled to overtime pay increase and the real value of the salary threshold falls. By way of this rulemaking, the Department seeks to update the salary level to ensure that the FLSA’s intended overtime protections are fully implemented, and to simplify the identification of overtime-eligible employees, thus making the exemptions easier for employers and workers to understand. For similar reasons, the Department also proposes to update the total annual compensation required for VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 the HCE exemption, since it too has been unchanged since 2004, and the current level could lead to inappropriate classification given the minimal duties test for that exemption. In a further effort to respond to changing conditions in the workplace, the Department is also considering whether to allow nondiscretionary bonuses to satisfy some portion of the standard test salary requirement. Currently, such bonuses are only included in calculating total annual compensation under the HCE test, but some stakeholders have urged broader inclusion, pointing out that in some industries, particularly the retail and restaurant industries, significant portions of salaried EAP employees’ earnings may be in the form of such bonuses. The Department also proposes automatically updating the salary levels based on changes in the economy to prevent the levels from becoming outdated with the often lengthy passage of time between rulemakings. The Department proposes to automatically update the standard salary test, the annual compensation requirement for highly compensated employees, and the special salary levels for American Samoa and for motion picture industry employees, in order to ensure the continued utility of these tests over time. As explained in the Weiss Report, the salary test is only a strong measure of exempt status if it is up to date, and a weakness of the salary test is that increases in wage rates and salary levels over time gradually diminish its effectiveness. See Weiss Report at 8. In the 1970 rulemaking, in response to a comment requesting that the regulations provide for annual review and updating of the salary level, the Department noted that the idea ‘‘appears to have some merit particularly since past practice has indicated that approximately 7 years elapse between amendment of these salary requirements,’’ but concluded that such a proposal would require further study. 35 FR 884. In the 2004 Final Rule, the Department declined to adopt a process for automatically updating the salary level and instead stated our intent ‘‘in the future to update the salary levels on a more regular basis’’ as we did prior to 1975. Yet competing regulatory priorities, overall agency workload, and the timeintensive nature of the notice and comment process have hindered the Department’s ability to achieve this goal, which would require nearly continuous future rulemaking. A rule providing for automatic updates to the salary level using a methodology that has been subject to notice and comment PO 00000 Frm 00009 Fmt 4701 Sfmt 4702 38523 rulemaking would maintain the utility of the dividing line set by the salary level without the need for frequent rulemaking. This modernization of the regulations would provide predictability for employers and employees by replacing infrequent, and thus more drastic, salary level increases with gradual changes occurring at set intervals. Regular annual increases in the salary and compensation levels, instead of large changes that result from sporadic rulemaking, will provide more certainty and stability for employers. The Department is also considering revisions to the duties tests in order to ensure that they fully reflect the purpose of the exemption. Possible revisions include requiring overtimeineligible employees to spend a specified amount of time performing their primary duty (e.g., a 50 percent primary duty requirement as required under California state law) or otherwise limiting the amount of nonexempt work an overtime-ineligible employee may perform, and adding to the regulations additional examples illustrating how the exemption may apply to particular occupations. As previously discussed, during listening sessions held in advance of this proposed rule, the Department asked stakeholders what, if any, changes should be made to the existing duties tests for exemption. Stakeholders from the business community, while noting the uncertainty caused by litigation surrounding their application of the current duties tests, generally advocated for no changes to the current duties tests and raised specific concerns about the difficulty of imposing any limit on the amount of nonexempt work that exempt employees may perform. These stakeholders indicated that the uncertainty which would result from any changes in the duties tests would be much more problematic than the challenges encountered with the current tests. Employees and stakeholders representing employee interests, however, generally advocated for stricter requirements to ensure that overtime-ineligible employees spend a sufficient amount of time performing exempt duties, and do not spend excessive amounts of time on nonexempt work. These stakeholders argued that such requirements would clarify the application of the exemption and restore overtime protection to employees whose duties are not, in fact, those of a bona fide executive, administrative, or professional employee. Several business stakeholders also suggested that adding additional examples of how the exemptions apply E:\FR\FM\06JYP2.SGM 06JYP2 38524 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules to particular occupations would simplify application of the exemption for employers and increase the clarity of the current duties tests. V. Proposed Regulatory Revisions The Department’s current proposal focuses primarily on updating the salary and compensation levels by proposing that the standard salary level be set at the 40th percentile of weekly earnings for full-time salaried workers, proposing to increase the HCE annual compensation requirement to the annualized value of the 90th percentile of weekly earnings of full-time salaried workers, and proposing a mechanism for automatically updating the salary and compensation levels going forward to ensure that they will continue to provide a useful and effective test for exemption. While the primary regulatory changes proposed are in §§ 541.600 and 541.601, additional conforming changes are proposed to update references to the salary level throughout part 541 as well as to update the special salary provisions for American Samoa and the motion picture industry. The proposal also discusses the inclusion of nondiscretionary bonuses to satisfy a portion of the standard salary requirement but does not propose specific regulatory changes. Additionally, the proposal discusses the duties tests, requests comments on the current requirements, and solicits suggestions for additional occupation examples, but does not make any specific proposals for revisions to these sections. A. Setting the Standard Salary Level i. History The FLSA became law on June 25, 1938, and the first version of part 541, issued later that year, set a minimum salary level of $30 per week for executive and administrative employees. 3 FR 2518. Since 1938, the Department has increased the salary levels seven times—in 1940, 1949, 1958, 1963, 1970, 1975, and 2004. See Table A. While the Department’s method for calculating the salary level has evolved to fulfill its mandate, the purpose of the salary level requirement has remained consistent—to define and delimit the scope of the executive, administrative, and professional exemptions. 29 U.S.C. 213(a)(1). The Department has long recognized that the salary paid to an employee is the ‘‘best single test’’ of exempt status (Stein Report at 19) and that setting a minimum salary threshold provides a ‘‘ready method of screening out the obviously nonexempt employees’’ while furnishing a ‘‘completely objective and precise measure which is not subject to differences of opinion or variations in judgment.’’ Weiss Report at 8–9. The Department reaffirmed this position in the 2004 Final Rule, explaining that the ‘‘salary level test is intended to help distinguish bona fide executive, administrative, and professional employees from those who were not intended by Congress to come within these exempt categories[,]’’ and reiterating that any increase in the salary level must ‘‘have as its primary objective the drawing of a line separating exempt from nonexempt employees.’’ 69 FR 22165. TABLE A—WEEKLY SALARY LEVELS FOR EXEMPTION Long test Professional Short test (all) ........................ $50 75 95 115 140 170 ........................ ........................ $100 125 150 200 250 Date enacted Executive 1938 1940 1949 1958 1963 1970 1975 ............................................................................................................. ............................................................................................................. ............................................................................................................. ............................................................................................................. ............................................................................................................. ............................................................................................................. ............................................................................................................. Administrative $30 30 55 80 100 125 155 $30 50 75 95 100 125 155 Standard Test tkelley on DSK3SPTVN1PROD with PROPOSALS2 2004 ............................................................................................................. In 1940, the Department maintained the $30 per week salary level set in 1938 for executive employees, increased the salary level for administrative employees, and established a salary level for professional employees. The Department used salary surveys from federal and state government agencies, experience gained under the National Industrial Recovery Act, and federal government salaries to determine the salary level that was the ‘‘dividing line’’ between employees performing exempt and nonexempt work. Stein Report at 9, 20–21, 31–32. The Department recognized that the salary level falls within a continuum of salaries that overlaps the outer boundaries of exempt and nonexempt employees. Specifically, the Department stated: VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 $455 To make enforcement possible and to provide for equity in competition, a rate should be selected in each of the three definitions which will be reasonable in the light of average conditions for industry as a whole. In some instances the rate selected will inevitably deny exemption to a few employees who might not unreasonably be exempted, but, conversely, in other instances it will undoubtedly permit the exemption of some persons who should properly be entitled to the benefits of the act. Id. at 6. Taking into account the average salary levels for employees in numerous industries, and the percentage of employees earning below these amounts, the Department set the salary level for each exemption slightly below the ‘‘dividing line’’ suggested by these averages. In 1949, the Department again looked at salary data from state and federal PO 00000 Frm 00010 Fmt 4701 Sfmt 4702 agencies, including the Bureau of Labor Statistics (BLS). The data reviewed included wages in small towns and lowwage industries, earnings of federal employees, average weekly earnings for exempt employees, starting salaries for college graduates, and salary ranges for different occupations such as bookkeepers, accountants, chemists, and mining engineers. Weiss Report at 10, 14–17, 19–20. The Department noted that the ‘‘salary level adopted must exclude the great bulk of nonexempt persons if it is to be effective’’. Id. at 18. Recognizing that the ‘‘increase in wage rates and salary levels’’ since 1940 had ‘‘gradually weakened the effectiveness of the present salary tests as a dividing line between exempt and nonexempt employees,’’ the Department calculated the percentage increase in weekly E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules earnings from 1940 to 1949, and then adopted new salary levels ‘‘at a figure slightly lower than might be indicated by the data’’ in order to protect small businesses. Id. at 8, 14. The Department also cautioned that ‘‘a dividing line cannot be drawn with great precision but can at best be only approximate.’’ Id. at 11 In 1949, the Department also established a second, less-stringent duties test for each exemption, but only for those employees who were paid at or above a higher ‘‘short test’’ salary level. Those paid above the higher salary level were exempt if they also met a ‘‘short’’ duties test, which lessened the duties requirements for exemption.15 The rationale for this short test was that employees who met the higher salary level were more likely to meet all the requirements for exemption, and thus a ‘‘short-cut test for exemption . . . would facilitate the administration of the regulations without defeating the purposes of section 13(a)(1).’’ Id. at 22–23. Employees who met only the lower ‘‘long test’’ salary level, and not the higher short test salary level, were still required to satisfy the default ‘‘long’’ duties test, which included a 20 percent limitation on the amount of nonexempt work that could be performed by an exempt employee. While the long test salary level was set based on an analysis of the defined sample, the short test salary level was set in relation to the long test salary. The existence of separate short and long tests—with short test salary levels ranging from approximately 130 to 180 percent of the long test salary levels—remained part of the Department’s regulations until 2004.16 See Table A. In setting the long test salary level in 1958, the Department considered data collected during 1955 WHD investigations on the ‘‘actual salaries paid’’ to employees who ‘‘qualified for exemption’’ (i.e., met the applicable salary and duties tests), grouped by geographic region, broad industry groups, number of employees, and city size, and supplemented with BLS and Census data to reflect income increases of white collar and manufacturing employees during the period not covered by the Department’s investigations. Kantor Report at 6. The Department then set the salary level 15 These higher salary levels are presented under the ‘‘Short Test’’ heading in Table A. 16 The smallest ratio was in 1963 between the long test salary requirement for professionals ($115) and the short test salary level ($150). The largest ratio was in 1949 between the long test salary requirement for executives ($55) and the short test salary level ($100). VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 tests for exempt employees ‘‘at about the levels at which no more than about 10 percent of those in the lowest-wage region, or in the smallest size establishment group, or in the smallestsized city group, or in the lowest-wage industry of each of the categories would fail to meet the tests.’’ Id. at 6–7. In other words, the Department set the salary level so that only a limited number of workers performing EAP duties (about 10 percent) in the lowestwage regions and industries would fail to meet the salary level test and therefore be overtime protected. In laying out this methodology, the Department echoed comments from the Weiss Report that the salary tests ‘‘simplify enforcement by providing a ready method of screening out the obviously nonexempt employees[,]’’ and that ‘‘[e]mployees that do not meet the salary test are generally also found not to meet the other requirements of the regulations.’’ Id. at 2–3. The Department also noted that in our experience misclassification of overtime-protected employees occurs more frequently when the salary levels have ‘‘become outdated by a marked upward movement of wages and salaries.’’ Id. at 5. The Department followed a similar methodology when determining the appropriate long test salary level increase in 1963, using data regarding salaries paid to exempt workers collected in a 1961 WHD survey. 28 FR 7002. The salary level for executive and administrative employees was increased to $100 per week, for example, when the 1961 survey data showed that 13 percent of establishments paid one or more exempt executives less than $100 per week, and 4 percent of establishments paid one or more exempt administrative employees less than $100 a week. 28 FR 7004. The professional exemption salary level was increased to $115 per week, when the 1961 survey data showed that 12 percent of establishments surveyed paid one or more professional employees less than $115 per week. Id. The Department noted that these salary levels approximated the same percentages used in 1958: Salary tests set at this level would bear approximately the same relationship to the minimum salaries reflected in the 1961 survey data as the tests adopted in 1958, on the occasion of the last previous adjustment, bore to the minimum salaries reflected in a comparable survey, adjusted by trend data to early 1958. At that time, 10 percent of the establishments employing executive employees paid one or more executive employees less than the minimum salary adopted for executive employees and 15 percent of the establishments employing administrative or professional employees PO 00000 Frm 00011 Fmt 4701 Sfmt 4702 38525 paid one or more employees employed in such capacities less than the minimum salary adopted for administrative and professional employees. Id. The Department continued to use a similar methodology when updating the long test salary level in 1970. After examining data from 1968 WHD investigations, 1969 BLS wage data, and information provided in a report issued by the Department in 1969 that included salary data for executive, administrative, and professional employees,17 the Department increased the long test salary level for executive employees to $125 per week when the salary data showed that 20 percent of executive employees from all regions and 12 percent of executive employees in the West earned less than $130 a week. 35 FR 884–85. The Department also increased the long test salary levels for administrative and professional employees to $125 and $140, respectively. In 1975, instead of following these prior approaches, the Department set the long test salary levels based on increases in the Consumer Price Index (CPI), although the Department adjusted the salary level downward ‘‘in order to eliminate any inflationary impact.’’ 40 FR 7091. As a result of this recalibration of the 1970 levels, the long test salary level for the executive and administrative exemptions was set at $155, while the professional level was set at $170. The salary levels adopted were intended as interim levels ‘‘pending the completion and analysis of a study by [BLS] covering a six month period in 1975[,]’’ and were not meant to set a precedent for future salary level increases. Id. at 7091–92. Although the Department intended to increase the salary levels after completion of the BLS study of actual salaries paid to employees, the envisioned process was never completed, and the ‘‘interim’’ salary levels remained unchanged for the next 29 years. As reflected in Table A, the short test salary level increased in tandem with the long test level throughout the various rulemakings since 1949. Because the short test was designed to capture only those white collar employees whose salary was high enough to indicate a stronger likelihood of exempt status and thus warrant a less stringent duties requirement, the short test salary level was always set significantly higher than the long test 17 Earnings Data Pertinent to a Review of the Salary Tests for Executive, Administrative and Professional Employees As Defined in Regulations Part 541, (1969), cited in 34 FR 9935. E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38526 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules salary level. Thus, in 1975 while the long test salary levels ranged from $155 to $170, the short test level was $250. The salary level test was most recently updated in 2004, when the Department abandoned the concept of separate long and short tests, opting instead for one ‘‘standard’’ test, and set the salary level under a new standard duties test at $455 for executive, administrative, and professional employees. Due to the lapse in time between the 1975 and 2004 rulemakings, the salary threshold for the long duties tests (i.e., the lower salary level) did not reflect salaries being paid in the economy and had become ineffective at distinguishing between overtime-eligible and overtimeineligible white collar employees. For example, at the time of the 2004 Final Rule, the salary levels for the long duties tests were $155 for executive and administrative employees and $170 for professional employees, while a fulltime employee working 40 hours per week at the federal minimum wage ($5.15 per hour) at that time earned $206 per week. 69 FR 22164. Even the short test salary level at $250 per week was not far above the minimum wage. The Department in the 2004 Final Rule based the new ‘‘standard’’ duties tests on the short duties tests (which did not limit the amount of nonexempt work that could be performed), and tied them to a single salary test level that was updated from the long test salary (which historically had been paired with a cap on nonexempt work). 69 FR 22164, 22168–69; see also 68 FR 15570 (‘‘Under the proposal, the minimum salary level to qualify for exemption from the FLSA minimum wage and overtime requirements as an executive, administrative, or professional employee would be increased from $155 per week to $425 per week. This salary level would be referred to as the ‘standard test,’ thus eliminating the ‘short test’ and ‘long test’ terminology. The separate, higher salary level test for professional employees also would be eliminated.’’). The Department concluded that it would be burdensome to require employers to comply with a more complicated long duties test given that the passage of time had rendered the long test salary level largely obsolete. 69 FR 22164; 68 FR 15564–65. The Department believed at the time that the new standard test salary level accounted for the elimination of the long duties test. 69 FR 22167. In determining the new salary level in 2004, the Department reaffirmed our oftrepeated position that the salary level is the ‘‘best single test’’ of exempt status. 69 FR 22165. Consistent with prior VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 rulemakings, the Department relied on actual earnings data and set the salary level near the lower end of the current range of salaries. Specifically, the Department used Current Population Survey (CPS) data that encompassed most salaried employees, and set the salary level to exclude roughly the bottom 20 percent of these salaried employees in each of the subpopulations: (1) The South and (2) the retail industry. Although several prior salary levels were based on salaries of approximately the lowest 10 percent of exempt salaried employees (the Kantor method), the Department stated that the change in methodology was warranted in part to account for the elimination of the short and long duties tests, and because the utilized data sample included nonexempt salaried employees, as opposed to only exempt salaried employees. However, as the Department acknowledged, the salary arrived at by this method was, in fact, equivalent to the salary derived from the Kantor method. 69 FR 22168. Based on the adopted methodology, the Department ultimately set the salary level for the new standard test at $455 per week. In the 2004 Final Rule the Department also created a test for highly compensated employees, which provided a minimal duties test for workers within the highest compensation range. Reasoning that an especially high salary level negated the need for a probing duties analysis, the Department provided that employees who earned at least $100,000 in total annual compensation (of which at least $455 was paid weekly on a salary or fee basis) were covered by the exemption if they customarily and regularly spent time on one or more exempt duties, and were not engaged in manual work. 69 FR 22172. In summary, the regulatory history reveals a common methodology used, with some variations, to determine appropriate salary levels. In almost every case, the Department examined a broad set of data on actual wages paid to salaried employees and then set the salary level at an amount slightly lower than might be indicated by the data. In 1940 and 1949, the Department looked to the average salary paid to the lowest level of exempt employees. Beginning in 1958, the Department set salary levels to exclude approximately the lowest-paid 10 percent of exempt salaried employees in low-wage regions, employment size groups, city size, and industry sectors, and we followed a similar methodology in 1963 and 1970. The levels were based on salaries in low-wage categories in order to protect PO 00000 Frm 00012 Fmt 4701 Sfmt 4702 the ability of employers in those areas and industries to utilize the exemptions and in order to mitigate the impact of higher-paid regions and sectors. In 1975, the Department increased the salary levels based on changes in the CPI, adjusting downward to eliminate any potential inflationary impact. 40 FR 7091 (‘‘However, in order to eliminate any inflationary impact, the interim rates hereinafter specified are set at a level slightly below the rates based on the CPI.’’). In 2004, the Department raised the salary level to $455 per week using earnings data of full-time salaried employees (both exempt and nonexempt) in the South and in the retail sector. As in the past, the use of lower-salary data sets was intended to accommodate those businesses for which salaries were generally lower due to geographic or industry-specific reasons. This most recent revision eliminated the short and long duties requirements in favor of a standard duties test for each exemption and a single salary level for executive, administrative, and professional employees. Between 1938 and 1975, the Department increased the salary level every five to nine years. Following the 1975 rulemaking, however, 29 years passed before the salary level was again raised. In the 2004 Final Rule, the Department expressed a commitment to updating the salary levels ‘‘on a more regular basis,’’ particularly when ‘‘wage survey data and other policy concerns support such a change.’’ 69 FR 22171. Regular updates to the salary level test are imperative to ensuring that the salary level does not become obsolete over time, and providing predictability for employers and employees. Not only does the annualized current salary level of $23,660 a year not reflect increases in nationwide salary levels since 2004, but this figure, as noted above, is below the 2014 poverty threshold of $24,008 per year for a family of four.18 Moreover, since the salary level test was last increased in 2004, the federal minimum wage has increased three times, from $5.15 to the current rate of $7.25 an hour,19 raising the wages of overtimeprotected employees. The absence of an 18 The 2014 poverty threshold for a family of four with two related people under 18 in the household. Available at: https://www.census.gov/hhes/www/ poverty/data/threshld/. 19 The U.S. Troop Readiness, Veterans’ Care, Katrina Recovery, and Iraq Accountability Appropriations Act, 2007, Public Law 110–28, 121 Stat. 112 (Mary 25, 2007), included an amendment to the FLSA that increased the applicable Federal minimum wage under section 6(a) of the FLSA in three steps: To $5.85 per hour effective July 24, 2007; to $6.55 per hour effective July 24, 2008; and to $7.25 per hour effective July 24, 2009. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules increase in the salary level when combined with past (and future) increases to the minimum wage further undermines the effectiveness of the salary level to serve as a line of demarcation between overtimeprotected and exempt workers. Mindful of such developments, the Department proposes to increase the salary level annually to ensure the test’s ability to serve as an effective dividing line between exempt and nonexempt employees. ii. Purpose of the Salary Level Requirement The Department has long recognized that the line of demarcation between the salaries of white collar employees who are overtime-protected and those who are exempt EAP employees cannot be reduced to a standard formula. There will always be white collar overtimeeligible employees who are paid above the salary threshold, and employees performing EAP duties who are paid below the salary threshold. The salary level selected will inevitably affect the number of workers falling into each of these categories. As the Department has noted: Inevitably, if the salary tests are to serve their purpose in a situation where salaries and wages have risen, some employees who have been classified as exempt under the present salary tests will no longer be within the exemption under any new tests adopted. Such employees include some whose status in management or the professions is questionable in view of their low salaries. Also included in the group who would not be exempt are employees whose exempt status, on the basis of their duties and responsibilities, is questionable. tkelley on DSK3SPTVN1PROD with PROPOSALS2 Kantor Report at 5. Historically, when setting the lower, long test salary level, the Department strived to ensure that the salary threshold reasonably served to reduce instances where obviously overtime-protected white collar employees were classified as exempt, while avoiding undue exclusions from exemption of employees performing bona fide executive, administrative, and professional duties. In 1949, the Department noted: Regulations of general applicability such as these must be drawn in general terms to apply to many thousands of different situations throughout the country. In view of the wide variation in their applicability the regulations cannot have the precision of a mathematical formula. The addition to the regulations of a salary requirement furnishes a completely objective and precise measure which is not subject to differences of opinion or variations in judgment. The usefulness of such a precise measure as an aid in drawing the line between exempt and nonexempt employees, particularly in borderline cases, seems . . . to be established beyond doubt. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 Weiss Report at 9. Since 1958, the Department’s approach has emphasized minimizing the number of white collar employees performing bona fide EAP duties who are excluded from the exemption by the salary level. This approach was appropriate when there was a long duties test with a specific cap on the amount of time that overtime-ineligible employees could spend performing nonexempt work. However, this approach is not effective in the absence of that limitation, as it does not take into sufficient account the inefficiencies (in terms of the administrative costs of classifying positions) of applying the duties test to large numbers of overtime-eligible white collar employees and the possibility of misclassification of those employees as exempt (and possible litigation costs associated with misclassification). A thorough review of the regulatory history of the seven previous increases to the salary levels reveals an essentially common methodology to determine the appropriate level, which has been refined periodically in order to better meet the salary level test’s goals. In almost every case, the Department considered a broad set of salary data and then set the salary level at an amount slightly lower than the dividing line between exempt and nonexempt that might be indicated by the data, or otherwise set it ‘‘at points near the lower end of the current range of salaries for each of the [EAP] categories.’’ Kantor Report at 5. The exact line of demarcation set by the Department, however, has varied, and is guided by practical considerations that allow it to best serve the underlying principles of the exemption, that is, to differentiate exempt and nonexempt white collar employees. With that objective in mind, the Department proposes to increase the minimum salary level required to qualify for the EAP exemptions from $455 per week to the 40th percentile of weekly earnings for full-time salaried workers ($921 per week).20 This proposed methodology is conceptually similar to the methodology utilized by the Department in the 2004 Final Rule, which in turn was largely modeled on the salary level methodology first set forth in the Kantor Report in 1958 and used by the Department in nearly every salary level rulemaking thereafter. See BLS sample used for this rulemaking consists of usual weekly earnings for full-time (defined as at least 35 hours per week) non-hourly paid employees. For the purpose of this rulemaking, the Department considers data representing compensation paid to non-hourly workers to be an appropriate proxy for compensation paid to salaried workers. PO 00000 20 The Frm 00013 Fmt 4701 Sfmt 4702 38527 69 FR 22167–68; Kantor Report at 6–7. Both the proposed methodology and its predecessors set the salary level based on a percentile of the salaries actually paid to a specified pool of salaried employees. iii. Sources for the Salary Level Requirement After a careful review of the guidance articulated in the Department’s previous part 541 rulemakings, and observing more than a decade of experience since the 2004 salary level test update, the Department has chosen to rely on the general methodology used in every previous update except 1975, with a few changes designed to simplify and improve the methodology as a tool for differentiating exempt and nonexempt workers. Specifically, in the interest of making the salary methodology simpler and more transparent, the Department is using nationwide CPS data on full-time salaried employees (both exempt and nonexempt) to set the proposed salary level. As discussed infra, the Department is not further modifying the sample as we did in 2004. See 69 FR 22168.21 This is not the first time the Department has modified the methodology, in part because the specific sources of the Department’s data have changed over the years. In 1940, the Department considered salary surveys by government agencies, experience under the NIRA, state laws, and federal government salaries. Stein Report at 9, 20–21, 31–32. In 1949, the Department looked at salary data collected by state and federal agencies, including the BLS, and considered wages in small towns and low-wage industries, earnings of federal employees, average weekly earnings for exempt employees, wages of clerical employees, and starting salaries for college graduates. Weiss Report at 10, 13–20. In 1958, the Department used a data set that consisted of data collected during WHD investigations on actual salaries paid to employees who qualified for the exemption, grouped by geographic region, broad industry groups, number of employees, and size of city, and the Department supplemented the investigation data with BLS and Census data on the income increases of white collar and 21 As discussed infra, the CPS data on full-time salaried workers which the Department is now proposing to use excludes certain groups, such as the self-employed, unpaid volunteers, workers under age 16, and members of the military on active duty. However, BLS automatically excludes these groups when it generates the sample. In 2004, the Department took additional steps to exclude other categories of workers from the sample. E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38528 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules manufacturing employees for the period not covered by the Department’s investigations. Kantor Report at 6–9. Subsequent salary level updates in 1963 and 1970 followed a similar approach, looking to WHD data on actual salaries paid to exempt employees and augmenting the 1970 analysis with BLS data. 28 FR 7002; 35 FR 884. The Department diverged from our practice of looking to actual salary data in the 1975 rule, when the Department increased the salary levels set in 1970 based on the CPI and adjusted slightly ‘‘in order to eliminate any inflationary impact’’; those salary levels, however, were intended to be ‘‘interim’’ levels, pending receipt and review of data on actual salary levels. 40 FR 7091. The Department made some adjustments in 2004 to broaden the data set used, rather than continuing to rely upon WHD’s limited enforcement data. The Department continued to carefully review actual salary levels, but did so by using the CPS as the data source. The CPS is a large, statistically robust survey jointly administered by the Census Bureau and BLS, and it is widely used and cited by industry analysts. It surveys 60,000 households a month, covering a nationally representative sample of workers, industries, and geographic areas.22 Households are surveyed for four months, excluded from the survey for eight months, surveyed for an additional four months, then permanently dropped from the sample. During months 4 and 16 in the sample (the outgoing rotation months), employed respondents complete a supplementary questionnaire (the merged outgoing rotation group or MORG) in addition to the regular survey, which contains the detailed information on earnings necessary to estimate a worker’s exemption status. However, because the Department was unable to precisely identify which workers would qualify for the exemption, the Department based the salary level in the 2004 Final Rule on a pool of employees that generally included those full-time salaried employees covered by the FLSA and by the part 541 regulations. Where possible, the Department excluded from our analysis workers who were excluded entirely from the FLSA’s overtime requirements or from the salary tests.23 69 FR 22167–68. The 22 https://www.census.gov/cps; https:// www.census.gov/cps/methodology. 23 The 2004 pool of salaried employees excluded: (1) The self-employed, unpaid volunteers and religious workers who are not covered by the FLSA; (2) agricultural workers, certain transportation workers, and certain automobile dealership employees who are exempt from overtime under VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 Department concluded that it was preferable to move away from using a sample limited to exempt salaried employees, as was done in the Kantor method, because in order to create such a pool of likely-exempt salaried employees one would have to rely upon ‘‘uncertain assumptions regarding which employees are actually exempt.’’ Id. at 22167. In addition, the Department used CPS data rather than salary data from the limited pool of our own investigations because there would have been too few observations from these investigations to yield statistically meaningful results. In this proposed rule, the Department continues to adhere to the basic methodological principle of looking to actual salaries paid to employees, but as in the 2004 rulemaking, the Department has reexamined the precise contours of the sample to ensure that it is as transparent, accessible, and easily replicated as possible. By moving to an even more standardized sample than the one used in 2004—the proposed rule includes all full-time salaried employees nationwide, without exclusions—the Department seeks to further improve upon the methodology. The proposed rule uses CPS data comprising all full-time salaried employees to determine the proposed salary levels, and the Department is not further restricting the sample. Inclusion of those employees previously excluded by the Department in 2004 achieves a more robust sample that is more representative of salary levels throughout the economy. For example, while teachers, physicians, lawyers, outside sales employees, and federal employees were excluded from the 2004 sample because they are not subject to the part 541 salary level test, they nonetheless are part of the universe of salaried employees and, as such, their salaries shed light on the salaries paid to employees performing exempt EAP duties. Furthermore, replicating this sample from the CPS public-use files would require no adjustments, making it easier for members of the public to access it and use it.24 In contrast, the other provisions of the Act; (3) teachers, academic administrative personnel, certain medical professionals, outside sales employees, lawyers and judges who are not subject to the part 541 salary tests; and (4) federal employees who are not subject to the part 541 regulations. 69 FR 22168. 24 The Department notes that the public will not be able to exactly replicate the weekly earnings and percentiles used in this NPRM from the public-use data files made available by BLS. As with all BLS data, to ensure the confidentiality of survey respondents, data in the public-use files use adjusted weights and therefore minor discrepancies between internal BLS files and public-use files exist. BLS publishes quarterly the earnings deciles PO 00000 Frm 00014 Fmt 4701 Sfmt 4702 sample from the 2004 rulemaking required filtering out various employees based on interpretations of a number of statutory and regulatory exclusions from coverage or the salary requirement—a process that is inconsistent with the simplification, streamlining, and transparency objectives of the current rulemaking. Using a broader sample does not diminish the soundness of the ultimate salary level derived. As the Department noted with respect to our change in the sample for the 2004 rulemaking, different ‘‘approaches are capable of reaching exactly the same endpoint [i.e., a percentile that accomplishes the purpose of the salary level test].’’ 69 FR 22167. iv. Setting the Required Salary Level In addition to looking to a lessrestricted sample, this proposed rule also differs from the 2004 Final Rule in that the Department proposes to set the standard salary level at a higher percentile of the salary distribution and relies upon salaries nationwide rather than salaries in a limited geographic area or industry. The Department is also proposing to set the salary level as a percentile of weekly earnings of fulltime salaried workers rather than a specific dollar amount because we believe a percentile serves as a better proxy for distinguishing between overtime-eligible and exempt white collar workers as it is rooted in the relative distribution of earnings which are linked to the type of work undertaken by salaried workers. The proposed standard salary level of the 40th percentile of weekly earnings for all full-time salaried employees is higher than the percentile used by the Department in either the 2004 Final Rule or the Kantor method. In the 2004 Final Rule, the Department set the required standard salary level at approximately the 20th percentile of salaried employees in the South region and in the retail industry, and in 1958, using the Kantor method which had both the long and short tests, the Department set the required salary level at approximately the 10th percentile of exempt EAP workers’ salaries in lowwage regions, employment size groups, city size, and industries. As explained in the 2004 Final Rule, those two methods produced roughly equivalent salary levels when taking into account their differing samples. See 69 FR 22167–68; Kantor Report at 6. Applying of full-time salaried workers on which the Department relies to set the proposed salary level at https://www.bls.gov/cps/research_series_ earnings_nonhourly_workers.htm. E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules these methods today would result in salary levels of $577 per week (2004 method) or $657 per week (Kantor method), which would equate to approximately the 15th and 20th percentiles of weekly earnings for all full-time salaried workers. However, the higher percentile proposed here is necessary to correct for the current pairing of a salary based on the lower salary long test with a duties test based on the less rigorous short duties test, and ensure that the proposed salary is consistent with the Department’s longstanding goal of finding an appropriate line of demarcation between exempt and nonexempt employees. See, e.g., Weiss Report at 11 (‘‘The salary tests in the regulations are essentially guides to help in distinguishing bona fide executive, administrative, and professional employees from those who were not intended by the Congress to come within these categories.’’). Currently, approximately 85 percent of white collar salaried workers who fail the EAP duties test earn at least $455 per week. Because the current salary level is only screening from exemption approximately 15 percent of overtimeeligible white collar salaried employees, it is not an effective test for exemption and does not serve the intended purpose of simplifying application of the exemption by reducing the number of employees for whom employers must perform a duties analysis. Increasing the standard salary level to the 40th percentile of weekly earnings for fulltime salaried workers would reduce by 6.3 million the number of white collar employees whose exemption status currently can only be determined by applying the duties test.25 Conversely, only approximately 4 percent of all white collar salaried employees who meet the duties test earn less than the current salary level. The proposed increase in the standard salary level would increase the number of overtimeeligible white collar salaried employees who meet the duties test and earn less than the proposed salary level to approximately 25 percent. The proposed percentile diverges from the percentiles adopted in both the 2004 Final Rule and the Kantor method because it more fully accounts for the Department’s elimination of the long duties test. As discussed in detail below, the Department acknowledged in the 2004 Final Rule that it was 25 These workers are salaried, white collar workers who do not satisfy the EAP duties tests and who earn at least $455 per week but less than the proposed salary level. Some workers in this group may be overtime ineligible due to another non-EAP exemption. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 necessary in setting the salary level to account for the shift to a single standard duties test that was equivalent to the less rigorous short duties test. The Department intended the change from the 10th to the 20th percentile to address, in part, the elimination of the long duties test. 69 FR 22167. The Department also intended this change, however, to account for the use of a different data set. 69 FR 22168. Based on further consideration of our analysis of the 2004 salary, the Department has now concluded that the $455 salary level did not adequately account for both the shift to a sample including all salaried workers covered by the part 541 regulations, rather than just EAP exempt workers, and the elimination of the long duties test that had historically been paired with the lower salary level. Accordingly, this proposal is intended to correct for that error by setting a salary level that fully accounts for the fact that the standard duties test is significantly less rigorous than the long duties test and, therefore, the salary threshold must play a greater role in protecting overtime-eligible employees. This proposal is also responsive to the President’s desire to simplify the exemption, and it addresses the Department’s concern that overtimeeligible workers may be misclassified as exempt based solely on the salaries they receive. This is the first time that the Department has needed to correct for such a mismatch between the existing salary level and the applicable duties test. Under the old short test/long test structure, the Department routinely focused on setting a long test salary level that would minimize the number of employees performing bona fide EAP duties deemed overtime-eligible based on their salaries (keeping the number of such excluded employees to about 10 percent of those who qualified for exemption based upon their duties). This approach was possible because the long duties test included a limit on the amount of nonexempt work that could be performed and thus provided an adequate safeguard against the exemption of white collar workers who should be overtime-protected but who exceeded the salary level. The creation of a single standard test based on the less rigorous short duties test caused new uncertainty as to what salary level is sufficient to ensure that employees intended to be overtime-protected are not subject to inappropriate classification as exempt, while minimizing the number of employees disqualified from the exemption even PO 00000 Frm 00015 Fmt 4701 Sfmt 4702 38529 though their primary duty is EAP exempt work. A brief history of the long duties test illustrates the importance of offsetting its elimination with a corresponding increase in the salary level. The socalled long test was the sole test for all employees until 1949. The Department devised a separate short test in 1949 to supplement the long test with a shortcut, more permissive, method for determining exempt status for only those employees meeting a higher salary requirement. For example, the long duties test in effect from 1949 to 2004 for administrative employees required that an exempt employee: (1) Have a primary duty consisting of the performance of office or non-manual work directly related to management policies or general business operations of the employer or the employer’s customers; (2) customarily and regularly exercise discretion and independent judgment; (3) regularly and directly assist a proprietor or a bona fide executive or administrative employee, or perform under only general supervision work along specialized or technical lines requiring special training, experience, or knowledge, or execute under only general supervision special assignments and tasks; and (4) not devote more than 20 percent (or 40 percent in a retail or service establishment) of hours worked in the workweek to activities that are not directly and closely related to the performance of the work described above. 29 CFR 541.2 (2003). By contrast, the short duties test in effect during the 1949 to 2004 period provided that an administrative employee paid at or above the short test salary level qualified for exemption if the employee’s primary duty consisted of the performance of office or non-manual work directly related to management policies or general business operations of the employer or the employer’s customers which includes work requiring the exercise of discretion and independent judgment. Id. Between 1949 and 2004, employers were only able to claim the exemption based on the less-stringent short duties test for employees who were paid a specified higher salary level. The Department reasoned that, ‘‘in the categories of employees under consideration the higher the salaries paid the more likely the employees are to meet all the requirements for exemption, and the less productive are the hours of inspection time spent in analysis of the duties performed.’’ Weiss Report at 22. The original, more thorough duties test became known as the long test, and remained for decades E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38530 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules the test employers were required to satisfy for those employees whose salary was insufficient to meet the higher short test salary level. Apart from the differing salary requirements, the most significant difference between the short test and the long test was the long test’s limit on the amount of time an exempt employee could spend on nonexempt duties while allowing the employer to claim the exemption. For all three EAP exemptions, the long duties test imposed a limit on nonexempt duties. A bright-line, 20 percent cap on nonexempt work was instituted in 1940 for executive and professional employees, and in 1949 for administrative employees.26 The short duties tests did not include a limitation on nonexempt work because employees paid the higher short test salary level were likely to ‘‘meet all of the requirements of the Administrator’s basic definitions of exempt employees, including the requirements with respect to nonexempt work.’’ Weiss Report at 23. The Department reasoned that if the test were to exempt those for whom ‘‘the nonexempt work is substantial,’’ this would be ‘‘contrary to the objectives of the Fair Labor Standards Act.’’ Id. at 33. In 2004 the Department discontinued the use of the long duties test because it had effectively become dormant due to the passage of time since the required salary level had last been raised in 1975, and because the Department believed that reinstituting it would be administratively burdensome. Instead the Department essentially adopted the short duties tests as the standard duties tests, stating that the new standard duties tests ‘‘are substantially similar to the current short duties tests,’’ 69 FR 22214, and that ‘‘it is impossible to quantitatively estimate the number of exempt workers resulting from the de minimis differences in the standard duties tests compared to the current short duties tests.’’ Id. at 22192–93. The Department recognized the need to adjust the salary percentile previously used to set the long test salary level upward to account for the transition to a single more lenient duties test. Indeed, the Department stated that the increase to the 20th percentile instead of the 10th percentile was intended to account for two changes made in 2004: ‘‘because of the proposed change from the ‘short’ and ‘long’ test structure and because the data included nonexempt salaried employees.’’ 69 FR 22167; see 68 FR 26 By statute, beginning in 1961, retail employees could spend up to 40 percent of their hours worked performing nonexempt work and still be found to meet the duties tests for EAP exemption. 29 U.S.C. 213(a)(1). VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 15571. However, although the Department recognized the need to make an adjustment because of the elimination of the long duties test, the amount of the increase in the required salary actually only accounted for the fact that the data set used to set the salary level included nonexempt workers while the Kantor method considered only the salaries paid to exempt employees. As the data tables in the 2004 Final Rule show, a salary of $455 excluded from the exemption 20.2 percent of all salaried employees in the South and 20.0 percent of all salaried employees in retail. 69 FR 22169, Table 3. However, that same $455 salary level excluded only 8.2 percent of likely exempt employees in the South and 10.2 percent of likely exempt employees in retail. 69 FR 22169, Table 4. In other words, ‘‘by setting a salary level excluding from the exemptions approximately the lowest 20 percent of all salaried employees, rather than the Kantor report’s 10 percent of exempt employees,’’ the Department in 2004 actually adopted a percentile that produced a salary amount roughly equivalent to the long test salary yielded at the 10th percentile using the Kantor method’s data set. Id. at 22168 (emphases in original). The Department had not, in fact, made any additional adjustment to account for the elimination of the long duties test. Thus, although the Department had identified the need to adjust the required salary percentile to account for the elimination of the long duties test, the Department effectively paired the short test’s less stringent duties requirements with the lower salary level historically associated with the long duties test.27 The long duties tests had limited the amount of nonexempt work that could be performed by employees for whom the employer claimed the EAP exemption; only employees who were paid the higher short test salary level were not required to meet the nonexempt duties caps. Because the standard duties tests do not contain a cap on the amount of nonexempt work that may be performed, after the 2004 rulemaking the salary level test must play a larger role in screening out 27 Throughout both the 2003 NPRM and 2004 Final Rule, the Department emphasized that it was increasing the standard salary level from the $155 long test salary level last previously updated in 1975. See, e.g., 68 FR 15570; 69 FR 22123 (‘‘The final rule nearly triples the current $155 per week minimum salary level required for exemption to $455 per week.’’); id. at 22171. Neither the 2003 NPRM nor the 2004 Final Rule compared the magnitude of the new standard salary level against the former $250 per week short test salary level. PO 00000 Frm 00016 Fmt 4701 Sfmt 4702 overtime-protected white collar employees. While the role of the salary level test as an initial test for exemption increased in 2004, the Department has always recognized the impact of the threshold on overtime-eligible white collar employees. In the Stein Report, the Department looked at the impact of various salary thresholds on overtimeeligible bookkeepers, noting that approximately 50 percent of surveyed bookkeepers earned more than the then applicable $30 weekly salary threshold, while that number decreased to approximately 8 percent at the $50 dollar level at which the applicable salary level was ultimately set. Stein Report at 32. The Department went on to note that evidence that a salary of $50 ‘‘would not also exclude persons who properly deserve the exemption is illustrated by the fact that almost 50 percent of the accountants and auditors [many of whom are properly considered administrative or professional] earn at least $50 a week.’’ Id. Similarly, the Weiss Report noted that ‘‘[a]nother guide of value in determining the appropriate levels of a salary test for administrative and professional employees is the probable percentage of persons in clerical, subprofessional, or other nonexempt occupations who would meet the various salary requirements. The salary level adopted must exclude the great bulk of nonexempt persons if it is to be effective.’’ Weiss Report at 18. The Weiss Report went on to look at salaries paid to bookkeepers in New York and nine other surveyed cites and noted that, at a salary of $80 per week, some hand-bookkeepers in 9 of the 10 cities surveyed would exceed the salary level; at $75 per week, the salary test would be met by some hand-bookkeepers in all 10 cities. The report noted that the data ‘‘all tend to indicate that a salary requirement of about $75 or $80 a week for administrative employees is necessary in order to provide adequate protection against misclassification since many obviously nonexempt employees earn salaries at or near these figures.’’ Id. The Department set the salary level for administrative employees at $75 per week. The Department’s 2004 pairing of the lower long test salary level with the short test duties requirements also runs contrary to the Department’s rationale for the short duties test that ‘‘the higher the salaries paid the more likely the employees are to meet all the requirements for exemption,’’ and at ‘‘the higher salary levels in such classes of employment, the employees have almost invariably been found to meet all E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules the other requirements of the regulations for exemption.’’ Weiss Report at 22. Further, in establishing the short test the Department cautioned that ‘‘the salary level must be high enough to include only those persons about whose exemption there is normally no question.’’ Id. at 23. Setting the standard salary level at the 40th percentile of earnings for full-time salaried workers would effectively correct for the Department’s establishment in the 2004 Final Rule of a single standard duties test that was equivalent to the former short duties test without a correspondingly higher salary level. In the absence of the protection provided by the long duties test, the lower salary level increased the risk that employees who should be entitled to overtime protection might be inappropriately classified as exempt and denied that protection. The lower salary level associated with the former long duties test was never intended to ensure that the employees earning that amount meet ‘‘all the requirements for exemption . . . including the requirement with respect to nonexempt work.’’ Id. at 22– 23. Therefore, without a more rigorous duties test, the salary level set in the 2004 Final Rule is inadequate to serve the salary’s intended purpose of the ‘‘drawing of a line separating exempt from nonexempt employees[.]’’ 69 FR 22165. The importance of adjusting the salary level threshold upward to account for the lack of a long duties test is illustrated by the Department’s Burger King litigation in the early 1980’s, when the long test was still actively in use. The Department brought two actions arguing that Burger King restaurants in the northeast had misclassified their assistant managers as exempt executive employees and that these employees were, in fact, entitled to overtime protection. Sec’y of Labor v. Burger King Corp., 675 F.2d 516 (2d Cir. 1982); Sec’y of Labor v. Burger King Corp., 672 F.2d 221 (1st Cir. 1982). The assistant managers at issue all performed the same duties, which included spending significant amounts of time performing the same routine, nonexempt work as their subordinates. One group of assistant managers was paid between $155 and $249 per week—and therefore subject to the long duties test; the other group was paid $250 or more—and therefore subject to the short duties test. The Department argued that neither group of assistant managers had management as their primary duty. Both appellate courts found that the employees did have management as their primary duty; however, for the VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 lower paid group, both courts found the employees to be overtime protected because they spent more than 40 percent of their time performing nonexempt work and therefore did not satisfy the requirements of the long duties test. Accordingly, the lower paid employees were protected by application of the more rigorous long duties test, while the higher paid employees were found to be exempt under the easier short duties test. If the less rigorous short duties test had been paired with the long test’s lower salary threshold—as the Department did in 2004—the lower paid assistant managers would have lost their overtime protection. The continued extensive litigation regarding employees for whom employers assert the EAP exemption also demonstrates that using the 20th percentile of salaried employees in the South and in retail as the threshold has not met the Department’s goals as stated in the 2004 Final Rule of simplifying enforcement and reducing litigation. Id. According to a recent Government Accountability Office (GAO) report, statistics from the Federal Judicial Center show that the number of wage and hour lawsuits filed in federal courts ‘‘has increased substantially, with most of this increase occurring in the last decade.’’ GAO–14–69, ‘‘Fair Labor Standards Act,’’ December 2013, at 2, 6.28 A ‘‘total of 8,148 FLSA lawsuits [were] filed in fiscal year 2012. Since 2001, when 1,947 FLSA lawsuits were filed, the number of FLSA lawsuits has increased sharply.’’ Id. at 6. Stakeholders advised GAO that one of the reasons for the increased litigation was employer confusion about which workers should be classified as EAP exempt. Id. at 11. Adjusting the salary level upward to account for the absence of a more rigorous duties test will ensure that the salary threshold serves as a more clear line of demarcation between employees who are entitled to overtime and those who are not, and will reduce the number of white collar employees who may be misclassified and therefore decrease litigation related to application of the EAP duties test. At the 40th percentile of full-time salaried workers, there will be 10.9 million fewer white collar employees for whom employers could be subject to potential litigation regarding whether they meet the duties test for exemption (4.6 million who would be newly entitled to overtime due to the increase in the salary threshold and 6.3 million who previously failed the duties test and PO 00000 28 https://gao.gov/products/GAO-14-69. Frm 00017 Fmt 4701 Sfmt 4702 38531 would now also fail the salary level test). As discussed previously, the salary component of the EAP test for exemption has always worked hand-inhand with the duties test in order to simplify the application of the exemption. At a lower salary level, more overtime-eligible employees will exceed the salary threshold, and a more rigorous duties test would be required to ensure that they are not classified as falling within an EAP exemption and therefore denied overtime pay. At a higher salary level, more employees performing bona fide EAP duties will become entitled to overtime because they are paid a salary below the salary threshold. Setting the salary threshold too low reduces the risk that workers who pass the duties test become entitled to overtime protection, but does so at the cost of increasing the number of overtime-eligible employees exceeding the salary level who are subject to the duties test and possible misclassification. In contrast, setting the salary level too high reduces the number of overtime-protected employees subject to the duties test and eliminates their risk of misclassification, but at the cost of requiring overtime protection for workers who pass the duties test. With those concerns in mind, the Department has reviewed a variety of data sources to ascertain the appropriate amount to increase the required salary level in order to ensure that it works effectively with the standard duties tests to distinguish between overtime-eligible white collar employees and employees performing bona fide EAP duties. In the 1949, 1958, 1963, 1970 and 1975 updates to the salary level, all of which featured a long test/short test structure, the short test salary level was set at approximately 130 to 180 percent of the long duties test salary level to adequately establish a salary level that obviated the need to engage in a more probing duties analysis. To remedy the Department’s error from 2004 of pairing the lower long test salary with the less stringent short test duties, the Department is setting the salary level within the range of the historical short test salary ratio so that it will work appropriately with the current standard duties test. The Department recognizes that the proposed salary amount is only about 140 percent of the long duties test salary level under the Kantor method, and thus may be viewed as slightly out of line with the historic average of approximately 150 percent of the long test at which the short-test salary has E:\FR\FM\06JYP2.SGM 06JYP2 38532 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules been set.29 This suggests that a salary significantly lower than the 40th percentile of full-time salaried workers would pose an unacceptable risk of inappropriate classification of overtimeprotected employees without a change in the standard duties test. The Department believes that setting the salary level at the 40th percentile of weekly wages for all full-time salaried employees will result in a salary threshold that properly distinguishes between employees who may meet the duties requirements of the EAP exemption and those who likely do not, without necessitating a return to the more detailed long duties test. The Department notes that currently approximately 75 percent of white collar employees who do not meet the duties test earn less than the proposed salary threshold. The Department believes that the 40th percentile is appropriate because there is no longer a lower salary/long duties test for EAP exemption to which employers can turn if employees do not satisfy the standard salary level. By proposing a lower salary level than traditionally used for the short duties test, the Department intends to minimize the potential that additional bona fide exempt employees might become entitled to overtime because they fall below the proposed salary level. The Department notes that currently approximately 78 percent of all exempt EAP workers—those who are paid on a salary basis of at least $455 per week and meet the duties test—earn at least $921 per week. This salary level also accounts for the fact that the salary threshold will apply to all employees nationwide, including employees who work in low-wage regions and low-wage industries. In this rulemaking, we are proposing a salary level of the 40th percentile of the weekly wages of all full-time salaried workers nationwide. The Department believes that setting the salary level based on nationwide salary data is consistent with the goals of modernizing and simplifying the regulations. Using nationwide salary data will also produce a salary level appropriate to both low- and high-wage areas and industries. While the proposed salary level is lower than the average historical short test salary ratio under the Kantor method, a higher percentile more in line with the historical short duties test could have a negative impact on the ability of employers in low-wage regions and industries to claim the EAP exemptions for employees who have bona fide executive, administrative, or professional duties as their primary duty, particularly in the absence of a long duties test as an alternative. As will be discussed in section VII.D., the Department believes this proposal is appropriate in low-wage areas and lowwage industries. The proposal also is consistent with the Department’s practice in prior rulemakings, including the 2004 Final Rule, of establishing a national salary level, rather than multiple levels for different regions or industries. As stated in the 2004 Final Rule, the Department does not believe that having different salary levels for different areas of the country or for different kinds or sizes of businesses ‘‘is administratively feasible because of the large number of different salary levels this would require.’’ 69 FR 22171. The Department came to the same conclusion in 1940 when the Department rejected suggestions for varying salary levels, stating that it would present serious difficulties in enforcement, and that the FLSA is a national law that cannot take into account every small variation occurring over the length and breadth of the country. To make enforcement possible and to provide for equity in competition, a rate should be selected . . . which will be reasonable in light of average conditions for industry as a whole. In some instances the rate selected will inevitably deny exemption to a few employees who might not unreasonably be exempted, but, conversely, in other instances it will undoubtedly permit the exemption of some persons who should properly be entitled to the benefits of the act. Stein Report at 6; see Weiss Report at 9 (‘‘Regulations of general applicability such as these must be drawn in general terms to apply to many thousands of different situations throughout the country.’’). Setting the salary level at the 40th percentile of full-time salaried workers places it far enough above the minimum wage to provide an effective means of screening out workers who should be overtime protected. As the Stein Report noted, ‘‘[i]t must be assumed that [executive employees] enjoy compensatory privileges and this assumption will clearly fail if they are not paid a salary substantially higher than the wages guaranteed as a mere minimum under section 6 of the act.’’ Stein Report at 19. Furthermore, the failure to require a salary level of substantially more than the minimum wage would ‘‘invite evasion of section 6 and 7 for large numbers of workers to whom the wage-and-hour provisions should apply.’’ Id. Accordingly, following each update from 1949 to 1975 (those which included a short duties test similar to the current standard test), the ratio of the short test salary level to the earnings of a fulltime, nonexempt, minimum wage worker equaled between approximately 3.0 and 6.25.30 See Table B. For instance, the ratio was its highest in 1949 at 6.25 ($100 salary level divided by the product of $0.40 and 40 hours) and its lowest in 1975 at 2.98 ($250/($2.10 × 40)). Because the 2004 standard salary level was based on the 1975 long test salary and not the short test salary, it deviated from the pattern observed over the previous decades, resulting in a salary threshold of just 2.21 times fulltime minimum wage earnings ($455/ ($5.15 × 40)). The proposed salary level is 3.18 times full-time minimum wage earnings ($921/($7.25 × 40)), which is consistent with the historical average. Therefore, the Department believes that the proposed salary level is appropriate in comparison with prior minimum wage ratios. TABLE B—RATIOS OF SALARY TEST LEVELS TO FULL-TIME MINIMUM WAGE EARNINGS Minimum wage (MW) tkelley on DSK3SPTVN1PROD with PROPOSALS2 Year 1949 ................................................................................................. 1958 ................................................................................................. 29 The Department estimated the average historic ratio of 149 percent as the simple average of the fifteen historical ratios of the short duties salary level to the long duties salary level (salary levels were set in 5 years and in each year the salary level varied between the three exemptions: executive, administrative, and professional). If the Department VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 MW earnings for a 40-hour workweek $0.40 1.00 had weighted the average ratio based on the length of time the historic salary levels were in effect, this would have yielded an average historic ratio of 152 percent. 30 The 6.25 ratio is an outlier that was set in December 1949 (when the short test was created) PO 00000 Frm 00018 Fmt 4701 Sfmt 4702 Exempt short test salary level $16 40 $100 125 Ratio of short salary test to MW earnings 6.25 3.13 and the minimum wage increased from $.40 to $.75 per hour one month later (which reduced the ratio to 3.33). To return to the 6.25 ratio, the weekly salary level would have to be set at $1,812.50, which is around the 80th percentile of all full-time salaried employees. E:\FR\FM\06JYP2.SGM 06JYP2 38533 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE B—RATIOS OF SALARY TEST LEVELS TO FULL-TIME MINIMUM WAGE EARNINGS—Continued Minimum wage (MW) Year 1963 ................................................................................................. 1970 ................................................................................................. 1975 ................................................................................................. Year 1.25 1.60 2.10 Minimum wage (MW) tkelley on DSK3SPTVN1PROD with PROPOSALS2 2004 ................................................................................................. 2015 ................................................................................................. Moreover, the median earnings for all salaried workers provides further support for the proposed salary level. The Weiss Report observed approvingly that in the Stein Report, the ‘‘dividing line [between subprofessional and professional employees was] based on the midpoint salaries’’ of federal government service classifications of administrative and professional employees, and thus suggested that a midpoint value of the aggregated earnings of such workers is an appropriate benchmark for the salary level. Weiss Report at 16–17 (referencing Stein Report at 43). In 1947, 1962, 1969, and 2003, data showing median increases in earnings for all employees in various industries were generated and considered instructive to a determination of an appropriate salary level.31 The 2013 national median earnings for all full-time salaried workers was $1,065 per week, giving support to the Department’s proposed salary level of $921. Thus, using median earnings as a point of comparison supports that the 40th percentile of fulltime salaried workers would provide an appropriate line of demarcation between overtime-eligible white collar employees and potentially exempt EAP employees. The Department’s proposed salary level is further supported by its increased ability to distinguish overtime-eligible employees. The primary objective of the salary level test has always been the drawing of a line separating overtime-eligible white collar salaried employees from employees who may be bona fide EAP employees. At the current salary threshold, there are 11.6 million salaried white collar workers who are overtime protected but are paid 31 Statistical Materials Bearing on the Salary Requirement in Regulations Part 541 (1947), at 2, 6, 27–30, 56–57; Salary Tests for EAP Employees DOL Report—Wage and Hour Public Contracts Division (1962), at 3, 7–15, 18, 20; Salary Tests WHD Report (1969), at 19, 48. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 MW earnings for a 40-hour workweek 50 64 84 MW earnings for a 40-hour workweek $5.15 7.25 at or above the $455 salary level and therefore must be subjected to a duties analysis to determine their overtime eligibility. At the proposed salary level, the number of overtime-eligible salaried white collar employees paid at or above the salary level would be reduced by more than 50 percent. Thus a salary level at the 40th percentile of weekly earnings for salaried workers would be more efficient at distinguishing overtime-eligible employees. v. Alternatives Considered While the Department has largely followed historical precedent in determining the proposed salary threshold by basing it on the level of salaries that employers currently pay and making only modest changes to our time-tested model, the Department did consider other approaches to determine the appropriate salary test level.32 First, the Department considered adjusting either the 2004 standard salary test level or the 1975 short test salary level for inflation using the CPI, similar to the methodology used to set the salary levels in the 1975 interim update. The Department noted in 1975 that ‘‘[t]he rapid increase in the cost of living since the salary tests were last adjusted justifies an interim increase in those tests . . . [and] the widely accepted [CPI] may be utilized as a guide for establishing these interim rates.’’ 40 FR 7091. However, the Department noted at that time that the adoption of interim rates, while necessary to expeditiously provide protection for workers affected by a salary level rendered obsolete by rapid cost-of-living changes, was not considered a precedent for future rulemaking (and those same inflationary conditions do not exist today). Id. at 7092. In other years, however, the Department has looked at inflation when increasing the salary level, but has 32 The alternatives the Department considered are discussed in more detail in section VII.C. PO 00000 Frm 00019 Fmt 4701 Sfmt 4702 Exempt short test salary level 150 200 250 Exempt short test salary level $206 290 $455 921 (proposed) Ratio of short salary test to MW earnings 3.00 3.13 2.98 Ratio of short salary test to MW earnings 2.21 3.18 never established the actual numerical salary level based on inflation. The Department has thus recognized that measures of inflation and losses in purchasing power provide helpful background for setting the salary level because they indicate how far the levels erode between updates and underscore the need for an update. They can also point very generally to ranges in which new salary levels might be considered. Indeed, with respect to the current rulemaking, looking at inflation provides added support for the proposed salary level. Updating the 2004 standard salary level for inflation based on the Consumer Price Index for all urban consumers (CPI–U) would result in a salary level of $561 per week (approximately the 15th percentile of weekly earnings for all full-time salaried workers). Updating the 1975 short test salary level with the CPI–U would result in a salary level of $1,083 per week (approximately the 50th percentile of weekly earnings for all full-time salaried workers). Considering that the standard test most closely approximates the historic short duties test, looking at an inflation adjustment would support a higher salary level than that being proposed. However, inflation has been used as a method for setting the precise salary level only in the breach, as in 1975 when practical considerations prevented a more complete analysis of actual salaries. The Department continues to believe that looking to the actual earnings of workers provides the best evidence of the rise in prevailing salary levels and, thus, constitutes the best source for setting the proposed salary requirement. This viewpoint reflects guidance from previous updates, including the Weiss Report, where the Department rejected suggestions to base the salary level on the change in the cost of living. Weiss Report at 12 (‘‘The change in the cost of living which was urged by several witnesses as a basis for determining the appropriate levels is, in E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38534 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules my opinion, not a measure for the rise in prevailing minimum salaries.’’). The Department also considered setting the salary level using the 2004 method (20th percentile of full-time salaried employees in the South and retail) or Kantor method (10th percentile of likely exempt employees in low-wage regions, employment size groups, city size, and industries). While these methods produced similar salaries in 2004 when the Department last revised the salary levels, over time they have diverged significantly and today would result in salaries of $577 and $657 per week, respectively (approximately the 15th and 20th percentiles of weekly earnings for all full-time salaried workers). Because the Kantor method was based on the long test duties requirements (which limited the amount of nonexempt work that EAP employees could perform), the Department concluded that the resulting salary level was inappropriately low when paired with the standard duties test (which was based on the short test). For similar reasons the Department concluded that the 2004 method (which paired the lower long test salary level with a standard duties test based on the short duties test) also resulted in an inappropriately low salary level. The Department further considered setting the standard salary level equal to the median earnings for all full-time wage and salaried workers combined (i.e., not just salaried, also workers paid by the hour). This median provides a rough dividing line between the generally lower-paid hourly workers who are overtime protected and the generally higher-paid salaried workers who may be exempt. The national median earnings for all full-time workers, both wage and salary, in all occupations and industries, and across metropolitan and rural areas, was $776 per week (approximately the 30th percentile of weekly earnings for all full-time salaried workers). The Department concluded, however, that it would not be appropriate to include the wages of hourly workers in setting the EAP salary threshold and that the resulting salary level was too low to work effectively with the standard duties test. The Department also considered updating the Kantor long test salary level of $657 to a short test level, reflecting the historical relationship of the short test to the long test which has ranged from approximately 130 percent to 180 percent of the long test level and averaged approximately 150 percent. This would result in a salary level between $854 and $1,183 per week, with the historical average yielding a VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 salary level of $979 per week. The end points of the historical range are approximately the 35th and 55th percentiles of weekly earnings for all full-time salaried workers, respectively. While the Department thought that salaries throughout this historical salary range would work appropriately with the standard duties test, we were concerned that the top end of the resulting range would be too high for low-wage regions and industries, particularly because employers no longer have a long duties test to fall back on for purposes of exempting lower-salaried workers performing bona fide EAP duties. Finally, the Department considered setting the standard salary equal to the 50th percentile, or median, of weekly earnings for all full-time salaried workers. This method would be similar to the proposed method but would use a higher percentile. Using the 50th percentile would result in a standard salary level of $1,065 per week. The Department believes that the salary level generated with this method would be too high for low-wage regions and industries, particularly in light of the absence of a lower salary long duties test. When measured against inflation or previous methods of setting the salary levels (standard, short, and long), the proposed salary level is within the range that was the historical norm until the 2004 update. For instance, this level falls well below the 1975 inflationadjusted short test level ($1,083 per week) and is lower than the salary level comparable to the average historical ratio between the short and long test salary ($979 per week). But the proposed salary exceeds the inflationadjusted 2004 salary level and the levels suggested by the Kantor and 2004 methods (all of which were based on the long test salary). While, for the reasons stated herein, none of these alternative measures was used as a methodology to establish the proposed salary test level, they confirm that the 40th percentile of weekly earnings of all full-time salaried employees ($921) proposed by the Department is in line with previous updates. vi. Summary of Proposed Change to Standard Salary Level Therefore, for the reasons stated above, the Department proposes to increase the standard salary level to qualify for exemption from the FLSA minimum wage and overtime requirements as an executive, administrative, or professional employee from $455 a week to the weekly earnings of the 40th percentile PO 00000 Frm 00020 Fmt 4701 Sfmt 4702 of full-time salaried employees ($921 a week). The Department reached the proposed salary level after considering available data on actual salary levels currently being paid in the economy. The Department believes that, in view of the regulatory history and all other relevant considerations, using the earnings of all full-time salaried workers (exempt and nonexempt) as the basis for setting the proposed salary level is appropriate here, and setting the salary level at the 40th percentile establishes an appropriate dividing line helping differentiate between white collar workers who are overtime-eligible and those who are not. The Department invites comments on this proposed salary level and on any alternative salary level amounts, or methodologies for determining the salary level, that appropriately distinguish between overtime-eligible white collar workers and bona fide EAP workers. In addition, the Department invites comments on the effectiveness of the proposed salary level to both limit the number of employees who pass the EAP duties tests but become overtime eligible because of the increased salary level, and reduce the number of employees who fail the EAP duties test but are subject to a duties analysis and possible misclassification by their employers. B. Special Salary Tests i. American Samoa The Department has historically applied a special salary level test to employees in American Samoa because minimum wage rates in that jurisdiction have remained lower than the federal minimum wage. See 69 FR 22172. Prior to July 24, 2007, industry-specific minimum wage rates for American Samoa were set by a special industry committee appointed by the Department. See Sec. 5, Pub. L. 87–30, 75 Stat. 67 (May 5, 1961). The Fair Minimum Wage Act of 2007 replaced this methodology with a system of incremental increases. See Sec. 8103, Pub. L. 110–28, 121 Stat. 188 (May 25, 2007). As amended, this law provides that the American Samoa minimum wage for each industry will increase by $0.50 on September 30, 2015, and continue to increase every three years thereafter until each equals the federal minimum wage. See Sec. 4, Pub. L. 112– 149, 126 Stat. 1145 (July 26, 2012). The minimum wage in American Samoa currently ranges from $4.18 to $5.59 an hour depending on the industry,33 and 33 See WHD Minimum Wage Poster for American Samoa, available at https://www.dol.gov/whd/ minwage/americanSamoa/ASminwagePoster.pdf. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules so the disparity with the federal minimum wage is expected to remain for the foreseeable future. Accordingly, the Department proposes to maintain a special salary level test for employees in American Samoa. Consistent with our practice since 1975, in the 2004 Final Rule the Department set the special salary level test for employees in American Samoa at approximately 84 percent of the standard salary test level—which computed to $380 per week. See 69 FR 22172. The Department believes that our approach in the 2004 Final Rule remains appropriate given the continued gap between American Samoa and federal minimum wage rates. Accordingly, the Department proposes to set the American Samoa special salary level test at $774, which equals approximately 84 percent of the proposed standard salary level of the 40th percentile of weekly earnings for full-time salaried workers ($921). The Department also proposes that when the minimum wage rate for any industry in American Samoa equals the federal minimum wage, the standard salary level will then apply in full for all EAP employees in all industries in American Samoa. The Department invites comments on this special salary level proposal. tkelley on DSK3SPTVN1PROD with PROPOSALS2 ii. Motion Picture Producing Industry The Department currently permits employers to classify as exempt employees in the motion picture producing industry who are paid at a base rate of at least $695 per week (or a proportionate amount based on the number of days worked), so long as they meet the duties tests for the EAP exemptions. § 541.709. This exception from the ‘‘salary basis’’ requirement was created to address the ‘‘peculiar employment conditions existing in the [motion picture] industry’’ (18 FR 2881 (May 19, 1953)), and applies, for example, when a motion picture industry employee works less than a full workweek and is paid a daily base rate that would yield at least $695 if six days were worked. Id. The Department has provided this industry-specific exception to the salary basis requirement since 1953. 18 FR 3930 (July 7, 1953). In the 2004 Final Rule the Department increased the base rate for motion picture industry employees by the same percentage that the salary level tests, on average, increased.34 See 69 FR 22190. 34 Specifically, in the 2004 Final Rule the Department increased the standard salary level test by approximately 170 percent for professional employees (from a long test salary level of $170 to VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 Consistent with the 2004 Final Rule methodology, the Department proposes to increase the required base rate proportionally to the proposed increase in the standard salary level test. The Department is proposing to increase the standard salary level by approximately 102 percent—from $455 to $921. Accordingly, in § 541.709, the Department proposes to increase the current base rate for employees in the motion picture industry by approximately 102 percent—from $695 to $1,404 per week (or a proportionate amount based on the number of days worked). The Department invites comments on this base rate proposal. C. Inclusion of Nondiscretionary Bonuses in the Salary Level Requirement The Department has consistently assessed compliance with the salary level test by looking only at actual salary or fee payments made to employees and, with the exception of the highly compensated test, has not included bonus payments of any kind in this calculation. During stakeholder listening sessions several business representatives asked the Department to include nondiscretionary bonuses and incentive payments as a component of any revised salary level requirement. These stakeholders conveyed that nondiscretionary bonuses and incentive payments are an important component of employee compensation in many industries and stated that such compensation might be curtailed if the standard salary level was increased and employers had to shift compensation from bonuses to salary to satisfy the new standard salary level. They asserted that such a change would have a negative impact on the workplace and would undermine managers’ sense of ‘‘ownership’’ in their organizations. A few employer stakeholders also raised the possibility of counting fringe benefits and/or commissions toward the salary level requirement. The Department’s longstanding position has been to allow employers to pay additional compensation in the form of bonuses in addition to the required salary. § 541.604(a). However, in recognition of the increased role bonuses play in many compensation systems, and as part of the Department’s a standard test salary level of $455), and by roughly 190 percent for executive and administrative employees (from a long test salary level of $155 to a standard test salary level of $455). The Department averaged these two percentiles and increased the base rate for motion picture industry employees by 180 percent—from $250 to $695. See 69 FR 22190. PO 00000 Frm 00021 Fmt 4701 Sfmt 4702 38535 efforts in this rulemaking to modernize these regulations, the Department is now considering whether to also permit nondiscretionary bonuses and incentive payments to count toward a portion of the standard salary level test for the executive, administrative, and professional exemptions.35 Such payments may include, for example, nondiscretionary incentive bonuses tied to productivity and profitability. Thus, the Department is considering whether compensation such as a nondiscretionary bonus for meeting specified performance metrics, in combination with a minimum weekly salary amount, may be counted in satisfying the standard salary level test. The Department is also considering how to include nondiscretionary bonuses and incentive payments as part of the salary level test, if such a change is implemented. Compliance with the HCE exemption’s $100,000 total compensation requirement is assessed annually, and employers are permitted to make a ‘‘catch-up’’ payment at or shortly after the end of the year that counts toward this amount. Employees for whom the HCE exemption is claimed must receive the full standard salary amount, currently $455, weekly on a salary or fee basis. See § 541.601(b). The Department believes that a different approach would be needed for the standard salary test. Because the only compensation guaranteed to employees for whom the employer claims the standard EAP exemption is the standard salary threshold amount, the Department believes it is important to strictly limit the amount of the salary requirement that could be satisfied through the payment of nondiscretionary bonuses and incentive pay. The Department is considering whether to permit such payments to satisfy 10 percent of the standard weekly salary level. The Department recognizes that some businesses pay significantly larger bonuses and where larger bonuses are paid, the amount attributable toward the EAP standard salary requirement would be capped at 10 percent of the salary level if such a provision were adopted. The 35 The Department notes that overtime-eligible (i.e., nonexempt) employees may also receive such bonuses. Where nondiscretionary bonuses or incentive payments are made to overtime-eligible employees, the payments must be included in the regular rate when calculating overtime pay. The Department’s regulations at §§ 778.208-.210 explain how to include nondiscretionary bonuses in the regular rate calculation. One way to calculate and pay such bonuses is as a percentage of the employee’s total earnings. Under this method, the payment of the bonus includes the simultaneous payment of overtime due on the bonus payment. See § 778.210. E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38536 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules Department also believes that the time period over which such compensation should be considered must be limited. Permitting bonuses to be paid as much as a year out would significantly undermine the crucial protection provided by the salary basis requirement, which ensures that exempt workers receive a minimum level of compensation on a consistent basis. Accordingly, the Department envisions that in order for employers to be permitted to credit such compensation toward the weekly salary requirement employees would need to receive the bonus payments monthly or more frequently. For similar reasons, the Department is not considering permitting employers to make a yearly catch-up payment like under the HCE exemption. With these parameters in mind, the Department seeks comments on whether it should modify the standard exemption for executive, administrative, and professional employees to permit nondiscretionary bonuses and incentive payments to count toward partial satisfaction of the salary level test. The Department seeks information on what industries commonly have pay arrangements that include nondiscretionary bonuses and incentive payments, what types of employees typically earn nondiscretionary bonuses and incentive payments, the types of nondiscretionary compensation employees receive, and to what extent including nondiscretionary bonuses and incentive payments as part of the salary level would advance or hinder that test’s ability to serve as a dividing line between exempt and nonexempt employees. The Department also seeks comments on whether payment on a monthly basis is the appropriate interval for such nondiscretionary compensation that will be credited toward the weekly salary requirement, and whether 10 percent is the appropriate limit on the amount of the salary requirement that can be satisfied by nondiscretionary bonuses and incentive payments (with the remaining 90 percent paid on a salary or fee basis in accordance with the regulations). Consistent with the rule for highly compensated employees (which counts nondiscretionary bonuses toward the total annual compensation requirement), the Department is not considering expanding the salary level test calculation to include discretionary bonuses. The Department is also not considering changing the exclusion of board, lodging, or other facilities from the salary calculation, a position that it has held consistently since the salary requirement was first adopted. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 Similarly, the Department also declines to consider including in the salary requirement payments for medical, disability, or life insurance, or contributions to retirement plans or other fringe benefits. See § 541.601(b)(1). The Department is also concerned it would be inappropriate to count commissions toward the salary level requirement, as employees who earn commissions are usually sales employees who—with the exception of outside sales employees—are generally unable to satisfy the standard duties test (which is more stringent than the HCE duties test) for the EAP exemptions. However, the Department seeks comments on the appropriateness of including commissions as part of nondiscretionary bonuses and other incentive payments that could partially satisfy the standard salary level test. D. Highly Compensated Employees In the 2004 Final Rule, the Department created a new highly compensated exemption for EAP employees. Section 541.601(a) provides that such employees are exempt if they earn at least $100,000 in total annual compensation and customarily and regularly perform any one or more of the exempt duties or responsibilities of an executive, administrative, or professional employee. Section 541.601(b)(1) states that employees must receive at least $455 per week on a salary or fee basis, while the remainder of the total annual compensation may include commissions, nondiscretionary bonuses, and other nondiscretionary compensation. It also clarifies that total annual compensation does not include board, lodging, and other facilities, and does not include payments for medical insurance, life insurance, retirement plans, or other fringe benefits. Pursuant to § 541.601(b)(2), an employer is permitted to make a final payment (catch-up pay) during the final pay period or within one month after the end of the 52-week period to bring an employee’s compensation up to the required level. If an employee does not work for a full year, § 541.601(b)(3) permits an employer to pay a pro rata portion of the required annual compensation, based upon the number of weeks of employment (and one final payment may be made, as under paragraph (b)(2), within one month for employees who leave employment during the year). In the 2003 NPRM, where the HCE test was first introduced, the Department had proposed to require total annual compensation of at least $65,000. The Department stated that, ‘‘[t]o determine an appropriate salary PO 00000 Frm 00022 Fmt 4701 Sfmt 4702 level for highly compensated employees, the Department looked to points near the higher end of the current range of salaries and found that the top 20 percent of all salaried employees earned above $65,000 annually. This level is consistent with setting the proposed standard test salary level at the bottom 20 percent of salaried employees.’’ 68 FR 15571. However, in the 2004 Final Rule, the Department recognized that the required compensation level had to ‘‘be set high enough to avoid the unintended exemption of large numbers of employees—such as secretaries in New York City or Los Angeles—who clearly are outside the scope of the exemptions and are entitled to the FLSA’s minimum wage and overtime pay provisions.’’ 69 FR 22174. Therefore, the Department increased the required annual compensation to $100,000, to ‘‘address commenters’ concerns regarding the associated duties test, the possibility that workers in high-wage regions and industries could inappropriately lose overtime protection, and the effect of future inflation.’’ Id. at 22175. The Department set the level at $100,000 because our experience demonstrated that virtually every salaried ‘‘white collar’’ employee with a total annual compensation of $100,000 per year would satisfy any duties test. Employees earning $100,000 or more per year are at the very top of today’s economic ladder, and setting the highly compensated test at this salary level provides the Department with the confidence that, in the words of the Weiss report: ‘‘in the rare instances when these employees do not meet all other requirements of the regulations, a determination that such employees are exempt would not defeat the objectives of section 13(a)(1) of the Act.’’ Id. at 22174 (quoting Weiss Report at 22–23). The Department further noted that ‘‘[o]nly roughly 10 percent of likely exempt employees who are subject to the salary tests earn $100,000 or more per year,’’ which the Department noted was ‘‘broadly symmetrical with the Kantor approach of setting the minimum salary level for exemption at the lowest 10 percent of likely exempt employees. In contrast, approximately 35 percent of likely exempt employees subject to the salary tests exceed the proposed $65,000 salary threshold.’’ Id. The Department continues to believe that an HCE test for exemption is an appropriate means of testing whether highly compensated employees qualify as bona fide executive, administrative, or professional employees. In the 2004 Final Rule, the Department concluded that the requirement for $100,000 in total annual compensation struck the E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules right balance by matching a much higher compensation level than was required for the standard salary level test with a duties test that was more flexible than the standard duties test, thereby creating a bright-line test that allowed only appropriate workers to qualify for exemption. See 69 FR 22174. This total annual compensation requirement was set more than four times higher than the standard salary requirement of $455 per week, which totals $23,660 per year. Id. at 22175. Such a balancing of a substantially higher compensation requirement with a minimal duties test still is appropriate, so long as the required annual compensation threshold is sufficiently high to ensure that it covers only employees who ‘‘have almost invariably been found to meet all the other requirements of the regulations for exemption.’’ Id. at 22174. Therefore, the Department proposes to increase the total annual compensation required by § 541.601 in order to ensure that it remains a meaningful and appropriate standard when matched with the minimal duties test. Just as with the standard salary level test, it is imperative to increase the compensation level that was established more than ten years ago to ensure that it continues to allow for the exemption of only bona fide exempt employees. Over the past decade, the percentage of salaried employees who earn more than $100,000 annually has increased substantially to approximately 17 percent of full-time salaried workers. Accordingly, the Department proposes to increase the total annual compensation requirement to the annualized weekly earnings of the 90th percentile of all full-time salaried workers ($122,148). As discussed earlier with respect to the standard salary level, the Department is proposing to set the annual compensation requirement as the annualized value of a percentile of weekly earnings of full-time salaried workers rather than a specific dollar amount because we believe it serves as a better proxy for distinguishing those white collar workers who meet the requirements of the HCE exemption. Consistent with the current regulations, the Department also proposes that at least the standard salary requirement must be paid on a salary or fee basis.36 The Department is not proposing any 36 Should the Department implement in the final rule resulting from this proposed rule a provision allowing employers to take a credit against the standard salary level for nondiscretionary bonuses paid to the employee, that credit would not be applicable in determining compliance with the standard salary requirement for HCE workers. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 changes to the HCE duties test created in 2004. The Department believes that the 90th percentile of full-time salaried workers is appropriate because it brings the required compensation level more in line with the level established in 2004; therefore, it will ensure that, as in 2004, the HCE exemption covers only those employees who are at the very top of today’s economic ladder and minimizes ‘‘the possibility that workers in highwage regions and industries could inappropriately lose overtime protection.’’ 69 FR 22175. The proposed $122,148 requirement also generally corresponds to the increase that would result from updating the $100,000 level by the amount of the increase in the CPI–U between 2004 and 2013 (the CPI– U increase would result in a compensation level of approximately $123,000). The Department invites comments on whether the 90th percentile is the correct HCE total annual compensation level and whether the Department should make any other changes to the requirements for the use of the HCE exemption. E. Automatically Updating the Salary Levels As previously discussed, the salary level test plays a crucial role in ensuring that the EAP exemptions effectively differentiate between exempt and overtime-protected workers. But even a well-calibrated salary level that is fixed becomes obsolete as wages for nonexempt workers increase over time. Since the EAP regulations were first issued in 1938, the Department has increased the salary level only seven times—in 1940, 1949, 1958, 1963, 1970, 1975, and 2004. The lapses between rulemakings have resulted in salary levels that are based on outdated salary data and thus ill-equipped to help employers assess which employees are unlikely to meet the duties tests for the exemptions. During stakeholder listening sessions several employee advocates called on the Department to index the EAP salary level requirement to ensure that the revised salary test set in this rulemaking does not suffer the same fate as the salary tests in the Department’s prior rulemakings. After careful consideration of the history of EAP salary increases and the impact on the regulated community of routine updating of the salary test, the Department is proposing to modernize the EAP exemptions by establishing a mechanism for automatically updating the standard salary test, as well as the total annual compensation requirement for highly compensated employees. The addition of automatic updating will PO 00000 Frm 00023 Fmt 4701 Sfmt 4702 38537 ensure that the salary test level is based on the best available data (and thus remains a meaningful, bright-line test), produce more predictable and incremental changes in the salary required for the EAP exemptions, and therefore provide certainty to employers, and promote government efficiency by removing the need to continually revisit this issue through resource-intensive notice and comment rulemaking. The Department also proposes to update annually the special salary level test for employees in American Samoa and the base rate test for motion picture industry employees, as described infra. The Department is considering two alternative methodologies for annually updating the salary and compensation thresholds. One method would update the thresholds based on a fixed percentile of earnings for full-time salaried workers. The other method would update the thresholds based on changes in the CPI–U. Both methods are described in detail below and the Department seeks comments on which methodology would be the most appropriate basis for annual updates to the salary and compensation thresholds. i. History of Automatically Updating the Salary Levels The Department has only directly commented twice on the subject of automatically updating the salary level test for the EAP exemptions. In the 1970 rulemaking, the Department stated that a comment ‘‘propos[ing] to institute a provision calling for an annual review and adjustment of the salary tests . . . appears to have some merit, particularly since past practice has indicated that approximately 7 years elapse between amendment of the salary level requirements.’’ 35 FR 884. Despite recognizing the potential value of this approach, the Department ultimately determined that ‘‘such a proposal will require further study.’’ Id. In the 2004 Final Rule the Department declined to adopt commenter requests for automatic increases to the salary level, reasoning in part that ‘‘the salary levels should be adjusted when wage survey data and other policy concerns support such a change’’ and that ‘‘the Department finds nothing in the legislative or regulatory history that would support indexing or automatic increases.’’ 69 FR 22171. Although the Department acknowledged the lack of historical guidance related to the automatic updating of salary levels, in the 2004 Final Rule we did not discuss the Department’s authority to promulgate such an approach through notice and comment rulemaking. Rather than explore in greater depth whether E:\FR\FM\06JYP2.SGM 06JYP2 38538 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 automatic updates to the salary levels posed a viable solution to problems created by lapses between rulemakings, the Department expressed our intent ‘‘in the future to update the salary levels on a more regular basis, as it did prior to 1975.’’ Id. As discussed below, difficulties in achieving this goal have led the Department to examine the possibility of automatically updating salary levels in greater detail. The lack of Congressional guidance either supporting or prohibiting automatic updating is unsurprising given the origin and evolution of the salary level test, and does not foreclose the Department’s proposal. Congress did not specifically set forth precise criteria, such as a salary level test, for defining the EAP exemptions, but instead delegated that task to the Secretary. The Department established the first salary level tests by regulation in 1938, using our delegated authority to define and delimit the EAP exemptions. See 29 U.S.C. 213(a)(1). The fact that the salary level tests were created by regulation after the FLSA was enacted helps explain why the FLSA’s early legislative history does not address the salary level tests or methods for updating the salary level. Despite numerous amendments to the FLSA over the past 75 years, Congress has continued to entrust the Department with promulgating, updating, and enforcing the salary test regulations. Significant regulatory changes since 1938 include adding a separate salary level for professional employees in 1940, adopting separate VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 short and long test salary levels in 1949, and creating a single standard salary level test and a new HCE exemption in 2004. These changes were all made without express Congressional guidance, and none have been superseded by statute. Other than directing the Department in 1990 to include in the section 13(a)(1) exemption regulations certain computer employees paid at least six-and-a-half times the minimum wage on an hourly basis, see Sec. 2, Pub. L. 101–583, 104 Stat. 2871 (Nov. 15, 1990), Congress has never amended the FLSA in a manner that affects the salary level tests. It has also never enacted limits on the Department’s ability to update the salary levels. Just as the Department has authority under 29 U.S.C. 213(a)(1) to establish and update the salary level tests, it likewise has authority to adopt a methodology through notice and comment rulemaking for automatically updating the salary levels to ensure that the tests remain effective. This interpretation is consistent with the well-settled principle that agencies have authority to ‘‘ ‘fill any gap left, implicitly or explicitly, by Congress.’ ’’ Long Island Care at Home, Ltd. v. Coke, 551 U.S. 158, 165 (2007) (quoting Chevron, U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 843 (1984)). ii. Rationale for Automatically Updating Salary Levels The addition of an automatic updating mechanism will ensure that PO 00000 Frm 00024 Fmt 4701 Sfmt 4702 the standard salary level and the HCE total annual compensation requirement remain meaningful tests for distinguishing between bona fide EAP workers who are not entitled to overtime and overtime-protected white collar workers. Experience has shown that the salary level test is only a strong measure of exempt status if it is up to date. Left unchanged, the test becomes substantially less effective as wages for overtime-protected workers increase over time. See Weiss Report at 8 (‘‘The increase in wage rates and salary levels gradually weakened the effectiveness of the present salary tests as a dividing line between exempt and nonexempt employees.’’); see also 69 FR 22164 (explaining that 1975 salary levels had grown outdated and were ‘‘no longer useful in distinguishing between exempt and nonexempt employees’’). For example, in 2005 18.6 million workers subject to the FLSA were potentially covered by the EAP exemptions and in 2013 that number had grown to 21.4 million—an increase of 15 percent—while the number of workers subject to the FLSA grew only 5.8 percent during that period. See Figure A. Automatically updating the salary level using the most recent data ensures that the salary level test continues to accurately reflect current salary conditions. This specific proposal also helps fulfill the President’s instruction to modernize the part 541 regulations. 79 FR 18737. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38539 Automatically updating the salary level will ensure that it continues to be a reliable proxy for identifying overtime-eligible white collar employees, thus reducing one source of uncertainty for employers and employees. Regular updates to the salary level will also prevent the more drastic and unpredictable salary level increases that have resulted from the differing time periods between rulemakings. For example, between 1940 and 2004 the time between salary level updates ranged from five to 29 years. In part as a result of these breaks, long test salary level increases between 1940 and 1975 ranged from roughly five to 50 percent, and the 2004 standard salary level test represented an average 180 percent increase from the 1975 long test salary levels. Automatically updating the standard salary level test will ensure that future salary level increases occur at regular intervals and at more even increments. The Department recognizes that instituting a mechanism for automatically updating the salary level is a change to the part 541 regulations. As explained in the 2004 Final Rule, the Department’s reluctance to institute automatic updating was tied in part to VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 our preference for issuing new salary level regulations when new wage survey data necessitated such action. 69 FR 22171. However, a review of salary test history shows that the Department has updated the salary level only once since 1975, and has gone nine or more years between updates on several occasions. This history underscores the difficulty in maintaining an up-to-date and effective salary level test, despite the Department’s best intentions. Competing regulatory priorities, overall agency workload, and the time-intensive nature of notice and comment rulemaking have all contributed to the Department’s difficulty in updating the salary level test as frequently as necessary to reflect changes in workers’ salaries. These impediments are exacerbated because unlike most regulations, which can remain both unchanged and forceful for many years if not decades, in order for the salary level test to be effective, frequent updates are imperative to keep pace with changing employee salary levels. Confronted with this regulatory landscape, the Department believes automatic updating is the most viable and efficient way to ensure that the standard salary level test and the HCE PO 00000 Frm 00025 Fmt 4701 Sfmt 4702 total annual compensation requirement remain current and can serve their intended function of helping differentiate between white collar workers who are overtime-eligible and those who are not. iii. Proposal for Automatic Updating of the Standard Salary Level Test The Department proposes to insert a new provision in the regulations in the Final Rule that will establish a set methodology for recalculating the required salary level annually. The Department is not proposing specific regulatory text because it has not chosen the updating methodology and is instead seeking comments on two alternatives—using a fixed percentile of wage earnings or using the CPI–U. In the 1970 rulemaking, the Department recognized the potential merit of automatically updating the salary level test, but determined that such action would ‘‘require further study.’’ 35 FR 884. The Department has now examined a range of possible updating methodologies and concluded, for the reasons stated herein, that either maintaining the standard salary level at the 40th percentile of weekly wages of all full-time salaried workers or E:\FR\FM\06JYP2.SGM 06JYP2 EP06JY15.005</GPH> tkelley on DSK3SPTVN1PROD with PROPOSALS2 Figure A: Employees Subject to EAP Salary Level Requirement tkelley on DSK3SPTVN1PROD with PROPOSALS2 38540 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules updating the standard salary threshold based on changes in the CPI–U would maintain the effectiveness of the salary level in distinguishing overtime-eligible white collar salaried employees from those who may be exempt. Regardless of the updating method used, the Department proposes to publish the revised salary and compensation levels annually using the most recent data as determined and published by BLS. The Department will publish a notice with the new salary level in the Federal Register, as well as on the WHD Web site, at least sixty days before the updated rates would become effective. Should the Department choose to make any changes to the updating methodology in the future, such changes would require notice and comment rulemaking. 1. Fixed Percentile Approach to Automatically Updating the Standard Salary Level The ‘‘fixed percentile’’ approach would permit the Department to reset the salary level test by applying the same methodology proposed in this rulemaking to update the standard salary level. As explained at length in section V.A. of this preamble, the proposed salary level test methodology closely tracks prior rulemakings, with a few adjustments drawn from the Department’s long history of administering the part 541 regulations. The chosen population—all full-time salaried workers—represents the broadest pool of workers who could potentially be denied overtime pay as bona fide EAP workers. The BLS data for this pool is readily available and transparent (all full-time salaried workers in the CPS data set are included), and at the 40th percentile level is representative of those employees who may be bona fide executive, administrative or professional workers. The Department has proposed raising the salary percentile to the 40th percentile in part to reflect our conclusion that the 20th percentile figure used in the 2004 Final Rule did not fully account for the elimination of the more stringent long duties test; by updating the long—rather than the short—test salary level, and effectively pairing it with the less rigorous short duties test, we inadvertently made the exemptions over-inclusive and increased the risk of misclassification. The proposed salary level percentile reflects the Department’s best estimate of the appropriate line of demarcation between exempt and nonexempt workers, and maintaining the salary level at the 40th percentile by updating it annually would ensure that the salary level test VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 continues to fulfill its intended purpose. Further, because annual salary level updates would be based on actual salaries that employers are currently paying, it is consistent with the methodology the Department has used in prior rulemakings when setting the required salary level. Other factors make the fixed percentile approach well-suited for automatic updating. For example, on a quarterly basis, BLS publishes a table of deciles of the weekly wages of full-time salaried workers, calculated using CPS data,37 which would provide employers with information on changes in salary levels prior to the annual updates. While employers may be more familiar with the CPI–U, the quarterly publishing of weekly earnings deciles would provide employers with information on changes in wages and allow them to plan for changes in the salary threshold. The Department would be able to update the salary level test annually using this published BLS table, without modifying the data in any way or otherwise engaging in complex data analysis. This transparent process would further the President’s instruction to simplify and modernize the part 541 regulations. It would also ensure that salary level updates occur in a manner established in the regulations and, thus, do not require additional, time-consuming notice and comment rulemaking. Additionally, maintaining the standard salary level test at the 40th percentile would ensure that increases in overtime-protected employee salaries do not render the salary level threshold obsolete; such increases have lessened the effectiveness of the salary level test in the past when they were not promptly recognized. For all of these reasons, the Department believes that automatically updating the standard salary level test annually by maintaining the salary level at the 40th percentile of weekly earnings for all full-time salaried workers would ensure the standard salary level remains a meaningful test for distinguishing between overtimeprotected and potentially exempt white collar employees. 2. Automatically Updating the Standard Salary Level Using the CPI–U The Department could also automatically update the salary level test based on changes to the CPI–U—a commonly used economic indicator for measuring inflation. The CPI–U calculates inflation by measuring the average change over time in the prices paid by urban consumers for a set basket 37 https://www.bls.gov/cps/research_series_ earnings_nonhourly_workers.htm. PO 00000 Frm 00026 Fmt 4701 Sfmt 4702 of consumer goods and services.38 The CPI–U is the ‘‘broadest and most comprehensive’’ of the many CPI statistics calculated by BLS, and is published monthly.39 The Department has generally discussed inflation adjustments in the context of determining how to raise the salary level from a prior rulemaking, not as a method for ensuring the salary level’s ongoing effectiveness. The Department has expressed concern in prior part 541 rulemakings with setting a new salary level test by using inflationary indicators to update the prior salary level. These sentiments were first raised in 1949 in the Weiss Report, which considered and rejected proposals to use cost-of-living increases to update the 1940 salary levels. Weiss Report at 12. More recently, in the 2003 NPRM the Department considered whether to calculate the new salary level by adjusting the 1975 salary levels for inflation, and expressed concern that the 1975 figure was a potentially inaccurate benchmark and that an inflation-based adjustment would not account for changes in working conditions over the preceding 28 years. See 68 FR 15570. We also noted in the 2003 NPRM that setting the salary level based on inflation was inconsistent with the Department’s past practice of looking at actual salaries and incomes, not inflation-adjusted amounts, id., and we expressed concern in the 2004 Final Rule that this approach ‘‘could have an inflationary impact or cause job losses.’’ 69 FR 22168. Although the Department acknowledges these prior concerns regarding whether the CPI–U will accurately track the actual salaries and incomes, we believe that using the CPI– U to update the proposed salary level, which will be set using current data on wages being paid to full-time salaried workers, would ensure that the salary level remains a useful tool for distinguishing between overtimeeligible white collar employees and those who may be exempt. Many of the concerns raised in prior rulemakings are less troublesome here because the Department is only proposing to use the CPI–U to automatically update the proposed salary level going forward; it is not being used to update the salary from its 2004 level. The related concerns about using an outdated salary level as a baseline for inflation-based adjustments, and the inability of inflation-based indicators to account for changes in working conditions, are not cause for concern in the context of 38 https://stats.bls.gov/cpi/cpifaq.htm. 39 Id. E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules automatically updating a newly set salary level going forward. The proposed salary level provides the most appropriate baseline to subsequently update using the CPI–U, and year-toyear changes in working conditions should be negligible (especially compared to the changes between 1975 and 2004). While the Department considers it unlikely that cumulative changes in job duties, compensation practices, and other relevant working conditions would undermine application of the CPI–U over an extended period of time, should such changes occur the Department could adjust the salary level test through notice and comment rulemaking. The Department expressed concern in the 2003 NPRM about the effect that adjusting the 1975 salary levels for inflation ‘‘would have on certain segments of industry and geographic areas of the county, particularly in the retail industry and in the South, which tend to pay lower salaries.’’ 68 FR 15570. In the 2004 Final Rule the Department explained that these concerns applied ‘‘equally when considering automatic increases to the salary levels’’ and declined to adopt commenter requests to institute a mechanism for automatically updating the salary level. 69 FR 22171–72. The Department continues to believe that any automatic updating mechanism must adequately protect low-wage industries and geographic areas. However, two related factors have led the Department to conclude that updating the salary level using the CPI– U would not harm vulnerable business sectors or have other negative economic effects. First, the Department’s proposal to set the salary level test at the 40th percentile of the salaries of all full-time salaried workers already accounts for and protects low-wage industries and geographic areas. In choosing to set the salary level as a percentile of full-time salaried workers, the Department set the salary level at the 40th percentile rather than a higher percentile to account for low-wage regions and industries. Second, the Department has analyzed the historical relationship between the 40th percentile benchmark and the CPI– U, and determined that the data does not substantiate the Department’s past concerns about the likely effects on lowwage regions and industries of updating the salary level test using an inflationbased updating mechanism. As discussed in section VII.E., the CPI–U has largely tracked the earnings rates of the 40th percentile of weekly wages of full-time salaried workers. The two updating methodologies are thus expected to produce roughly equivalent VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 salary growth in the future; or, put another way, past evidence suggests that updating the salary level using the CPI– U would result in a comparable salary level to updating using the fixed percentile approach. Since the 40th percentile figure adequately protects low-wage industries and areas, it follows that CPI–U based updating would do likewise, while also maintaining the appropriate line of demarcation between white collar workers who are overtime-eligible and those who are not. This congruence also supports the conclusion that updating the salary level using the CPI–U, as opposed to actual salary and income data, would not produce an appreciably different result. Automatically updating the salary level test using the CPI–U would provide a familiar and well understood method for updating the salary level and ensure that the real value of the salary level does not degrade over time. The CPI–U is commonly applied as an automatic updating mechanism. For example, the Internal Revenue Service uses the CPI–U to adjust personal tax brackets, 26 U.S.C. 1(f)(3)–(5), and multiple federal agencies use the CPI–U to determine eligibility for a wide range of government programs.40 And although it was not intended to serve as a precedent for future rulemakings, in 1975 the Department set salary levels using the consumer price index. 40 FR 7092. Most importantly, given the comparable growth rates of the 40th percentile benchmark and the CPI–U between 1998 and 2013, the Department believes that updating the salary levels using the CPI–U would maintain the effectiveness of the standard salary level test. The Department seeks comments on both methods to update the standard salary level test—the fixed percentile approach and the CPI–U—including comments on whether one approach is better suited to maintaining the effectiveness of the salary level test. Additionally, the Department seeks comments on whether to schedule updates based on the effective date of the Final Rule, on January 1, or some other specified date. The Department also seeks comments on how often automatic updates to the salary level test should occur. In order to ensure that the salary level tests are based on the best available data, the Department proposes to update the salary level annually, which will produce predictable and incremental changes. However, we seek comments identifying PO 00000 40 See whether a different updating period would be more appropriate. v. Automatic Updates to the Special Salary Test for American Samoa As discussed in subpart V.B., the Department has historically set a special salary test for employees in American Samoa because minimum wage rates there are lower than the federal minimum wage. This gap is likely to remain for the foreseeable future since American Samoa’s industry-specific minimum wage rates are scheduled to increase only every three years (Sec. 4, Pub. L. 112–149), and as a result the industry with the highest minimum wage will not equal the current federal minimum wage ($7.25 an hour) until September 30, 2027. Consistent with the 2004 Final Rule, the Department is proposing to set the special salary level for employees in American Samoa at 84 percent of the proposed standard salary level ($774 per week). In future years, the Department proposes to automatically update the special salary level test in American Samoa with the same frequency as the standard salary level and to maintain the 84 percent ratio. The Department will publish the updated American Samoa special salary level and standard salary level simultaneously. Once any industry-specific minimum wage rate in American Samoa equals the federal minimum wage, the special salary level will no longer be operative and the standard salary level test will apply in full to all EAP employees in all industries in American Samoa. The Department seeks comments on this proposal. vi. Automatic Updates to the Base Rate for Motion Picture Industry Employees As discussed in subpart V.B., the Department is proposing to increase the base rate for the motion picture industry exception from the salary basis requirement with the same frequency and by the same percentage as the proposed increase to the standard salary level test. This updating method will ensure that the base rate remains a meaningful test for helping determine exempt status for motion picture industry employees who work partial workweeks and are paid a daily rate, rather than a weekly salary. The Department will publish the updated base rate and the standard salary test level simultaneously. The Department seeks comments on this proposal. https://fas.org/sgp/crs/misc/R42000.pdf. Frm 00027 Fmt 4701 Sfmt 4702 38541 E:\FR\FM\06JYP2.SGM 06JYP2 38542 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 vii. Proposal for Automatically Updating the Total Annual Compensation Requirement for Highly Compensated Employees The Department is also proposing to automatically update the total annual compensation requirement for highly compensated employees. This change is needed to ensure that only those who are ‘‘at the very top of [the] economic ladder’’ satisfy the total annual compensation requirement and are thus subject to a minimal duties test analysis. 69 FR 22174. Leaving the total annual compensation requirement at a fixed dollar amount would risk exempting increasingly large numbers of employees, thus diluting the effectiveness of the HCE total annual compensation test and allowing exemption of increasing numbers of employees who do not meet the standard duties test. Id. Only by automatically updating the requirement so that it does not become obsolete can the Department ensure that the workers who satisfy the HCE compensation test continue to ‘‘almost invariably . . . meet all the other requirements’’ for exemption. Id. The Department proposes to update the HCE total annual compensation requirement with the same method and frequency used to update the standard salary level test—either by maintaining the required total annual compensation level at the annualized value of the 90th percentile of the weekly wages of all full-time salaried workers or by updating the total annual compensation requirement based on changes in the CPI–U. As discussed with regard to the standard salary level, either method for updating the required compensation would preserve what the Department has identified as the appropriate dividing line for the use of the minimal duties test. The Department also proposes to update the portion of the total annual compensation that employers are required to pay on a salary basis (proposed to be $921 per week) so that it continues to mirror the standard salary requirement as it is updated. The Department seeks comments on both methods of updating the HCE total annual compensation requirement, including comments on whether one method is better suited to maintaining the effectiveness of the compensation test. F. Duties Requirements for Exemption While the Department has long viewed the salary level test as an initial bright-line test for white collar overtime eligibility, we have always recognized the salary level test works in tandem VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 with the duties test. As previously explained, the part 541 regulations set forth three criteria that, in most instances, must be met for an employee to be excluded from the Act’s minimum wage and overtime pay protections. Employees must (1) be paid on a salary basis, (2) be paid at least a fixed minimum salary per week, and (3) meet certain requirements as to their job duties.41 From the outset, examination of the duties performed by the employee was an integral part of the determination of exempt status, and employers must establish that the employee’s ‘‘primary duty’’ is the performance of exempt work in order for the exemption to apply. Each of the categories included in section 13(a)(1) has separate duties requirements. From 1949 until 2004 the regulations contained two different duties tests for executive, administrative, and professional employees depending on the salary level paid—a long duties test for employees paid a lower salary, and a short duties test for employees paid at a higher salary level. The long duties test included a 20 percent limit on the time spent on nonexempt tasks (40 percent for employees in the retail or service industries). In the 2004 Final Rule, the Department replaced the differing short and long duties tests with a single standard test for executive, administrative, and professional employees that did not include a cap on the amount of nonexempt work that could be performed. The duties test has always worked in conjunction with the salary requirement to correctly identify exempt EAP employees. The Department has often noted that as salary levels rise a less robust examination of the duties is needed. This inverse correlation between the salary level and the need for an extensive duties analysis was the basis of the historical short and long duties tests. While the salary provides an initial bright-line test for EAP exemption, application of a duties test is imperative to ensure that overtimeeligible employees are not swept into the exemption. While the contours of the duties tests have evolved over time, the Department has steadfastly maintained that meeting a duties test remains a core requirement for the exemptions.42 41 The exemptions for outside sales employees, doctors, lawyers, teachers, and computer employees are distinct from the other exemptions with respect to their salary requirements. 42 Over the years since the original EAP regulations were first implemented, commenters have repeatedly suggested that salary should be the sole basis for the exemption. For example, at a 1949 hearing, ‘‘some of the management witnesses were PO 00000 Frm 00028 Fmt 4701 Sfmt 4702 During the stakeholder listening sessions held in advance of this proposed rule, the Department heard from employer stakeholders, particularly in the retail and restaurant industries, who advocated for the need to maintain flexibility in the duties tests. These stakeholders stated that the ability of a store or restaurant manager or assistant manager to ‘‘pitch in’’ and help line employees when needed was a key part of their organizations’ management culture and necessary to enhancing the customer experience. They emphasized that the employees in these entry-level management positions are critically important to their organizations and that the experience they gain in these positions will lead to higher level management opportunities. Employer stakeholders universally urged the Department not to consider any changes to the current duties tests, explaining that while the duties tests are sometimes difficult to apply and may not be perfect, employers have an understanding of the meaning and application of the current duties tests and any changes might engender costly litigation as parties try to adapt to and interpret the new rules. Employee stakeholders, on the other hand, stated that the current duties tests, particularly the 50 percent primary duty rule of thumb (§ 541.700(c)) and the concurrent duties doctrine for executives (§ 541.106), are insufficiently protective of employees. In particular, they expressed concern with cases in which the exemption has been applied to employees who have spent large amounts of time (sometimes more than 90 percent) performing nonexempt work. They asserted that some businesses, particularly in the retail industry, have built into their business model having exempt store managers perform significant amounts of nonexempt work in order to keep labor costs down. These employee stakeholders argued that where employees are essentially required to perform significant amounts of nonexempt work, the employees do not, in fact, have a primary duty of management in any meaningful sense. sufficiently convinced of the desirability of salary tests to propose the adoption of a salary level as the sole basis of exemption.’’ Weiss Report at 9. The Department declined to use salary as the sole basis for exemption, stating that the ‘‘Administrator would undoubtedly be exceeding his authority if he included within the definition of these terms craftsmen, such as mechanics, carpenters, or linotype operators, no matter how highly paid they might be.’’ Weiss Report at 23. As recently as the 2004 Final Rule, the Department has maintained the view ‘‘that the Secretary does not have authority under the FLSA to adopt a ‘salary only’ test for exemption’’ and rejected suggestions from employer groups to do so. 69 FR 22173. E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules In response to this concern, a few employer stakeholders argued that the concurrent duties regulation already addresses this issue by distinguishing between exempt executive employees who choose when to perform nonexempt duties and nonexempt employees who must perform duties as they are assigned. § 541.106(a). The Department appreciates the views shared by employer and employee stakeholders on this important issue. The Department understands the importance of managers ‘‘pitching in’’ and leading by example. At the same time, the Department is concerned that employees in lower-level management positions may be classified as exempt and thus ineligible for overtime pay even though they are spending a significant amount of their work time performing nonexempt work. The Department believes that, at some point, a disproportionate amount of time spent on nonexempt duties may call into question whether an employee is, in fact, a bona fide EAP employee. The Department is concerned that the removal of the more protective long duties test in 2004 has exacerbated these concerns and led to the inappropriate classification as EAP exempt of employees who pass the standard duties test but would have failed the long duties test. The issue sometimes arises when a manager is performing exempt duties less than 50 percent of the time, but it is argued that those duties are sufficiently important to nonetheless be considered the employee’s primary duty. The issue also arises when a manager who is performing nonexempt duties much of the time is deemed to perform exempt duties concurrently with those nonexempt duties, and it is argued the employee is exempt on that basis. While the regulations provide that exempt executives can perform exempt duties concurrently with nonexempt duties, § 541.106, this rule can be difficult to apply and can lead to varying results. Compare In re Family Dollar FLSA Litigation, 637 F.3d 508 (4th Cir. 2011) (manager of retail chain store considered an executive exempt from overtime pay requirements under the FLSA whether collecting cash, sweeping the floor, stocking shelves, working with employee schedules, or running a cash register); with Morgan v. Family Dollar Stores, Inc., 551 F.3d 1233 (11th Cir. 2008) (store managers not exempt executives where they spent most of their time performing manual, not managerial, tasks). California has addressed this issue by requiring that exempt EAP employees spend at least 50 percent of their time performing their VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 primary duty, and not counting time during which nonexempt work is performed concurrently. Cal. Lab. Code Sec. 515(a), (e); see Heyen v. Safeway Inc., 157 Cal. Rptr. 3d 280, 302 (Cal. Ct. App. 2013). Taking into account the views of stakeholders, the Department is seeking to determine whether, in light of our salary level proposal, changes to the duties tests are also warranted. The duties test must adequately protect overtime-eligible white collar employees who exceed the salary threshold from misclassification as exempt EAP employees. The Department is proposing to set the salary threshold at the 40th percentile of weekly earnings of fulltime salaried employees. As previously discussed, because the standard duties test is based on the short duties test— which was intended to work with a higher salary level—and the proposed salary level is below the historic average for the short test salary, a salary level significantly below the 40th percentile would necessitate a more robust duties test to ensure proper application of the exemption. The Department believes that the salary level increase proposed in this NPRM, coupled with automatic updates to maintain the effectiveness of the salary level test, will address most of the concerns relating to the application of the EAP exemption. A regularly updated salary level will assist in screening out employees who spend significant amounts of time on nonexempt duties and for whom exempt work is not their primary duty. However, the Department invites comments on whether adjustments to the duties tests are necessary, particularly in light of the proposed change in the salary level test. The Department recognizes that duties remain a critical metric of exempt status and invites comment on the effectiveness of the duties tests found in the current regulations. While the Department is not proposing specific regulatory changes at this time, the Department is seeking additional information on the duties tests for consideration in the Final Rule. Specifically, the Department seeks comments on the following issues: A. What, if any, changes should be made to the duties tests? B. Should employees be required to spend a minimum amount of time performing work that is their primary duty in order to qualify for exemption? If so, what should that minimum amount be? C. Should the Department look to the State of California’s law (requiring that 50 percent of an employee’s time be PO 00000 Frm 00029 Fmt 4701 Sfmt 4702 38543 spent exclusively on work that is the employee’s primary duty) as a model? Is some other threshold that is less than 50 percent of an employee’s time worked a better indicator of the realities of the workplace today? D. Does the single standard duties test for each exemption category appropriately distinguish between exempt and nonexempt employees? Should the Department reconsider our decision to eliminate the long/short duties tests structure? E. Is the concurrent duties regulation for executive employees (allowing the performance of both exempt and nonexempt duties concurrently) working appropriately or does it need to be modified to avoid sweeping nonexempt employees into the exemption? Alternatively, should there be a limitation on the amount of nonexempt work? To what extent are exempt lower-level executive employees performing nonexempt work? In addition to seeking comments on the duties tests, the Department is also considering whether to add to the regulations examples of additional occupations to provide guidance in administering the EAP exemptions. Employer stakeholders have indicated that examples of how the exemptions may apply to specific jobs, such as those provided in current §§ 541.203, 541.301(e), and 541.402, are useful in determining exempt status and should be expanded. The Department agrees that examples of how the general executive, administrative, and professional exemption criteria may apply to specific occupations are useful to the regulated community and seeks comments on what specific additional examples of nonexempt and exempt occupations would be most helpful to include. Computer Related Occupations In further effort to provide effective guidance to the public on the administration of the EAP exemptions, the Department is considering the suggestions of employer stakeholders from the computer and information technology sectors to include additional examples of the application of the EAP exemptions to occupational categories in computer-related fields. The Department has, as a threshold matter, reviewed the authority by which it might include additional examples of computer-related occupations. For the reasons articulated in the preamble to the 2004 Final Rule, the Department continues to believe that we should not expand the EAP exemption beyond the computer exemption currently set forth in section 13(a)(17), given the clarity E:\FR\FM\06JYP2.SGM 06JYP2 38544 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 with which Congress has set forth the scope of that exemption.43 69 FR 22160. However, in the 2004 Final Rule, the Department did add additional examples of occupations within the computer industry such as systems analysts and computer programmers which, subject to a case-by-case duties analysis, might fall within the section 13(a)(1) administrative and executive exemptions. § 541.402. In response to stakeholder input and as part of our broader effort to simplify part 541, the Department is again exploring the possibility of listing additional illustrative examples that typically do or do not fall within the general criteria for the three basic EAP exemptions (see §§ 541.100, .200, .300), as opposed to those falling within the computerspecific exemption set forth in section 13(a)(17), to bring additional clarity to employers and employees within the computer and information technology industries. The Department continues to be cognizant of the ‘‘tremendously rapid pace of significant changes occurring in the information technology industry’’ (69 FR 22158), and therefore requests comments from employer and employee stakeholders in the computer and information technology sectors as to what additional occupational titles or categories should be included as examples in the part 541 regulations, along with what duties are typical of such categories and would thus cause them to generally meet or fail to meet the relevant EAP exemption criteria. To provide additional context, the Department, as an initial matter, expresses the view that a help desk operator whose responses to routine computer inquiries (such as requests to reset a user’s password or address a system lock-out) are largely scripted or dictated by a manual that sets forth well-established techniques or procedures would not possess the discretion and independent judgment necessary for the administrative exemption, nor would that individual likely qualify for any other EAP exemption. On the other hand, an information technology specialist who, without supervision, routinely 43 Although the 1990 amendments to the FLSA afforded the Department some discretion to elaborate on computer-specific exemption criteria distinct from the standard EAP exemption criteria (Sec. 2, Pub. L. 101–583, 104 Stat 2871 (Nov. 15, 1990)), the Department concluded in the 2004 Final Rule that, because Congress subsequently codified the criteria for a computer employee exemption in FLSA section 13(a)(17) (Sec. 2105(a), Pub. L. 104– 188, 110 Stat. 1755 (Aug. 20, 1996)), it would be ‘‘inappropriate’’ to engage in further rulemaking after Congress had spoken on the issue. 69 FR 22160. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 troubleshoots and repairs significant glitches in his company’s point of sale software for the company’s retail clients might be an example of an administrative employee pursuant to § 541.200 as this employee’s work appears to be directly related to the management or business operation of his employer or employer’s customers and requires the use of discretion and independent judgment with respect to matters of significance. VI. Paperwork Reduction Act The Paperwork Reduction Act of 1995 (PRA), 44 U.S.C. 3501 et seq., and its attendant regulations, 5 CFR part 1320, require the Department to consider the agency’s need for its information collections, their practical utility, as well as the impact of paperwork and other information collection burdens imposed on the public, and how to minimize those burdens. The PRA typically requires an agency to provide notice and seek public comments on any proposed collection of information contained in a proposed rule. See 44 U.S.C. 3506(c)(2)(B); 5 CFR 1320.8. Persons are not required to respond to the information collection requirements until they are approved by the Office of Management and Budget (OMB) under the PRA. This NPRM would revise the existing information collection requirements previously approved under OMB control number 1235–0018 (Records to be Kept by Employers—Fair Labor Standards Act) and OMB control number 1235–0021 (Employment Information Form) in that employers would need to maintain records of hours worked for more employees and more employees may file complaints to recover back wages under the overtime pay provision. As required by the PRA, the Department has submitted the information collection revisions to OMB for review in order to reflect changes that would result from this proposed rule were it to be adopted. Summary: FLSA section 11(c) requires all employers covered by the FLSA to make, keep, and preserve records of employees and of wages, hours, and other conditions of employment. A FLSA-covered employer must maintain the records for such period of time and make such reports as prescribed by regulations issued by the Secretary of Labor. The Department has promulgated regulations at 29 CFR part 516 to establish the basic FLSA recordkeeping requirements. No new information collection requirements would be imposed by the adoption of this NPRM; rather, burdens under existing requirements are expected to increase as more employees receive PO 00000 Frm 00030 Fmt 4701 Sfmt 4702 minimum wage and overtime protections due to the proposed increase in the salary level requirement. More specifically, the proposed changes in this NPRM may cause an increase in burden on the regulated community because employers will have additional employees to whom certain longestablished recordkeeping requirements apply (e.g., maintaining daily records of hours worked by employees who are not exempt from the both minimum wage and overtime provisions). Additionally, the proposed changes in this NPRM may cause an initial increase in burden if more employees file a complaint with WHD to collect back wages under the overtime pay requirements. We anticipate this increased burden will wane over time as employers adjust to the new rule. Purpose and Use: WHD and employees use employer records to determine whether covered employers have complied with various FLSA requirements. Employers use the records to document compliance with the FLSA, including showing qualification for various FLSA exemptions. Additionally, WHD uses the Employment Information form to document allegations of noncompliance with labor standards the agency administers. Technology: The regulations prescribe no particular order or form of records and employers may preserve records in forms of their choosing provided that facilities are available for inspection and transcription of the records. Minimizing Small Entity Burden: Although the FLSA recordkeeping requirements do involve small businesses, including small state and local government agencies, the Department minimizes respondent burden by requiring no specific order or form of records in responding to this information collection. Burden is reduced on complainants by providing a template to guide answers. Public Comments: As part of its continuing effort to reduce paperwork and respondent burden, the Department conducts a preclearance consultation program to provide the general public and Federal agencies with an opportunity to comment on proposed and continuing collections of information in accordance with the PRA. This program helps to ensure that requested data can be provided in the desired format, reporting burden (time and financial resources) is minimized, collection instruments are clearly understood, and the impact of collection requirements on respondents can be properly assessed. The Department seeks public comments regarding the E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules burdens imposed by the information collections associated with this NPRM. Commenters may send their views about this information collection to the Department in the same manner as all other comments (e.g., through the regulations.gov Web site). All comments received will be made a matter of public record and posted without change to https://www.regulations.gov, including any personal information provided. As previously noted, an agency may not conduct an information collection unless it has a currently valid OMB approval, and the Department has submitted information collection requests under OMB control numbers 1235–0018 and 1235–0021 in order to update them to reflect this rulemaking and provide interested parties a specific opportunity to comment under the PRA. See 44 U.S.C. 3507(d); 5 CFR 1320.11. Interested parties may receive a copy of the full supporting statements by sending a written request to the mail address shown in the ADDRESSES section at the beginning of this preamble. In addition to having an opportunity to file comments with the Department, comments about the paperwork implications may be addressed to OMB. Comments to OMB should be directed to: Office of Information and Regulatory Affairs, Attention OMB Desk Officer for the Wage and Hour Division, Office of Management and Budget, Room 10235, Washington, DC 20503; Telephone: 202–395–7316/Fax: 202–395–6974 (these are not toll free numbers). OMB will consider all written comments that the agency receives within 30 days of publication of this proposed rule. Commenters are encouraged, but not required, to send the Department a courtesy copy of any comments sent to OMB. The courtesy copy may be sent via the same channels as comments on the rule. OMB and the Department are particularly interested in comments that: • Evaluate whether the proposed collections of information are necessary for the proper performance of the functions of the agency, including whether the information will have practical utility; • Evaluate the accuracy of the agency’s estimate of the burden of the proposed collection of information, including the validity of the methodology and assumptions used; • Enhance the quality, utility, and clarity of the information to be collected; and • Minimize the burden of the collection of information on those who are to respond, including through the use of appropriate automated, VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 electronic, mechanical, or other technological collection techniques or other forms of information technology (e.g., permitting electronic submission of responses). Total annual burden estimates, which reflect both the existing and new responses for the recordkeeping and complaint process information collections at the proposed salary, are summarized as follows: Type of Review: Revisions to currently approved information collections. Agency: Wage and Hour Division, Department of Labor. Title: Records to be Kept by Employers—Fair Labor Standards Act. OMB Control Number: 1235–0018. Affected Public: Private sector businesses or other for-profits, farms, not-for-profit institutions, state, local and tribal governments, and individuals or households. Estimated Number of Respondents: 3,771,434 (unaffected by this rulemaking). Estimated Number of Responses: 50,467,523 (6,909,600 added by this rulemaking). Estimated Burden Hours: 1,235,161 hours (230,320 added by this rulemaking). Estimated Time per Response: Various (unaffected by this rulemaking). Frequency: Various (unaffected by this rulemaking). Other Burden Cost: 0. Title: Employment Information Form. OMB Control Number: 1235–0021. Affected Public: Businesses or other for-profit, farms, not-for-profit institutions, state, local and tribal governments, and individuals or households. Total Respondents: 38,138 (2,788 added by this rulemaking). Estimated Number of Responses: 38,138 (2,788 added by this rulemaking). Estimated Burden Hours: 12,713 (930 hours added by this rulemaking). Estimated Time per Response: 20 minutes (unaffected by this rulemaking). Frequency: once. Other Burden Cost: 0. VII. Analysis Conducted in Accordance With Executive Order 12866, Regulatory Planning and Review, and Executive Order 13563, Improving Regulation and Regulatory Review Executive Orders 12866 and 13563 direct agencies to assess all costs and benefits of available regulatory alternatives and, if the regulation is necessary, to select regulatory approaches that maximize net benefits (including potential economic, environmental, public health and safety PO 00000 Frm 00031 Fmt 4701 Sfmt 4702 38545 effects, distributive impacts, and equity). Executive Order 13563 emphasizes the importance of quantifying both costs and benefits, of reducing costs, of harmonizing rules, and of promoting flexibility. Under Executive Order 12866, the Department must determine whether a regulatory action is economically ‘‘significant,’’ defined as having an annual effect on the economy of $100 million or more, and therefore subject to review by OMB and the requirements of the Executive Order. This proposed rule is economically significant within the meaning of Executive Order 12866; therefore, the Department has prepared a Preliminary Regulatory Impact Analysis (PRIA) in connection with this proposed rule as required under section 6(a)(3) of Executive Order 12866, and OMB has reviewed the rule. A. Introduction i. Background The FLSA applies to all enterprises that have employees engaged in commerce or in the production of goods for commerce and have an annual gross volume of sales made or business done of at least $500,000 (exclusive of excise taxes at the retail level that are separately stated); or are engaged in the operation of a hospital, an institution primarily engaged in the care of the sick, the aged, or individuals with intellectual disabilities who reside on the premises; a school for intellectually or physically disabled or gifted children; a preschool, elementary or secondary school, or an institution of higher education (without regard to whether such hospital, institution or school is public or private, or operated for profit or not); or are engaged in an activity of a public agency. See 29 U.S.C. 203(s). There are two ways an employee may be covered by the provisions of the FLSA: (1) Enterprise coverage, in which any employee of an enterprise covered by the FLSA is covered, and (2) individual coverage, in which even employees of non-covered enterprises may be covered if they are engaged in interstate commerce or in the production of goods for commerce, or are employed in domestic service. The FLSA requires employers to: (1) Pay employees who are covered and not exempt from the Act’s requirements not less than the Federal minimum wage for all hours worked and overtime premium pay at a rate of not less than one and one-half times the employee’s regular rate of pay for all hours worked over 40 in a workweek, and (2) make, keep, and preserve records of the persons E:\FR\FM\06JYP2.SGM 06JYP2 38546 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules employed by the employer and of the wages, hours, and other conditions and practices of employment. It is widely recognized that the general requirement that employers pay a premium rate of pay for all hours worked over 40 in a workweek is a cornerstone of the Act, grounded in two policy objectives. The first is to spread employment by incentivizing employers to hire more employees rather than requiring existing employees to work longer hours, thereby reducing involuntary unemployment. The second policy objective is to reduce overwork and its detrimental effect on the health and well-being of workers. The FLSA provides a number of exemptions from the Act’s minimum wage and overtime pay provisions, including one for bona fide executive, administrative, and professional employees. Such employees typically receive more monetary and nonmonetary benefits than most blue collar and lower-level office workers and therefore are less likely to need the Act’s protection. Thus, Congress created the exemption from the FLSA’s minimum wage and overtime pay protections for employees employed in a bona fide executive, administrative, or professional capacity and for outside sales employees, as those terms are ‘‘defined and delimited’’ by the Department. 29 U.S.C. 213(a)(1). The Department’s regulations implementing those exemptions are codified at 29 CFR part 541. For an employer to exclude an employee from minimum wage and overtime protection pursuant to the EAP exemptions, the employee generally must meet three criteria: (1) The employee must be paid a predetermined and fixed salary that is not subject to reduction because of variations in the quality or quantity of work performed (the ‘‘salary basis test’’); (2) the amount of salary paid must meet a minimum specified amount (the ‘‘salary level test’’); and (3) the employee’s job duties must primarily involve executive, administrative, or professional duties as defined by the regulations (the ‘‘duties test’’). The regulations governing these tests have been updated periodically since the FLSA’s enactment in 1938, most recently in 2004 when, among other revisions, the Department increased the salary level test to $455 per week. As a result of inflation and the low value of the salary threshold, the annual value of this salary level test, $23,660 ($455 per week for 52 weeks), is now slightly below the 2014 poverty threshold for a family of four ($24,008),44 making it inconsistent with Congress’ intent to exempt only bona fide EAP workers, who typically earn salaries well above those of any workers they may supervise and presumably enjoy other privileges of employment such as above average fringe benefits, greater job security, and better opportunities for advancement. Stein Report at 21–22. In the 2004 Final Rule, the Department also changed the structure of the duties test. Between 1949 and 2004, the EAP exemptions included two versions of the duties test. Assuming that a worker was paid on a salary basis, the exemptions would be met if a worker passed either a ‘‘long’’ test, which involved a more rigorous set of duties criteria, paired with a lower salary level, or a ‘‘short’’ test, which imposed fewer duties requirements, paired with a higher salary level. In the 1975 update, the last before the 2004 Final Rule, the Department set the long test salary levels at $155 per week for executive and administrative employees and $170 per week for professional employees. The short test salary level was set at $250 per week for all three EAP categories. In 2004, the Department replaced the two-test structure with a single ‘‘standard’’ duties test for each category, which closely resembles the former short test duties requirements, and a single salary level test of $455 per week based on an update of the 1975 long test salary level. The Department also introduced a highly compensated employee (HCE) exemption in the 2004 Final Rule, under which an employee may be exempt if he or she passes a very minimal duties test, receives at least $455 per week paid on a salary basis, and is highly compensated, defined in the 2004 Final Rule as earning total annual compensation, which may include commissions and nondiscretionary bonuses in addition to a salary, of at least $100,000. The HCE duties test is much more abbreviated than the historical short test duties requirements. The premise behind the salary level tests is that employers are more likely to pay higher salaries to workers in bona fide EAP jobs than to workers performing nonexempt duties. A high salary is considered a measure of an employer’s good faith in classifying an employee as exempt, because an employer is less likely to have misclassified a worker as exempt if he or she is paid a high wage. Stein Report at 5; Weiss Report at 8. The salary level requirement was created to identify the dividing line distinguishing workers performing truly exempt duties from the nonexempt workers Congress intended to be protected by the FLSA’s minimum wage and overtime provisions. Throughout the regulatory history of the FLSA, the Department has considered the salary level test the ‘‘best single test’’ of exempt status. Stein Report at 19. This bright-line test is easily observed, objective, and clear. Id. ii. Need for Rulemaking The salary level test has been updated seven times since it was implemented in 1938. Table 1 presents the weekly salary levels associated with the EAP exemptions since 1938, organized by exemption and long/short/standard duties test. TABLE 1—HISTORICAL SALARY LEVELS FOR THE EAP EXEMPTIONS Long test Date enacted Short test (all) tkelley on DSK3SPTVN1PROD with PROPOSALS2 Executive 1938 1940 1949 1958 1963 ............................................................................................................. ............................................................................................................. ............................................................................................................. ............................................................................................................. ............................................................................................................. Administrative $30 30 55 80 100 $30 $50 75 95 100 44 The 2014 poverty threshold for a family of four with two related people under 18 in the household. Available at: https://www.census.gov/hhes/www/ poverty/data/threshld/. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00032 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 Professional ........................ $50 75 95 115 ........................ ........................ $100 125 150 38547 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 1—HISTORICAL SALARY LEVELS FOR THE EAP EXEMPTIONS—Continued Long test Date enacted Short test (all) Executive 1970 ............................................................................................................. 1975 ............................................................................................................. Administrative 125 155 125 155 Professional 140 170 200 250 Standard Test 2004 ............................................................................................................. $455 substantially, and as a result increasingly more workers earn above the salary threshold. By 2013 the real value of the standard salary level had declined 18.9 percent since 2004, calculated using the Consumer Price Index for all urban consumers (CPI– U).45 Figure 1 demonstrates how the real values of the salary levels have changed since 1949, measured in 2013 dollars. As a result of the erosion of the real value of the standard salary level, more and more workers lack the clear protection the salary level test is meant to provide. Each year that the salary level is not updated, its utility as a distinguishing mechanism between exempt and nonexempt workers declines. The Department has revised the levels just once in the 40 years since 1975. In contrast, in the 37 years between 1938 and 1975, salary test levels were increased approximately every five to nine years. In our 2004 rulemaking, the Department stated the intention to ‘‘update the salary levels on a more regular basis, as it did prior to 1975,’’ and added that ‘‘the salary levels should be adjusted when wage survey data and other policy concerns support such a change.’’ 69 FR 22171. The real value of the salary level test has fallen substantially both when measured against its 2004 level and the 1975 levels. If the standard EAP salary level established in 2004 had kept up with inflation (measured using the CPI– U), it would be $561 per week in 2013 dollars, a 23.3 percent increase relative 45 CPI–U data available at: https://data.bls.gov/cgibin/cpicalc.pl. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00033 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 EP06JY15.000</GPH> tkelley on DSK3SPTVN1PROD with PROPOSALS2 The standard salary level was set at $455 per week in 2004. Following more than ten years of inflation, the purchasing power, or real value, of the standard salary level test has eroded 38548 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules to its current level. If the EAP salary level for the short test established in 1975 had kept up with inflation, it would be $1,083 per week, a 137.9 percent increase relative to the current salary level. In order to restore the value of the standard salary level as a line of demarcation between those workers for whom Congress intended to provide minimum wage and overtime protections and those workers who may be performing bona fide EAP duties, and to maintain its continued validity, the Department proposes to set the standard salary level equal to the 40th percentile of weekly earnings for all full-time salaried workers. Based on 2013 salary data,46 this is equivalent to a standard salary level of $921 per week. The Department also proposes to automatically update the standard salary level annually in the future. Furthermore, the Department proposes to set the HCE compensation level at the 90th percentile of annualized weekly earnings for full-time salaried workers, equivalent to $122,148, and to update the level annually in the future. Automatic updating would preserve the value of these earnings thresholds, eliminate the volatility associated with previous changes in the thresholds, and provide certainty for employers with respect to future changes. It would also simplify the updating process, as the Department would simply publish a notice in the Federal Register of the updated salary and compensation thresholds on an annual basis, and additional notice and comment rulemaking to adjust the salary and annual compensation thresholds would not be necessary unless the Department determined in the future that the methodology for setting the standard salary or the HCE total compensation levels needed to be adjusted. iii. Summary of Affected Workers, Costs, Benefits, and Transfers The Department estimated the number of affected workers and quantified costs and transfer payments associated with this proposed rulemaking.47 All estimates are based on analysis of the Current Population Survey (CPS), a monthly survey of 60,000 households conducted by the U.S. Census Bureau. In 2013, there were an estimated 144.2 million wage and salary workers in the United States, of whom 128.5 million were subject to the FLSA and the Department’s regulations.48 49 Of these 128.5 million workers, the Department estimates that 43.0 million are white collar salaried employees who may be affected by a change to the Department’s part 541 regulations and are not covered by another (non-EAP) exemption.50 51 The remaining 85.5 million workers include blue collar workers, workers paid on an hourly basis, and workers eligible for another (non-EAP) overtime exemption. These workers were excluded because they will generally not be affected by this proposed rulemaking. Of the 43.0 million workers discussed above, the Department estimates that 28.5 million are exempt from the minimum wage and overtime pay provisions under the part 541 EAP exemptions, while 14.4 million do not satisfy the duties tests for EAP exemption and/or earn less than $455 per week.52 However, of the 28.5 million EAP exempt workers, 7.1 million were in ‘‘named occupations’’ and thus need only pass the duties tests to be subject to the standard EAP exemptions.53 Therefore, these workers were not considered in the analysis, leaving 21.4 million EAP exempt workers potentially affected by this proposed rule. The Department proposes to increase the standard salary level from $455 per week to the 40th percentile of weekly earnings for all full-time salaried workers, which translates to $921 per week, an increase of $466 over the current level (Table 2).54 The Department also proposes to increase the HCE annual compensation level to the 90th percentile of annualized weekly earnings for full-time salaried workers, which translates to $122,148 annually. TABLE 2—PROPOSED SALARY LEVELS Current salary level Salary level Standard exemption ......................................................................................... HCE exemption ................................................................................................ Proposed salary level $455/week 100,000/year $921/week 122,148/year Total increase $ % 466 22,148 102.4 22.1 tkelley on DSK3SPTVN1PROD with PROPOSALS2 The Department also proposes to annually update the standard salary level to ensure the ongoing effectiveness of the salary level as a means of delimiting workers who should not fall within the EAP exemption. Similarly, the Department proposes to annually update the HCE total annual compensation level to ensure the effectiveness of the annual compensation requirement as a test for which employees should be subject to the minimal duties test for the HCE exemption. In Year 1, an estimated 4.6 million workers would be affected by the increase in the standard salary level test (Table 3). This figure consists of currently EAP-exempt workers who earn at least $455 per week but less than the 40th percentile ($921) of all full- 46 Unless otherwise noted, the Department relied upon 2013 data in the development of the NPRM. The Department will update the data used in the Final Rule resulting from this proposal. 47 Because the Department has not proposed specific changes to the duties tests, potential changes to the duties tests are not included in this RIA. However, the Department discusses a potential methodology for determining the impact of any changes to the duties test in section VII.F. 48 Data on wage and salary workers are from the CPS, series ID: LNU02000000. 49 Workers not covered as employees by the FLSA and/or the Department’s regulations include: members of the military, unpaid volunteers, the self-employed, many religious workers, and most federal employees. The number of workers covered by the FLSA was estimated using the CPS Merged Outgoing Rotation Group (MORG) data. 50 As discussed in more detail later, the Department used pooled data from 2011–2013 to represent the 2013 population in order to increase sample size, and thus the granularity of results. 51 As discussed later, the Department excluded from this analysis certain workers for whom their employer could claim a non-EAP exemption from the FLSA’s minimum wage and overtime pay provisions, and certain workers for whom the employer could claim an overtime pay exemption. For simplicity, the Department refers to these exemptions as other (non-EAP) exemptions. 52 Here and elsewhere in this analysis, numbers are reported at varying levels of aggregation, and are generally rounded to a single decimal point. However, calculations are performed using exact numbers. Therefore, as in this case, some numbers may not match the reported total or the calculation shown due to rounding of components. 53 Workers not subject to the EAP salary level test include teachers, academic administrative personnel, physicians, lawyers, judges, and outside sales workers. 54 The BLS data set used for this rulemaking consists of earnings for all full-time (defined as at least 35 hours per week) non-hourly paid employees. For the purpose of this rulemaking, the Department considers data representing compensation paid to non-hourly workers to be an appropriate proxy for compensation paid to salaried workers. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00034 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38549 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules time salaried workers. Additionally, an estimated 36,000 workers would be affected by the increase in the HCE compensation test. In Year 10, with automatic updating, between 5.1 and 5.6 million workers are projected to be affected by the change in the standard salary level test and between 33,000 and 42,000 workers affected by the change in the HCE total annual compensation test, depending on the updating methodology used. TABLE 3—ESTIMATED NUMBER OF AFFECTED EAP WORKERS, TEN-YEAR PROJECTIONS WITH AND WITHOUT AUTOMATIC UPDATING Without automatic updating Affected EAP workers (1,000s) a Year 1 Standard exemption ......................................................... HCE exemption ................................................................ 4,646 36 Year 10 Updated with fixed percentiles Year 1 2,685 7 Year 10 4,646 36 5,568 42 Updated with CPI–U Year 1 4,646 36 Year 10 5,062 33 a Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels). Three direct costs to employers are quantified in this analysis: (1) Regulatory familiarization costs; (2) adjustment costs; and (3) managerial costs. Regulatory familiarization costs are the costs incurred to read and become familiar with the requirements of the rule. Adjustment costs are the costs accrued to determine workers’ new exemption statuses, notify employees of policy changes, and update payroll systems. Managerial costs associated with this proposed rulemaking occur because hours of workers who are newly entitled to overtime may be more closely scheduled and monitored to minimize or avoid paying the overtime premium. The costs presented here are the combined costs for both the change in the standard salary level test and the HCE annual compensation level (these will be disaggregated in section VII.D.iv.). With updating, total average annualized direct employer costs were estimated to be between $239.6 and $255.3 million, assuming a 7 percent discount rate; hereafter, unless otherwise specified, average annualized values will be presented using the 7 percent real discount rate (Table 4). Deadweight loss (DWL) is also a cost but not a direct employer cost. DWL is a function of the difference between the wage employers are willing to pay for the hours lost, and the wage workers are willing to take for those hours. In other words, DWL represents the decrease in total economic surplus in the market arising from the change in the regulation. Average annualized DWL was estimated to be between $9.5 and $10.5 million, depending on updating methodology. In addition to the direct costs, this proposed rulemaking will also transfer income from employers to employees in the form of wages. Average annualized transfers were estimated to be between $1,178.0 and $1,271.4 million, depending on updating methodology. The majority of these transfers are from employers to affected EAP workers who become overtime protected due to changes in the EAP regulations. Employers may incur additional costs, such as hiring new workers. These other costs are discussed in section VII.D.iv.5. Another potential impact of the rule’s proposed increase in the salary threshold is a reduction in litigation costs. Other unquantified transfers, costs, and benefits are discussed in section VII.D.vii. TABLE 4—SUMMARY OF REGULATORY COSTS AND TRANSFERS, STANDARD AND HCE SALARY LEVELS WITH AUTOMATIC UPDATING [Millions 2013$] Automatic updating method b Cost/transfer a Direct employer costs ..................................................... Transfers d ...................................................................... DWL ................................................................................ Percentile .... CPI–U .......... Percentile .... CPI–U .......... Percentile .... CPI–U .......... Future years c Average annualized value Year 1 Year 2 $592.7 592.7 1,482.5 1,482.5 7.4 7.4 Year 10 $188.8 181.1 1,160.2 1,126.4 10.8 10.3 $225.3 198.6 1,339.6 1,191.4 11.2 9.7 3% real rate 7% real rate $248.8 232.3 1,271.9 1,173.7 10.5 9.6 $255.3 239.6 1,271.4 1,178.0 10.5 9.5 a Costs and transfers for affected workers passing the standard and HCE tests are combined. percentile method sets the standard salary level at the 40th percentile of weekly earnings for full-time salaried workers and the HCE compensation level at the 90th percentile. The CPI–U method adjusts both levels based on the annual percent change in the CPI–U. c These costs/transfers represent a range over the nine-year span. d This is the net transfer from employers to workers. There may also be transfers of hours and income from some workers to other workers. tkelley on DSK3SPTVN1PROD with PROPOSALS2 b The iv. Terminology and Abbreviations The following terminology and abbreviations will be used throughout this Regulatory Impact Analysis (RIA). Affected EAP workers: The population of potentially affected EAP workers who either earn between $455 and the proposed salary level of the 40th VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 percentile of weekly earnings ($921) or qualify for the HCE exemption and earn between $100,000 and the 90th percentile of earnings ($122,148 PO 00000 annually). This is estimated to be 4.7 million workers.55 55 Setting the standard salary level at the 40th percentile is estimated to affect 4,646,000 workers. See Table 3. Additionally, 36,000 workers are potentially affected by the change in the HCE exemption’s total compensation level. Id. Continued Frm 00035 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 38550 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules BLS: Bureau of Labor Statistics. CPI–U: Consumer Price Index for all urban consumers. CPS: Current Population Survey. Duties test: To be exempt from the FLSA’s minimum wage and overtime requirements under section 13(a)(1), the employee’s primary job duty must involve bona fide executive, administrative, or professional duties as defined by the regulations. The Department distinguishes among four such tests: Standard duties test: The duties test used in conjunction with the standard salary level test, as set in 2004 and applied to date, to determine eligibility for the EAP exemptions. It replaced the short and long tests in effect from 1949 to 2004, but its criteria closely follow those of the former short test. HCE duties test: The duties test used in conjunction with the HCE compensation level test, as set in 2004 and applied to date, to determine eligibility for the HCE exemption. It is much less stringent than the standard and short duties tests to reflect that very highly paid employees are much more likely to be properly classified as exempt. Long duties test: One of two duties tests used from 1949 until 2004; this more restrictive duties test had a greater number of requirements, including a limit on the amount of nonexempt work that could be performed, and was used in conjunction with a lower salary level test to determine eligibility for the EAP exemptions (see Table 1). Short duties test: One of two duties tests used from 1949 to 2004; this less restrictive duties test had fewer requirements and was used in conjunction with a higher salary level test to determine eligibility for the EAP exemptions (see Table 1). DWL: Deadweight loss; the loss of economic efficiency that can occur when equilibrium in a market for a good or service is not achieved. EAP: Executive, administrative, and professional. HCE: Highly compensated employee; a category of EAP-exempt employee, established in 2004 and characterized by high earnings and a minimal duties test. MORG: Merged Outgoing Rotation Group supplement to the CPS. Named occupations: Workers in named occupations are not subject to Accordingly, throughout this NPRM we refer to the total affected workers as 4.7 million (4,646,000 + 36,000, rounded to the nearest 100,000 workers). However, when discussing only those workers affected by the change in the standard salary level test, the number decreases to 4.6 million (4,646,000, similarly rounded). VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 the salary level or salary basis tests. These occupations include teachers, academic administrative personnel,56 physicians,57 lawyers, and judges.58 Overtime Workers Occasional overtime workers: Workers who report they usually work 40 hours or less per week (identified with variable PEHRUSL1 in CPS MORG) but in the survey week worked more than 40 hours (variable PEHRACT1 in CPS MORG). Regular overtime workers: Workers who report they usually work more than 40 hours per week (identified with variable PEHRUSL1 in CPS MORG). Pooled data for 2011–2013: CPS MORG data from 2011–2013 with earnings inflated to 2013 dollars and sample observations weighted to reflect the population in 2013; used to increase sample size. Potentially affected EAP workers: EAP exempt workers who are not in named occupations and are included in the analysis (i.e., white collar, salaried, not eligible for another (non-EAP) overtime pay exemption). This is estimated to be 21.4 million workers. Price elasticity of demand (with respect to wage): The percentage change in labor hours demanded in response to a one percent change in wages. Real dollars (2013$): Dollars adjusted using the CPI–U to reflect the purchasing power they would have in 2013. Salary basis test: The EAP exemptions’ requirement that workers be paid on a salary basis, that is, a predetermined amount that cannot be reduced because of variations in the quality or quantity of the employee’s work. Salary level test: The salary a worker must earn in order to be subject to the EAP exemptions. The Department distinguishes among four such tests: Standard salary level: The weekly salary level associated with the standard 56 Academic administrative personnel (including admissions counselors and academic counselors) need to be paid either (1) the salary level or (2) a salary that is at least equal to the entrance salary for teachers in the educational establishment at which they are employed (see § 541.204). Entrance salaries at the educational establishment of employment cannot be distinguished in the data and so this alternative is not considered (thus these employees were excluded from the analysis, the same as was done in the 2004 Final Rule). 57 The term physician includes medical doctors including general practitioners and specialists, osteopathic physicians (doctors of osteopathy), podiatrists, dentists (doctors of dental medicine), and optometrists (doctors of optometry or with a Bachelor of Science in optometry). § 541.304(b). 58 Judges may not be considered ‘‘employees’’ under the FLSA definition. However, since this distinction cannot be made in the data, all judges are excluded from the analysis (the same as was done in the 2004 Final Rule). PO 00000 Frm 00036 Fmt 4701 Sfmt 4702 duties test that determines (in part) eligibility for the EAP exemptions. The standard salary level was set at $455 per week in the 2004 Final Rule. HCE compensation level: Workers who meet the standard salary level requirement but not the standard duties test nevertheless are exempt if they pass a minimal duties test and earn at least the HCE total annual compensation required amount. The HCE required compensation level was set at $100,000 per year in the 2004 Final Rule, of which at least $455 per week must be paid on a salary or fee basis. Short test salary level: The weekly salary level associated with the short duties test (eliminated in 2004). Long test salary level: The weekly salary level associated with the long duties test (eliminated in 2004). Workers covered by the FLSA and subject to the Department’s regulations: Includes all workers except those excluded from the analysis because they are not covered by the FLSA or subject to the Department’s requirements. Excluded workers include: members of the military, unpaid volunteers, the selfemployed, many religious workers, and federal employees (with a few exceptions).59 The Department also notes that the terms employee and worker are used interchangeably throughout this analysis. B. Methodology To Determine the Number of Potentially Affected EAP Workers i. Overview This section explains the methodology used to estimate the number and characteristics of workers who are subject to the EAP exemptions. In this proposed rule, as in the 2004 Final Rule, the Department estimated the number of EAP exempt workers because there is no data source that identifies workers as EAP exempt. Employers are not required to report EAP exempt workers to any central agency or as part of any employee or establishment survey. The methodology described here is largely based on the approach the Department used in the 2004 Final Rule. 69 FR 22196–209. All tables include estimates for 2013. Some tables also include estimates for 2005 59 Employees of firms with annual revenue less than $500,000 who are not engaged in interstate commerce are also not covered by the FLSA. However, these workers are not excluded from this analysis because the Department has no reliable way of estimating the size of this worker population, although the Department believes it composes a small percent of workers. These workers were also not excluded from the 2004 Final Rule. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38551 workforce was analyzed through successive stages to estimate the number of potentially affected EAP workers. workers in a specific region. To increase the sample size, the Department pooled together three years of CPS MORG data (2011 through 2013). Earnings for each 2011 and 2012 observation were inflated to 2013 dollars using the CPI–U, and the weight of each observation was adjusted so that the total number of potentially affected EAP workers in the pooled sample remained the same as the number represented by the 2013 CPS MORG. Thus, the pooled CPS MORG sample uses roughly three times as many observations to represent the same total number of workers in 2013. The additional observations allow the Department to better estimate certain attributes of the potentially affected labor force. Some assumptions had to be made to use these data as the basis for the analysis. For example, the Department eliminated workers who reported that their weekly hours vary and provided no additional information on hours worked. This was done because the Department cannot estimate impacts for these workers since it is unknown whether they work overtime and therefore unknown whether there would be any need to pay for overtime if their status changed from exempt to nonexempt. The Department reweighted the rest of the sample to account for this change (to keep the same total population estimates). This adjustment assumes that the distribution of hours worked by workers whose hours do not vary is representative of hours worked by workers whose hours do vary. The Department believes that without more information this is an appropriate assumption.60 To the extent these excluded workers are exempt, if they tend to work more overtime than other workers, then transfer payments, costs, and DWL may be underestimated. Conversely, if they work fewer overtime hours then transfer payments, costs, and DWL may be overestimated. The estimates of EAP exempt workers are based on data drawn from the CPS MORG, which is sponsored jointly by the U.S. Census Bureau and the BLS. The CPS is a large, nationally representative sample of the labor force. Households are surveyed for four months, excluded from the survey for eight months, surveyed for an additional four months, then permanently dropped from the sample. During the last month of each rotation in the sample (month 4 and month 16), employed respondents complete a supplementary questionnaire (the MORG) in addition to the regular survey. This supplement contains the detailed information on earnings necessary to estimate a worker’s exemption status. Although the CPS is a large scale survey, administered to 60,000 households representing the entire nation, it is still possible to have relatively few observations when looking at subsets of employees, such as exempt workers in a specific occupation employed in a specific industry, or VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00037 Fmt 4701 Sfmt 4702 60 This is justifiable because other employment characteristics are similar across these two populations. The share of all workers whose hours vary is 6.3 percent. E:\FR\FM\06JYP2.SGM 06JYP2 EP06JY15.001</GPH> the prevalence of the EAP exemption has changed from 2005 through 2013. Figure 2 illustrates how the U.S. civilian ii. Data tkelley on DSK3SPTVN1PROD with PROPOSALS2 (the first full calendar year after the most recent increase to the salary level was implemented) to demonstrate how 38552 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules iii. Number of Workers Covered by the Department’s Part 541 Regulations To estimate the number of workers covered by the FLSA and subject to the Department’s part 541 regulations, the Department first excluded workers who are not protected by the FLSA or are not subject to the Department’s regulations for a variety of reasons—for instance, they may not be covered by, or considered to be employees under, the FLSA. These workers include: • Military personnel, • unpaid volunteers, • self-employed individuals, • clergy and other religious workers, and • federal employees (with a few exceptions described below). Many of these workers are excluded from the CPS MORG: members of the military on active duty, unpaid volunteers, and the self-employed. For other categories that are not automatically excluded from the CPS data, such as unpaid workers, that is, workers with zero wages and earnings but who report being employed, the Department has implemented measures to screen them out. Religious workers were excluded from the analysis after being identified by their occupation codes: ‘clergy’ (Census occupational code 2040), ‘directors, religious activities and education’ (2050), and ‘religious workers, all other’ (2060). Most employees of the federal government are covered by the FLSA but are not subject to the Department’s part 541 regulations because their minimum wage and overtime pay are regulated by the Office of Personnel Management (OPM).61 See 29 U.S.C. 204(f). Exceptions exist for U.S. Postal Service employees, Tennessee Valley Authority employees, and Library of Congress employees. See 29 U.S.C. 203(e)(2)(A). These covered federal workers were identified and included in the analysis using occupation and/or industry codes.62 Employees of firms that have annual revenue of less than $500,000 and who are not engaged in interstate commerce are also not covered by the FLSA. The Department does not exclude them from the analysis because it has no reliable way of estimating the size of this worker population, although the Department believes it is a small percentage of workers. The 2004 Final Rule analysis similarly did not adjust for these workers. Table 5 presents the Department’s estimates of the total number of workers, and the number of workers covered by the FLSA and subject to the Department’s part 541 regulations in 2005 and 2013. The Department estimated that in 2013 there were 144.2 million wage and salary workers in the United States. Of these, 128.5 million were covered by the FLSA and subject to the Department’s regulations (89.1 percent). The remaining 15.7 million workers were excluded from coverage by the FLSA for the reasons described above and delineated in Table 6. TABLE 5—ESTIMATED NUMBER OF WORKERS COVERED BY THE FLSA AND SUBJECT TO THE DEPARTMENT’S PART 541 REGULATIONS, 2005 AND 2013 Civilian employment (1,000s) Year 2005 ............................................................................................................................................. 2013 ............................................................................................................................................. a Estimate they are unlikely to be affected by this proposed rule. These include: Reason • Blue collar workers, • workers paid hourly, and Total .......................................... 15,703 • workers who are exempt under Self-employed and unpaid certain other (non-EAP) exemptions. workers .................................. 12,130 In 2013 there were 46.6 million blue Religious workers ..................... 518 Federal employees a ................. 3,057 collar workers (Table 7). These workers were identified in the CPS MORG data Note: 2013 estimates use pooled data for using data from the U.S. Government 2011–2013. a Most employees of the federal government Accountability Office’s (GAO) 1999 are covered by the FLSA but are not covered white collar exemptions report 63 and by part 541. Exceptions are for U.S. Postal the Department’s 2004 regulatory Service employees, Tennessee Valley Authority employees, and Library of Congress impact analysis. Supervisors in traditionally blue collar industries are employees. classified as white collar workers iv. Number of Workers in the Analysis because their duties are generally After limiting the analysis to workers managerial or administrative, and covered by the FLSA and subject to the therefore they were not excluded as blue Number (1,000s) tkelley on DSK3SPTVN1PROD with PROPOSALS2 Number (1,000s) 122,716 a 128,511 Percent 86.3 89.1 uses pooled data for 2011–2013. TABLE 6—REASON NOT SUBJECT TO Department’s regulations, several other THE DEPARTMENT’S PART 541 REG- groups of workers are identified and excluded from further analysis since ULATIONS, 2013 61 Federal workers are identified in the CPS MORG with the class of worker variable PEIO1COW. 62 Postal Service employees were identified with Census industry code 6370. Tennessee Valley Authority employees were identified as federal VerDate Sep<11>2014 142,126 144,214 Subject to the Department’s regulations 19:25 Jul 02, 2015 Jkt 235001 workers employed in the electric power generation, transmission, and distribution industry (570) and in Kentucky, Tennessee, Mississippi, Alabama, Georgia, North Carolina, or Virginia. Library of Congress employees were identified as federal workers under Census industry ‘libraries and archives’ (6770) and residing in Washington, DC. PO 00000 Frm 00038 Fmt 4701 Sfmt 4702 collar workers. In 2013, 76.1 million workers were paid on an hourly basis.64 Also excluded from further analysis were workers who are exempt under certain other (non-EAP) exemptions. Although some of these workers may also be exempt under the EAP exemptions, even if these workers lost their EAP exempt status they would remain exempt from the minimum wage and/or overtime pay provisions and thus were excluded from the analysis. In 2013 an estimated 4.2 million workers, including some agricultural and transportation workers, were excluded from further analysis because they were subject to another (non-EAP) overtime exemption. See Appendix A: Methodology for Estimating Exemption Status, for details on how this population was identified. 63 Fair Labor Standards Act: White Collar Exemptions in the Modern Work Place, GAO/ HEHS–99–164, p. 40–41. 64 The CPS MORG variable PEERNHRY is used to determine hourly status. E:\FR\FM\06JYP2.SGM 06JYP2 38553 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 7—ESTIMATED NUMBER OF WORKERS COVERED BY THE FLSA AND SUBJECT TO THE DEPARTMENT’S REGULATIONS, 2005 AND 2013 (1,000S) Reason Excluded b Year Subject to DOL’s Part 541 Reg. 2005 ................................. 2013 ................................. Workers in the analysis a 122,716 128,511 Excluded from analysis 39,689 42,970 83,027 85,541 Another exemption c Blue collar workers 46,245 46,644 Hourly workers Agriculture 74,192 76,113 773 911 Transportation 1,944 1,827 Other 1,006 1,484 Note: 2013 estimates use pooled data for 2011–2013. a Wage and salary workers who are white collar, salaried, and not eligible for another (non-EAP) overtime exemption. b Numbers do not add to total due to overlap. c Eligible for another (non-EAP) overtime pay exemption. The Department excluded some of these workers from the population of potentially affected EAP workers in the 2004 Final Rule, but not all of them. Agricultural and transportation workers are two of the largest groups of workers excluded from this analysis, and they were similarly excluded in 2004. Agricultural workers were identified by occupational-industry combination.65 Transportation workers were defined as those who are subject to the following FLSA exemptions: section 13(b)(1), section 13(b)(2), section 13(b)(3), section 13(b)(6), or section 13(b)(10). This methodology is the same as in the 2004 Final Rule and is explained in Appendix A. The Department excluded 911,000 agricultural workers and 1.8 million transportation workers from the analysis. The remaining 1.5 million excluded workers are included in multiple FLSA minimum wage and overtime exemptions and are detailed in Appendix A. However, of these 1.5 million workers, all but 28,000 are either blue collar or hourly and thus the impact of excluding these workers is negligible. For 2013 there were a total of 85.5 million workers excluded from the analysis for the reasons denoted above. These eliminations left 43.0 million workers covered by the FLSA and potentially affected by this proposed rulemaking. tkelley on DSK3SPTVN1PROD with PROPOSALS2 v. Number of Potentially Affected EAP Workers After excluding workers not subject to the Department’s FLSA regulations and workers who are unlikely to be affected by this proposed rulemaking (i.e., blue collar workers, workers paid hourly, 65 In the 2004 Final Rule all workers in agricultural industries were excluded. 69 FR 22197. Here only workers also in select occupations were excluded since not all workers in agricultural industries qualify for the agricultural overtime pay exemptions. This method better approximates the true number of exempt agricultural workers and provides a more conservative—i.e., greater— estimate of the number of affected workers. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 workers who are subject to another (non-EAP) overtime exemption), the Department estimated the number of workers for whom employers might claim the EAP exemptions. There are two ways a worker can lose overtime protection pursuant to the EAP exemptions: the standard EAP test and the HCE test. To be exempt under the standard EAP test the employee must: • Be paid a predetermined and fixed salary that is not subject to reductions because of variations in the quality or quantity of work performed (the salary basis test), • earn at least a designated salary amount; the salary level has been set at $455 per week since 2004 (the salary level test), and • perform work activities that primarily involve executive, administrative, or professional duties as defined by the regulations (the duties test). The HCE test requires the employee to pass the same standard salary basis and salary level tests. However, the HCE duties test is much less restrictive than the standard duties test, and the employee must earn at least $100,000 in total annual compensation, including at least $455 per week paid on a salary or fee basis, while the balance may be paid as nondiscretionary bonuses and commissions. Hourly computer employees who earn at least $27.63 per hour and perform certain duties are exempt under section 13(a)(17) of the FLSA. These workers are considered part of the EAP exemptions but were excluded from the analysis because they are paid hourly and will not be affected by this proposed rulemaking (these workers were similarly excluded in the 2004 analysis). Salaried computer workers are exempt if they meet the salary and duties tests applicable to the EAP exemptions, and are included in the analysis since they will be impacted by this proposed rulemaking. PO 00000 Frm 00039 Fmt 4701 Sfmt 4702 Additionally, administrative and professional employees may be paid on a fee basis,66 as opposed to a salary basis, at a rate of at least the amount specified by the Department in the regulations. However, the CPS MORG does not identify workers paid on a fee basis (only hourly or non-hourly). Thus in the analysis, workers paid on a fee basis are considered with non-hourly workers and consequently classified as ‘‘salaried’’ (as was done in the 2004 Final Rule). Weekly earnings are also available in the data, which allowed the Department to identify which workers passed the salary level tests.67 The CPS MORG data do not capture information about job duties. Therefore, to determine whether a worker met the duties test, the Department used an analysis performed by officials from the WHD in 1998 in response to a request from the GAO. Because WHD enforces the FLSA’s overtime requirements and regularly assesses workers’ exempt status, WHD’s representatives were uniquely qualified to provide the analysis. The analysis was used in both the GAO’s 1999 white collar exemptions report 68 and the Department’s 2004 regulatory impact analysis. See 69 FR 22198. WHD’s representatives examined 499 occupational codes, excluding nine that 66 Payment on a ‘‘fee basis’’ occurs where an employee is paid an agreed sum for a single job regardless of the time required for its completion. § 541.605(a). Salary level test compliance for fee basis employees is assessed by determining whether the hourly rate for work performed (i.e., the fee payment divided by the number of hours worked) would total at least $455 per week if the employee worked 40 hours. § 541.605(b). 67 The CPS MORG variable PRERNWA, which measures weekly earnings, is used to identify weekly salary. The CPS variable includes nondiscretionary bonuses and commissions, which do not count toward the standard salary level test. This discrepancy between the earnings variable used and the FLSA definition of salary may cause a slight overestimate of the number of workers estimated to meet the standard test. 68 Fair Labor Standards Act: White Collar Exemptions in the Modern Work Place. (1999). GAO/HEHS–99–164, p. 40–41. E:\FR\FM\06JYP2.SGM 06JYP2 38554 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules were not relevant to the analysis for various reasons (one code was assigned to unemployed persons whose last job was in the Armed Forces, some codes were assigned to workers who are not FLSA covered, others had no observations). Of the remaining occupational codes, WHD’s representatives determined that 251 occupational codes likely included EAP exempt workers and assigned one of four probability codes reflecting the estimated likelihood, expressed as ranges, that a worker in a specific occupation would perform duties required to meet the EAP duties tests. The Department supplemented this analysis in the 2004 Final Rule regulatory impact analysis when the HCE exemption was introduced. The Department modified the four probability codes for highly paid workers based upon our analysis of the provisions of the highly compensated test relative to the standard duties test (Table 8). To illustrate, WHD representatives assigned exempt probability code 3 to the occupation ‘‘first-line supervisors/managers of construction trades and extraction workers’’ (Census code 6200), which indicates that a worker in this occupation has a 10 to 50 percent likelihood of meeting the standard EAP duties test. However, if that worker earns at least $100,000 annually, he or she has between a 58.4 percent and 60 percent probability of being exempt under the shorter HCE test. The occupations identified by the GAO in 1999 and used by the Department in the 2004 Final Rule map to an earlier occupational classification scheme (the 1990 Census Occupational Codes). Therefore, for this proposed rule an occupational crosswalk was used to map the previous occupational codes to the 2002 Census Occupational Codes which are used in the CPS MORG 2002 through 2010 data, and to the 2010 Census Occupational Codes which are used in the CPS MORG 2011 through 2013 data.69 If a new occupation is comprised of more than one previous occupation, then the new occupation’s probability code is the weighted average of the previous occupations’ probability codes, rounded to the closest probability code. TABLE 8—PROBABILITY WORKER IN CATEGORY PASSES THE DUTIES TEST The Standard EAP Test Probability code 0 1 2 3 4 Lower bound (percent) ....................................................................................................................... ....................................................................................................................... ....................................................................................................................... ....................................................................................................................... ....................................................................................................................... The HCE Test Upper bound (percent) 0 90 50 10 0 0 100 90 50 10 Lower bound (percent) 0 100 94 58.4 15 Upper bound (percent) 0 100 96 60 15 tkelley on DSK3SPTVN1PROD with PROPOSALS2 These codes provide information on the likelihood an employee met the duties test but they do not identify the workers in the CPS MORG who actually passed the test. Therefore, the Department designated workers as exempt or nonexempt based on the probabilities. For example, for every ten public relations managers, between five and nine were estimated to pass the standard duties test (based on probability category 2). However, it is unknown which of these ten workers are exempt; therefore, the Department must determine the status for these workers. Exemption status could be randomly assigned with equal probability, but this would ignore the earnings of the worker as a factor in determining the probability of exemption. The probability of qualifying for the exemption increases with earnings because higher paid workers are more likely to perform the required duties, an assumption adhered to by both the Department in the 2004 Final Rule and the GAO in its 1999 Report.70 The Department estimated the probability of exemption for each worker as a function of both earnings and the occupation’s exempt probability category using a gamma distribution.71 Based on these revised probabilities, each worker was assigned exempt or nonexempt status based on a random draw from a binomial distribution using the worker’s revised probability as the probability of success. Thus, if this method is applied to ten workers who each have a 60 percent probability of being exempt, six workers would be expected to be designated as exempt.72 However, which particular workers are designated as exempt may vary with each set of ten random draws. For details see Appendix A. The Department estimated that of the 43.0 million workers considered in the analysis, 28.5 million qualified for the EAP exemptions (Table 9). However, some of these workers were excluded from further analysis because they would not be affected by the proposed rule. This excluded group contains workers in named occupations who are not required to pass the salary requirements (although they must still pass the duties tests) and therefore whose exemption status is not dependent on their earnings. These occupations include physicians (identified with Census occupation codes 3010, 3040, 3060, 3120), lawyers (2100), teachers (occupations 2200–2550 and industries 7860 or 7870), academic administrative personnel (school counselors (occupation 2000 and industries 7860 or 7870) and educational administrators (occupation 0230 and industries 7860 or 7870)), and outside sales workers (a subset of occupation 4950). Out of the 28.5 69 Crosswalks and methodology available at: https://www.census.gov/people/io/methodology/. 70 For the EAP exemptions, the relationship between earnings and exemption status is not linear and is better represented with a gamma distribution. For the HCE exemption, the relationship between earnings and exemption can be well represented with a linear function because the relationship is linear at high salary levels (as determined by the Department in the 2004 Final Rule). Therefore, the gamma model and the linear model would produce similar results. See 69 FR 22204–08, 22215–16. 71 The gamma distribution was chosen because, during the 2004 revision, this non-linear distribution best fit the data compared to the other non-linear distributions considered (i.e., normal and lognormal). A gamma distribution is a general statistical distribution that is based on two parameters that control the scale (alpha) and shape (in this context, called the rate parameter, beta). 72 A binominal distribution is frequently used for a dichotomous variable where there are two possible outcomes; for example, whether one owns a home (outcome of 1) or does not own a home (outcome of 0). Taking a random draw from a binomial distribution results in either a zero or a one based on a probability of ‘‘success’’ (outcome of 1). This methodology assigns exempt status to the appropriate share of workers without biasing the results with manual assignment. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00040 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38555 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules million workers who are EAP exempt, 7.1 million, or 25.1 percent, were in named occupations in 2013. Thus these workers would be unaffected by changes in the standard salary level test. The 21.4 million EAP exempt workers remaining in the analysis are referred to in this proposed rulemaking as ‘‘potentially affected.’’ In addition to the 21.4 million potentially affected EAP exempt workers, the Department estimates that an additional 6.3 million salaried white collar workers who do not satisfy the duties test and who currently earn at least $455 per week but less than the proposed salary level will have their overtime protection strengthened because their exemption status will be clear based on the salary test alone without the need to examine their duties. TABLE 9—ESTIMATED PERCENTAGES OF EAP EXEMPT WORKERS IN NAMED OCCUPATIONS, 2005 AND 2013 Workers in the analysis (millions) a Year 2005 ................................................................................................................. 2013 ................................................................................................................. EAP exempt (millions) 39.7 43.0 25.0 28.5 EAP exempt in named occupations (millions) b % of EAP exempt in named occupations 6.4 7.1 25.7 25.1 Note: 2013 estimates use pooled data for 2011–2013. a Wage and salary workers who are white collar, salaried, and not eligible for another (non-EAP) overtime exemption. b Workers not subject to a salary level test includes, but is not limited to, teachers, academic administrative personnel, physicians, lawyers, and judges. There are three groups of workers who lose minimum wage and overtime protections under the EAP exemptions: (1) Those passing just the standard EAP tests (i.e., passing the standard duties test, the salary basis test, and the standard salary level test and not passing the HCE tests); (2) those passing just the HCE tests (i.e., passing the HCE duties test, salary basis test, and the total compensation test and not passing the standard duties tests); and (3) those passing both tests. Based on analysis of the occupational codes and CPS earnings data, the Department has concluded that in 2013, of the 21.4 million potentially affected EAP workers, approximately 15.7 million passed only the standard EAP test, 5.6 million passed both the standard and the HCE tests, and approximately 75,000 passed only the HCE test (Table 10). When impacts are discussed in section VII.D., workers who pass both tests will be considered with those who pass only the standard salary level test because this salary level test is more restrictive (i.e., the worker may continue to pass the standard salary level test even if he or she no longer passes the HCE compensation test). TABLE 10—ESTIMATED NUMBER OF WORKERS EXEMPT UNDER THE EAP EXEMPTIONS BY TEST TYPE, 2005 AND 2013 Potentially affected EAP workers (millions) Year Pass standard test only Total 2005 ................................................................................................................. 2013 ................................................................................................................. 18.6 21.4 15.8 15.7 Pass both tests 2.8 5.6 Pass HCE test only 0.03 0.08 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Note: 2013 estimates use pooled data for 2011–2013. vi. Characteristics of Potentially Affected EAP Workers After estimating the population of workers who are subject to the EAP exemptions and potentially affected by this proposed rulemaking, the Department tabulated the characteristics of these workers. The characteristics considered and presented here include: industry of employment, occupation, and Metropolitan Statistical Area (MSA) status. As previously noted, the Department estimated 2013 values using CPS MORG data pooled from 2011–2013 in order to increase the sample size. Table 11 presents the estimated number of potentially affected EAP workers broken down into 13 major industry groups.73 The industry with the most potentially affected EAP workers was professional and business services, with 4.2 million potentially affected EAP workers. Other industries where a large number of workers were potentially affected are education and health services (3.4 million), financial activities (3.3 million), and manufacturing (3.3 million). The industry with the smallest number of potentially affected EAP workers was agriculture, forestry, fishing, and hunting (33,000). Looking at exemption status by occupation, 10.8 million workers employed in the management, business, and financial occupations were potentially affected; this occupation category accounts for roughly half of all potentially affected EAP workers. Professional and related occupations also employed many of the potentially affected EAP workers (7.0 million, which is 32.9 percent of all potentially affected EAP workers). The Department considered MSA status because workers in cities and suburban areas tend to be paid more than workers in rural areas. The percentage of potentially affected EAP workers (92.0 percent) who live in MSAs is larger than the percentage of the total workforce (85.8 percent) who live in MSAs. 73 See Appendix B: Additional Tables, for potentially affected workers categorized into the more detailed 51 industry group classifications. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00041 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38556 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 11—POTENTIALLY AFFECTED EAP WORKERS BY INDUSTRY, OCCUPATION, AND MSA STATUS, NUMBER AND AS PERCENT OF TOTAL, 2013 Potentially affected EAP workers (millions) Industry, occupation, MSA status Total ......................................................................................................................................................................... As percent of potentially affected EAP workers 21.4 100.0 0.03 0.18 0.76 3.27 2.42 0.80 0.90 3.30 4.20 3.41 0.75 0.55 0.83 0.2 0.8 3.6 15.3 11.3 3.7 4.2 15.4 19.6 15.9 3.5 2.6 3.9 10.79 7.04 0.19 2.19 0.97 0.00 0.02 0.05 0.10 0.04 50.4 32.9 0.9 10.2 4.5 0.0 0.1 0.2 0.5 0.2 19.67 1.62 0.09 92.0% 7.6% 0.4% By Industry Agriculture, forestry, fishing, & hunting ................................................................................................................... Mining ...................................................................................................................................................................... Construction ............................................................................................................................................................. Manufacturing .......................................................................................................................................................... Wholesale & retail trade .......................................................................................................................................... Transportation & utilities .......................................................................................................................................... Information ............................................................................................................................................................... Financial activities .................................................................................................................................................... Professional & business services ............................................................................................................................ Education & health services .................................................................................................................................... Leisure & hospitality ................................................................................................................................................ Other services .......................................................................................................................................................... Public administration ................................................................................................................................................ By Occupation Management, business, & financial ......................................................................................................................... Professional & related ............................................................................................................................................. Services ................................................................................................................................................................... Sales & related ........................................................................................................................................................ Office & administrative support ............................................................................................................................... Farming, fishing, & forestry ..................................................................................................................................... Construction & extraction ........................................................................................................................................ Installation, maintenance, & repair .......................................................................................................................... Production ................................................................................................................................................................ Transportation & material moving ........................................................................................................................... By MSA Status MSA ......................................................................................................................................................................... Non-MSA ................................................................................................................................................................. Not Identified ............................................................................................................................................................ Note: Pooled data for 2011–2013. tkelley on DSK3SPTVN1PROD with PROPOSALS2 C. Determining the Revised Salary Level Test Values i. Background The Department proposes to set the EAP standard salary level at the 40th percentile of the weekly earnings distribution for all full-time salaried workers and to set the HCE compensation test equal to the 90th percentile (at an annual salary equivalent) of this distribution. These methods were used because they generate salary levels that (1) were deemed to be appropriate in distinguishing between workers who should and should not be exempt; (2) are easy to calculate and thus easy to replicate, creating transparency through simplicity; and (3) generate consistent salary levels.74 The Department believes 74 On a quarterly basis, BLS publishes a table of deciles of the weekly wages of full-time salaried workers, calculated using CPS data, which will provide employers with information on changes in salary levels prior to the annual updates. https:// VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 that setting the standard salary level at the 40th percentile earnings ($921 per week) allows for reliance on the current standard duties test without an unacceptably high risk of overtimeeligible employees being misclassified as EAP exempt and denied overtime protection. Additionally, the Department believes that setting the standard salary level at the 40th percentile earnings will not result in an unacceptably high risk that employees performing bona fide EAP duties will become entitled to overtime protection by virtue of the salary test. The methodologies used to revise the EAP salary levels have varied somewhat across the seven updates to the salary level test since it was implemented in 1938. To guide the determination of the proposed salary level, the Department considered methodologies used previously to revise the EAP salary www.bls.gov/cps/research_series_earnings_ nonhourly_workers.htm. PO 00000 Frm 00042 Fmt 4701 Sfmt 4702 levels. In particular, the Department focused on the 1958 revisions and the most recent revisions in 2004. The 1958 methodology is particularly instructive in that it synthesized previous approaches to setting the salary level, and the basic structures it adopted have been a touchstone in subsequent rulemakings (with the exception of 1975). The 1958 Revisions In 1958, the Department updated the salary levels based on a 1958 Report and Recommendations on Proposed Revision of Regulations, Part 541, by Harry S. Kantor (Kantor Report). To determine the revised salary levels the Department looked at data collected during WHD investigations on actual salaries paid to exempt EAP employees, grouped by geographic region, industry groups, number of employees, and size of city. The Department then set the salary level so that no more than about 10 percent of those in the lowest-wage E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules region, lowest-wage industry, smallest establishment group, or smallest city group would fail to meet the test. Kantor Report at 6.75 76 This methodology is referred to as the Kantor method and the Department followed a similar methodology in setting the salary levels in 1963 and 1970. The 2004 Revisions A significant change in 2004 from the Kantor method was that the salaries of both exempt and nonexempt full-time salaried workers in the South and retail industry were used to determine levels (hereafter referred to as the 2004 method), rather than the salaries of exempt workers only. However, because the salaries of exempt workers on average are higher than the salaries of all full-time salaried workers, the Department selected a higher earnings percentile for full-time salaried workers. Based on the Department’s 2004 analysis, the 20th percentile of earnings for exempt and nonexempt full-time salaried workers in the South and retail industry achieved a result very similar to the 10th percentile for workers in the lowest-wage regions and industries who were estimated to be exempt. 69 FR 22169. tkelley on DSK3SPTVN1PROD with PROPOSALS2 ii. Proposed Methodology for the Standard Salary Level The Department proposes to set the standard salary level at the 40th percentile of the distribution of weekly earnings for all full-time salaried workers nationwide. For the purposes of this proposed rulemaking, the Department relied on BLS calculations of the dollar value of the 40th earnings percentile from the CPS MORG data. BLS limited the population to salaried workers who work at least 35 hours per week and determined the specified percentile of the resulting weighted weekly earnings distribution.77 This methodology differs somewhat in specifics from previous revisions to the salary levels but the general concept 75 The Kantor method was based on an analysis of a survey of exempt workers as determined by investigations conducted by WHD. Subsequent analyses, including both the 2004 rulemaking and this proposed rule, have estimated exempt status using multiple data sources. 76 Because the salary level test is likely to have the largest impact on the low-wage categories of the economy (e.g., low-wage regions and industries), salaries in those regions/industries were selected as the basis for the required salary level under the Kantor method. 77 The Census Bureau publishes a public-use version of the CPS MORG data, which is very similar to the data used by BLS but involves a few changes to protect respondents’ confidentiality. The salary level found with the public-use files is only very slightly different from that obtained with the confidential data. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 holds: define a relevant population of workers, estimate an earnings distribution for that population, then set the salary level to a designated percentile of that distribution in order for the salary to serve as a meaningful line of demarcation between those Congress intended to protect and those who may qualify for exemption. The proposed method continues the evolution of the Department’s approach from the Kantor method to the 2004 method. The Department spent considerable time evaluating the previous methodologies. Where the proposed methodology differs from past methodologies, the Department believes the proposed methodology is an improvement. The Department compared the proposed method with the past methods, and the reasons for selecting the proposed method are detailed in the rest of this section. The Kantor and 2004 Methods The Department replicated the Kantor method and the 2004 method to evaluate and compare them to the proposed methodology.78 Although the Department was able to replicate the 1958 and 2004 methods reasonably well, we could not completely replicate those methods due to changes in data availability, occupation classification systems, and incomplete documentation. In general, there are four steps in the process: 1. Identify workers likely to be members of the population of interest. 2. Further narrow the population of interest by distinguishing that subpopulation employed in low-wage categories. 3. Estimate the distribution of earnings for these workers. 4. Identify the salary level that is equal to a pre-determined percentile of the distribution. The population of workers considered for purposes of setting the salary level depends on whether the 2004 method or the Kantor method is used. In replicating both methods, we limited the population to workers subject to the FLSA and covered by the Department’s part 541 provisions, and excluded EAP exempt workers in named occupations, and those exempt under another (nonEAP) exemption. For the 2004 method, the Department further limited the population to full-time salaried workers, 78 The Department followed the same methodology used in the 2004 Final Rule for estimating the Kantor method with minor adjustments. In an attempt to more accurately estimate the Kantor method, for example, this analysis included non-MSAs as a low-wage sector as Kantor did but the 2004 revisions did not. PO 00000 Frm 00043 Fmt 4701 Sfmt 4702 38557 and for the Kantor method we further limited the population of interest by only including those workers determined as likely to be EAP exempt (see more detailed methodology explanations in section VII.B. and Appendix A). During the 2004 revisions the Department identified two low-wage categories: The South (low-wage geographic region), and the retail industry (low-wage industry). For this proposed rule the Department identified low-wage categories by comparing average weekly earnings across categories for the populations of workers used in the Kantor method and the 2004 method. The South was determined to be the lowest-wage region and was used for the 2004 method; however, the Department chose to use a more detailed geographical break-down for the Kantor method to reflect the geographic categories Kantor used. Therefore, for the Kantor method the East South Central Division is considered the lowest-wage geographical area.79 The Department found that the industry with the lowest mean weekly earnings depends on whether the Kantor method or the 2004 method’s population was used. Therefore, three industries are considered low-wage: Leisure and hospitality, other services, and public administration. The Department also considered non-MSAs as a low-wage sector in the Kantor method. The 2004 revision did not consider population density but the Kantor method examined earnings across population size groups. In conclusion, the 2004 method looks at workers in the South and low-wage industries whereas the Kantor method looks at workers in the East South Central Division, non-MSAs, and the three low-wage industries. Next, the Department estimated the distributions of weekly earnings of two populations: (1) Workers who are in at least one of the low-wage categories and in the Kantor population, and (2) workers who are in at least one of the low-wage categories and in the 2004 population. From these distributions, alternate salary levels were identified based on pre-determined percentiles. For the Kantor method, the salary level for the long duties test is identified based on the 10th percentile of weekly earnings for the relevant population of likely EAP exempt workers, while the 2004 method salary level is identified based on the 20th percentile of weekly 79 The East South Central Division is a subset of the South and includes Alabama, Kentucky, Mississippi, and Tennessee. If the South is used instead, the resulting salary levels would increase slightly. E:\FR\FM\06JYP2.SGM 06JYP2 38558 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules earnings for the relevant population of both exempt and nonexempt salaried workers. Using 2013 CPS MORG data, the 2004 method resulted in a salary level of $577 per week and the Kantor method resulted in a salary level of $657 per week. Table 12 presents the distribution of weekly earnings used to estimate the salary levels under the proposed method, the 2004 method, and the Kantor method. TABLE 12—WEEKLY EARNINGS DISTRIBUTION, 2013 Distribution of annual earnings a Distribution of weekly earnings Percentile Full-Time Salaried 5 ....................................................................................... 10 ..................................................................................... 15 ..................................................................................... 20 ..................................................................................... 25 ..................................................................................... 30 ..................................................................................... 35 ..................................................................................... 40 ..................................................................................... 45 ..................................................................................... 50 ..................................................................................... 55 ..................................................................................... 60 ..................................................................................... 65 ..................................................................................... 70 ..................................................................................... 75 ..................................................................................... 80 ..................................................................................... 85 ..................................................................................... 90 ..................................................................................... 95 ..................................................................................... $378 490 586 645 726 773 852 921 981 1,065 1,154 1,248 1,363 1,478 1,626 1,828 2,000 2,349 3,077 2004 Method b $330 416 500 577 634 697 769 812 878 961 1,015 1,095 1,194 1,295 1,433 1,576 1,792 2,071 2,732 Kantor Method c $577 657 721 780 850 913 976 1,035 1,095 1,171 1,250 1,346 1,434 1,538 1,659 1,827 1,999 2,341 2,885 Full-Time Salaried $19,656 25,480 30,472 33,540 37,752 40,196 44,304 47,892 51,012 55,380 60,008 64,896 70,876 76,856 84,552 95,056 104,000 122,148 160,004 2004 Method b $17,148 21,632 26,000 30,000 32,968 36,247 39,988 42,209 45,659 49,972 52,762 56,960 62,093 67,317 74,533 81,952 93,208 107,707 142,050 Kantor Method c $30,000 34,176 37,500 40,586 44,200 47,486 50,732 53,817 56,960 60,879 65,000 69,992 74,566 80,000 86,245 95,000 103,958 121,721 150,000 Note: Estimates for the full-time salaried percentiles are from BLS. Estimates for the 2004 method and the Kantor method are based on pooled CPS MORG public-use data for 2011–2013. The use of pooled data allows us to better represent both earnings distributions and the characteristics of affected EAP workers. a Weekly earnings multiplied by 52. b Full-time salaried workers in the South or employed in a low-wage industry (excludes workers not subject to the FLSA, not subject to the salary level test, and in agriculture or transportation). c Salaried, white collar workers who earn at least $455 per week, pass the EAP duties test, and either live in the East South Central Division or a non-MSA or are employed in a low-wage industry (excludes workers not subject to FLSA, not subject to the salary level test, and in agriculture or transportation). tkelley on DSK3SPTVN1PROD with PROPOSALS2 iii. Rationale for the Methodology Chosen The salary level test has historically been intended to serve as an initial bright-line test for overtime eligibility for white collar employees. As discussed previously, however, there will always be white collar overtimeeligible employees who are paid above the salary threshold. A low salary level increases the number of these employees. The necessity of applying the duties test to these overtimeprotected employees consumes employer resources, may result in misclassification (which imposes additional costs to employers and society in the form of litigation), and is an indicator of the effectiveness of the salary level. Similarly, there will always be employees performing bona fide EAP duties who are paid below the salary threshold; the inability of employers to claim the EAP exemption for these employees is also an indicator of the effectiveness of the salary level. Selecting the standard salary level will inevitably affect the number of workers falling into each of these two categories. The Kantor method sought to minimize VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 the number of white collar employees who pass the duties test but were excluded from the exemption by the salary threshold and therefore set the salary level at the bottom 10 percent of exempt EAP employees in low wage regions and industries so as to prevent ‘‘disqualifying any substantial number of such employees.’’ Kantor Report at 5; see Weiss Report at 9. This method was based on the long/short test structure, in which employees paid at lower salary levels were protected by significantly more rigorous duties requirements than are part of the current standard duties test. This approach, however, does not take into sufficient account the inefficiencies of applying the duties test to large numbers of overtime-eligible white collar employees and the possibility of misclassification of those employees as exempt. In this rulemaking, the Department wants to correct for the elimination of the long duties test and set a salary level that appropriately classifies white collar workers as entitled to minimum wage and overtime protection or potentially exempt. Thus the Department’s proposed standard salary level is higher PO 00000 Frm 00044 Fmt 4701 Sfmt 4702 than the level the Kantor or 2004 methods would generate but still lower than the historical average for the short test. Setting the salary level at the 40th percentile of weekly earnings for fulltime salaried workers will reduce the number of employees subject to the standard duties test by raising the salary threshold; the Department believes that this will simplify the determination of exemption status for employers and will result in reduced misclassification of overtime-eligible white collar workers as exempt and reduced litigation. At the 40th percentile, 10.6 million white collar employees would no longer be subject to potential litigation over the duties they perform (4.6 million currently EAP exempt employees who would be newly entitled to overtime due to the increase in the salary threshold and 6.0 million overtimeeligible white collar employees who are paid between $455 and $921 per week whose exemption status would no longer depend on the application of the duties test). The proposed salary level will therefore more efficiently distinguish between employees who may meet the duties requirement of the E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules EAP exemption and those who do not, without necessitating a return to the more detailed long duties test. The proposed salary level also affects the likelihood of workers being misclassified as exempt from overtime pay. This provides an additional measure of the effectiveness of the salary level as a bright-line test delineating exempt and nonexempt workers. The Department estimated the number of workers misclassified as exempt as the number of salaried white collar workers who: Earn at least $455 per week; do not satisfy the EAP duties tests; are not in a named occupation (or exempt under another (non-EAP) exemption); usually work overtime; and do not usually receive overtime pay.80 The Department estimates that almost 20 percent of the 11.6 million salaried white collar workers who fail the duties test are misclassified as exempt. The Department estimates that at the proposed salary level, the number of overtime-eligible white collar workers earning at or above the salary level will decrease by 6.0 million, and that approximately 806,562 (13.5 percent) of these workers are currently misclassified as exempt. In this section the Department assesses the impact of the standard salary level as a bright-line test for the EAP exemptions by examining: (1) The number of white collar workers who pass the salary level test but not the duties test and (2) the number of white on workers’ response to the CPS–MORG question concerning whether they receive overtime pay, tips, or commissions at their job (‘‘PEERNUOT’’ variable). tkelley on DSK3SPTVN1PROD with PROPOSALS2 80 Based VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 collar workers who pass the duties test but not the salary level test. The Department makes this assessment at the current salary level and the proposed salary level, while holding all other factors determining exempt status constant (e.g., not considering whether the duties test is correctly applied or potential employer response to the change in the salary level test). Examining the impact of the salary threshold in isolation from the application of the duties test or employer adjustments to pay or hours does not provide a complete picture of the impact of a new salary threshold. It does, however, allow the Department to evaluate the effectiveness of the salary level in protecting overtime-eligible white collar employees without unduly excluding from the exemption employees performing EAP duties. In order to calculate the potential impact on the two groups of workers, the Department estimated: (1) The number of salaried white collar workers who are eligible for overtime pay because they do not pass the standard EAP duties test, but earn above a specific salary level; and (2) the number of salaried white collar workers who satisfy the standard duties test but earn less than a specific standard salary level.81 These numbers were estimated at the current salary level ($455) and the proposed standard salary level of the 81 These populations are limited to salaried, white collar workers subject to the FLSA and the Department’s part 541 regulations, and not eligible for another (non-EAP) exemption, not in a named occupation, and not HCE only. PO 00000 Frm 00045 Fmt 4701 Sfmt 4702 38559 40th percentile of weekly wages of all full-time salaried workers ($921). As a benchmark, the Department estimates that at the current standard salary threshold, there are 11.6 million salaried white collar workers who fail the standard duties test and are therefore overtime eligible, but earn at least the $455 threshold, while there are only 845,500 salaried white collar workers who pass the standard duties test but earn less than the $455 level. Thus the number of white collar workers who pass the current salary threshold test but not the duties test is nearly 14 times the number of white collar workers who pass the duties test but are paid below the salary threshold. This underscores the large number of overtime-eligible workers for whom employers must perform a duties analysis, and who may be at risk of misclassification as EAP exempt. If the salary threshold were raised to the 40th percentile, the number of overtimeeligible salaried white collar workers who would earn at least the threshold but do not pass the duties test would be reduced to 5.6 million. At the 40th percentile, the number of salaried white collar workers who would pass the standard duties test but earn less than the 40th percentile would be 4.6 million (approximately 25 percent of all white collar salaried employees who pass the standard duties test). While this number is higher than the number of such employees under the Kantor method, it includes employees who would not have passed the more rigorous long duties test and therefore were not included under that approach. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules As illustrated in Figure 3, as the salary level increases there is a decrease in the share of overtime-eligible white collar workers for whom employers would be required to make an assessment under the duties test and who would be subject to possible misclassification. At the same time, as the salary level increases there is an increase in the share of white collar workers who pass the duties test but are screened from exemption by the salary threshold. At the current salary level, there is a very large gap between white collar workers who are overtime eligible but earn at least the threshold (about 85 percent of all salaried white collar workers who fail the duties test are paid at least $455 per week) and white collar workers who pass the standard duties test but do not meet the current salary level (about 4 percent of all salaried white collar workers who pass the duties test are paid less than $455 per week). At the proposed salary level of the 40th percentile of weekly earnings of full-time salaried workers, the percentage of overtime-eligible white collar workers who earn above the threshold (and thus would be at risk of misclassification) remains substantially higher than the percentage of white collar workers who pass the duties test but earn less than the salary threshold (and would become overtime VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 protected).82 The salary threshold would have to be considerably higher (at a salary level of approximately $1,015, approximately the 50th percentile level of full-time salaried workers) before the percentage of white collar workers who earn less than the threshold but pass the duties test would equal the percentage who are overtime eligible but earn at least the salary threshold. The Department has also looked at the impact of the proposed salary level on these two groups of workers in lowwage (East South Central) and highwage (Pacific) regions in addition to nationally.83 For the East South Central region, the salary level at which the percentages of the two groups are about equal is approximately $914 per week, while in the Pacific region, the salary at which the percentages of the two groups are equal is approximately $1,154 per week. The Department’s proposed salary 82 Approximately 41 percent of white collar salaried workers who do not pass the duties test earn at least the proposed salary level ($921 per week). Conversely, approximately 25 percent of employees who pass the standard duties test (and 22 percent of employees who are currently exempt) earn less than the proposed salary level. 83 Of the nine Census Region Divisions, the East South Central and Pacific divisions correspond to the divisions with the lowest and highest earnings using the Kantor method. The East South Central includes Alabama, Kentucky, Mississippi, and Tennessee. The Pacific includes Alaska, California, Hawaii, Oregon, and Washington. PO 00000 Frm 00046 Fmt 4701 Sfmt 4702 level of the 40th percentile of weekly earnings of full-time salaried workers ($921 per week) falls appropriately within this range. This supports the Department’s use of nationwide data to set a salary level that is appropriate for classifying workers as entitled to minimum wage and overtime pay or potentially exempt, and takes into account the impact on employers in low-wage regions. Appropriateness. The standard salary level serves as a bright-line test for employers, intended to assist in identifying those workers with duties that may make them truly bona fide executive, administrative, or professional employees. As explained in the preceding analysis, the Department has determined that setting the proposed standard salary level at the 40th percentile of earnings for full-time salaried workers ($921) appropriately balances the tradeoff between denying the exemption for employees who are currently exempt and exposing workers who fail to meet the duties test to the risk of misclassification as exempt. In the absence of a long duties test which limits the amount of nonexempt work that can be performed, the Department believes a salary level at or above the proposed salary level appropriately distinguishes between overtimeprotected and potentially exempt employees. Of employees currently E:\FR\FM\06JYP2.SGM 06JYP2 EP06JY15.002</GPH> tkelley on DSK3SPTVN1PROD with PROPOSALS2 38560 38561 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules exempt under the part 541 regulations, that is, those who are paid on a salary basis of at least $455 and meet the duties test, approximately 78 percent earn at least the proposed level of $921 per week. Conversely, among overtimeeligible white collar employees (both salaried and hourly), approximately 75 percent earn less than the proposed salary level. Simplicity. The proposed method of basing the standard salary threshold on a particular percentile of weekly earnings for full-time salaried employees involves less estimation than previous updates, making it easier to implement, less prone to error, and more transparent than before. The proposed method reduces computation by simplifying the classification of workers to just two criteria: Wage or salaried, and full-time or part-time. Application of the Kantor method, in particular, would involve significant work to replicate since one would need to identify likely EAP exempt workers, a process which requires applying the standard duties test to determine the population of workers used in the earnings distribution. The proposed method is easier for stakeholders to replicate and understand because the standard duties test does not need to be applied to determine the population of workers used in the earnings distribution. Consistency. A method that produces very different salary levels in consecutive years will reduce confidence that the salary levels in any given year are optimal. Since 2003, the 40th percentile of full-time salaried workers’ weekly earnings has increased by an average of 2.6 percent annually. Similarly, the salary levels that would have been generated by the 2004 method increased 2.4 percent annually on average between 2003 and 2013. Conversely, since 2003 the salary levels that would have been generated by the Kantor method increased 3.6 percent on average annually. The larger growth rate for the Kantor method explains why despite the Kantor method and 2004 method generating very similar salary levels for the 2004 rulemaking, by 2013 these levels differ significantly (Kantor = $657; 2004 = $577). The primary reason the Kantor method generates a larger salary level than the 2004 method in 2013 is because the Kantor method uses the value of the current salary level test to identify the population of workers from which the earnings distribution is determined. Therefore, the Kantor method limits the pool of workers in the sample to those who meet the required salary level before evaluating the salaries of workers in low-wage regions and industries, while the 2004 method looks to all salaried workers in the South and retail industry but does not exclude workers with salaries below the current salary level. For example, in 2003 the Kantor method population of interest was limited to workers earning at least $155 per week (the 1975 long test salary level); in this proposed rule the Kantor method’s population was restricted to workers earning at least $455 per week. Therefore the population considered in Kantor’s method changes each time the salary level is changed. The Department’s proposed method, like the 2004 method, considers all full-time salaried workers and does not limit the pool to only those workers who meet the current salary level test, thus avoiding this potential shortcoming of the Kantor method. Based on the comparison of the characteristics of the methods reviewed in this section, the Department has determined that the proposed method, for the reasons identified, meets the objectives of appropriateness, simplicity, and consistency. iv. Standard Salary Levels With Alternative Methodologies When assessing the effects of the proposed standard salary level on the U.S. economy, the Department also evaluated several alternatives. This section presents the alternative salary levels considered and the bases for identifying those alternative levels. As shown in Table 13, the alternative salary levels evaluated are: • Alternative 1: Calculate the salary level by adjusting the 2004 salary level of $455 for inflation from 2004 to 2013 as measured by the CPI–U. This results in a salary level of $561 per week. • Alternative 2: Use the 2004 method to set the salary level at $577 per week. • Alternative 3: Use the Kantor method to set the salary level at $657 per week. • Alternative 4: Use the 50th earnings percentile of full-time hourly and salaried workers. This results in a salary level of $776 per week. • Alternative 5: Adjust the salary level from the Kantor method to reflect the historical ratio between the long and short test salary levels. This results in a salary level of $979 per week. • Alternative 6: Use the 50th earnings percentile of full-time salaried workers. This results in a salary level of $1,065 per week. • Alternative 7: Adjust the 1975 short test salary level of $250 for inflation from 1975 to 2013. This results in a salary level of $1,083 per week. TABLE 13—PROPOSED STANDARD SALARY LEVEL AND ALTERNATIVES, 2013 Salary level (weekly/ annually) Alternative tkelley on DSK3SPTVN1PROD with PROPOSALS2 Alternative #1: Inflate 2004 levels ....................................................................................................... Alternative #2: 2004 method ............................................................................................................... Alternative #3: Kantor method ............................................................................................................. Alternative #4: Median full-time hourly and salaried workers ............................................................. Proposed (40th percentile full-time salaried) ....................................................................................... Alternative #5: Kantor short test .......................................................................................................... Alternative #6: Median full-time salaried ............................................................................................. Alternative #7: Inflate 1975 short test level ......................................................................................... $561/$29,178 577/30,000 657/34,176 776/40,352 921/47,892 979/50,922 1,065/55,380 1,083/56,291 Total increase a $ % 106 122 202 321 466 524 610 628 23.3 26.8 44.4 70.5 102.4 115.2 134.1 137.9 Note: Pooled data for 2011–2013. a Average weekly change between proposed/alternative salary level and the salary level set in 2004 ($455 per week). Alternatives 2 (2004 method) and 3 (Kantor method) were already discussed. Alternative 5 (Kantor short test) is also based on the Kantor method VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 but, whereas alternative 3 generates the salary level associated with the long duties test, alternative 5 generates a level more closely resembling the salary PO 00000 Frm 00047 Fmt 4701 Sfmt 4702 associated with the short duties test. In the 2004 Final Rule, the Department replaced the structure of a short and a long duties test with a single standard E:\FR\FM\06JYP2.SGM 06JYP2 38562 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 duties test based on the less restrictive short duties test, which had historically been paired with a higher salary level test. However, the Department set the standard salary level in 2004 at a level that was equivalent to the Kantor long test salary level, which was associated with the long duties test and limited the amount of nonexempt work that the employee could perform. In alternative 5, the Department therefore considered revising the standard salary level to approximate the short test salary that better matches the standard duties test. On average, the salary levels set in 1949 through 1975 were 149 percent higher for the short test than the long test. Therefore, the Department inflated the 2013 Kantor estimate of $657 by 149 percent, which generated a short salary level equivalent of $979.84 While the Department used the average difference between the Kantor long and short tests for this alternative, the ratio of the short to long salary tests ranged from approximately 130 percent to 180 percent between 1949 and 2004. The low end of this range would result in a salary of $854; the high end would result in a salary of $1,183. Alternatives 1 (inflating the 2004 level) and 7 (inflating the 1975 short test level) use similar approaches to each other. Both begin with an exemption salary level set in an earlier rulemaking, and use the CPI–U to adjust that salary level to account for inflation between the year it was set and 2013. Where the two approaches differ is in the selection of the starting point. Alternative 1 assumes the 2004 standard salary level was set at an appropriate level, and that changes in earnings since that time can be reflected well by changes in prices. Alternative 1 is inappropriate because the salary level set in 2004 does not fully account for changes in the sample and the change from long and short duties tests to a single standard test that is comparable to the old short duties test. Alternative 7 assumes that the 1975 salary levels were set to a more appropriate level than the 2004 levels; inflating the 1975 short duties test salary level to 2013 results in a salary level of $1,083 per week. This alternative is inappropriate because it is based on interim salary rates. 40 FR 7091. Additionally, the Department 84 The Department estimated the average historic ratio of 149 percent as the simple average of the fifteen historical ratios of the short duties salary level to the long duties salary level (salary levels were set in 5 years and in each year the salary level varied between the three exemptions: executive, administrative, and professional). If the Department had weighted the average ratio based on the length of time the historic salary levels were in effect, this would have yielded an average historic ratio of 152 percent and a salary level of $999. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 thinks the salary level generated with this method is too high in light of the fact that there no longer is a long duties test with an associated lower salary level that employers may use to claim that employees are exempt. Alternatives 4 and 6 set the standard salary equal to the 50th percentile, or median, of weekly earnings for two groups of workers: full-time hourly and salaried workers and full-time salaried workers, respectively. These approaches are similar to the proposed method in that they set the salary level equal to a percentile of an earnings distribution. The 50th earnings percentile of all fulltime hourly and salaried workers results in a salary level of $776. The Department concluded, however, that it would not be appropriate to include the wages of hourly workers in setting the EAP salary threshold and that the resulting salary level was too low to work effectively with the standard duties test. Selecting the 50th earnings percentile of full-time salaried workers results in a standard salary level of $1,065, which is only $18 per week less than alternative 7. Like alternative 7, the Department believes that the salary level generated with this method is too high because there is no longer a long duties test with an associated lower salary level that employers may use to claim that employees are exempt. Section VII.D. will detail the transfers, costs, and benefits of the proposed salary levels and alternatives. A comparison of the costs and benefits justifies the Department’s decision to propose a standard salary level of the 40th percentile of weekly earnings for all full-time salaried workers ($921 per week). v. Proposed Methodology for the HCE Total Annual Compensation Level The Department proposes to set the HCE compensation level equal to the annual equivalent of the 90th percentile of the distribution of earnings for all full-time salaried workers. BLS calculated the salary level from the CPS MORG data by limiting the population to non-hourly workers who work fulltime (i.e., at least 35 hours per week) and determining the 90th percentile of the resulting weighted weekly earnings distribution. The 90th percentile of weekly earnings ($2,349) was then multiplied by 52 to determine the annual earnings equivalent ($122,148). This mirrors the method used to set the standard salary level but uses a percentile towards the top of the earnings distribution to reflect the minimal duties criteria associated with the highly compensated exemption. PO 00000 Frm 00048 Fmt 4701 Sfmt 4702 The Department also evaluated the following alternative HCE compensation levels: • HCE alternative 1: Leave the HCE compensation level unchanged at $100,000 per year. • HCE alternative 2: Set the HCE compensation level at $150,000 per year, which is approximately the annualized level of weekly earnings exceeded by 6.3 percent of full-time salaried workers. This is the same percent of such workers that exceeded the HCE compensation level in 2004. The Department concluded that HCE alternative 1 was inappropriate because leaving the HCE compensation level unchanged at $100,000 per year would ignore more than 10 years of wage growth. In 2013, approximately 17 percent of full-time salaried workers earned at least $100,000 annually, more than twice the share who earned that amount in 2004. Conversely, HCE alternative 2 would set the annual compensation level at $150,000.85 The Department believes this salary level would be too high to provide a meaningful alternative test for exemption. Thus, the Department believes its proposal to adjust the HCE total annual compensation to reflect the 90th percentile of earnings of full-time salaried workers strikes the appropriate balance. D. Impacts of Revised Salary and Compensation Level Test Values i. Overview Impacts due to the proposed increases in the EAP salary levels will depend on how employers respond. Employer response is expected to vary by the characteristics of the affected EAP workers. For workers who usually work 40 hours a week or less, the Department assumes that employers will reclassify these workers as overtime-eligible and will pay the same weekly earnings for the same number of hours worked. While these employees will become overtime eligible, employers can continue to pay their current salaries and need make no adjustments as long as the employees’ hours do not exceed 40 hours in a workweek.86 For employees who work overtime, employers may: (1) Pay the required overtime premium for the current number of overtime hours based upon the current implicit regular rate of pay; 85 This compensation level corresponds to the annual value of the highest weekly earnings reported in the CPS MORG public-use data. 86 Assuming the worker earns the minimum wage. Otherwise, wages and hours will be adjusted to reflect compliance with minimum wage requirements. E:\FR\FM\06JYP2.SGM 06JYP2 38563 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules (2) reduce the regular rate of pay so total weekly earnings and hours do not change after overtime is paid; (3) eliminate overtime hours; (4) increase employees’ salaries to the proposed salary level; or (5) use some combination of these responses. Transfers from employers to employees, direct employer costs, and DWL depend on how employers respond to the proposed rulemaking. The cost, benefit and transfer estimates appearing throughout this section represent nationwide aggregates. Given the potential for this proposed rule to have impacts that differ by region or industry, the Department invites detailed comment, data and analysis that would allow for estimation of impacts on a regional or industry basis. ii. Summary of Quantified Impacts Table 14 presents the aggregated projected costs, transfers, and DWL associated with increasing the standard EAP salary level from $455 per week to the 40th earnings percentile, $921 per week, and the HCE compensation level from $100,000 to the 90th earnings percentile, $122,148 annually (without automatic updating). The Department estimated that the direct employer costs of this proposal will total $592.7 million in the first year, with average annualized direct costs of $194.2 million per year over 10 years. In addition to the direct costs, this proposed rulemaking would also transfer income from employers to employees. Year 1 transfers would equal $1,482.5 million, with average annualized transfers estimated at $872.9 million per year over 10 years. Finally, the 10-year average annualized DWL was estimated to be $7.2 million. In order to increase the sample size and the reliability and granularity of results in this analysis, the Department used three years (2011–2013) of CPS MORG data to represent the 2013 labor market. Monetary values in 2011 and 2012 were inflated to 2013 dollars and the sample was reweighted to reflect the population of potentially affected workers in 2013. The potential employer costs due to reduced profits and additional hiring were not quantified but are discussed in section VII.D.iv.5. TABLE 14—SUMMARY OF REGULATORY COSTS AND TRANSFERS, STANDARD AND HCE SALARY LEVELS, WITHOUT AUTOMATIC UPDATING (millions 2013$) Future years b Cost/Transfer a Average annualized value Year 1 Year 2 Year 10 3% Real rate 7% Real rate Direct Employer Costs: Regulatory familiarization ................................................................. Adjustment ........................................................................................ Managerial ........................................................................................ $254.5 160.1 178.1 $0.0 1.1 169.0 $0.0 0.1 93.1 $29.0 18.4 135.9 $33.9 21.5 138.9 Total direct costs c ..................................................................... 592.7 170.0 93.1 183.2 194.2 Transfers from Employers to Due to minimum wage ...................................................................... Due to overtime pay ......................................................................... 46.7 1,435.8 44.0 1,017.1 9.9 490.2 27.9 815.7 29.3 843.6 Total transfers d ......................................................................... 1,482.5 1,061.2 500.1 843.6 872.9 DWL e DWL .................................................................................................. .................... 7.4 .................... 9.8 .................... 4.3 .................... 7.0 .................... 7.2 Workers d a Additional costs and benefits of the rule that could not be quantified or monetized are discussed in the text. costs/transfers represent a range over the nine-year span. c Components may not add to total due to rounding. d This is the net transfer from employers to workers. There may also be transfers of hours and income from some workers to others. e DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. Since the transfer associated with the minimum wage is negligible compared to the transfer associated with overtime pay, the vast majority of this cost is attributed to the overtime pay provision. b These iii. Affected EAP Workers tkelley on DSK3SPTVN1PROD with PROPOSALS2 1. Overview Costs, transfer payments, DWL, and benefits of this proposed rulemaking depend on the number of affected EAP workers and labor market adjustments made by employers. The Department estimated there were 21.4 million potentially affected EAP workers, that is EAP workers who either (1) passed the salary basis test, the standard salary level test, and the standard duties test, or (2) passed the salary basis test, passed the standard salary level test, the HCE total compensation level test, and the HCE duties test. This number excludes VerDate Sep<11>2014 20:30 Jul 02, 2015 Jkt 235001 workers in named occupations who are not subject to the salary tests or who qualify for another (non-EAP) exemption. The Department estimated that increasing the standard salary level from $455 per week to the 40th earnings percentile of all full-time salaried workers ($921 per week) would directly affect 4.6 million workers (i.e., the number of potentially affected workers who earn at least $455 per week but less than $921 per week). These affected workers compose 21.7 percent of potentially affected EAP workers. The Department also estimated that 36,000 workers would be directly affected by PO 00000 Frm 00049 Fmt 4701 Sfmt 4702 an increase in the HCE compensation level from $100,000 to the 90th earnings percentile (the number of potentially affected workers who earn between $100,000 and $122,148 annually and pass the minimal duties test but not the standard duties test; about 0.2 percent of the pool of potentially affected EAP workers). Table 15 presents the number of affected EAP workers, the mean number of overtime hours they work per week, and their average weekly earnings. The 4.6 million workers affected by the increase in the standard salary level average 1.6 hours of overtime per week and earn an average of $731 per week. E:\FR\FM\06JYP2.SGM 06JYP2 38564 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules The average number of overtime hours is low because most of these workers (3.7 million) do not usually work overtime.87 However, the estimated 988,000 affected workers who regularly work overtime average 11.1 hours of overtime per week. The 36,000 EAP workers affected by the proposed change in the HCE annual compensation level average 5.8 hours of overtime per week and earn an average of $2,103 per week. Although most affected EAP workers who typically do not work overtime might experience little or no change in their daily work routine, those who regularly work overtime may experience significant changes. The Department expects that workers who routinely work some overtime or who earn less than the minimum wage are most likely to be tangibly impacted by the revised salary level.88 Employers might respond by: converting such employees to overtime eligible, paying at least the minimum wage, and paying the overtime premium; reducing overtime hours; reducing workers’ regular wage rates (where the rate exceeds the minimum wage); increasing the employees’ salary to the proposed salary level; or use some combination of these responses. TABLE 15—NUMBER OF AFFECTED EAP WORKERS, MEAN OVERTIME HOURS, AND MEAN WEEKLY EARNINGS, 2013 Affected EAP workers a Type of affected EAP worker Number (1,000s) % of total Mean overtime hours Mean usual weekly earnings Standard Salary Level All affected EAP workers ................................................................................. Earn less than the minimum wage b ................................................................ Regularly work overtime .................................................................................. Occasionally work overtime c ........................................................................... 4,646 12 988 180 100 0.3 21.3 3.9 1.6 36.4 11.1 8.0 $731 529 743 729 36.2 100 5.8 2,103 14.5 1.0 40.1 2.6 14.3 6.5 2,119 2,120 HCE Compensation Level All affected EAP workers ................................................................................. Earn less than the minimum wage b ................................................................ Regularly work overtime .................................................................................. Occasionally work overtime c ........................................................................... Note: Pooled data for 2011–2013. a Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels). b The applicable minimum wage is the higher of the federal minimum wage and the state minimum wage. HCE workers will not be impacted by the minimum wage provision. c Workers who do not usually work overtime but did in the survey week. Mean overtime hours are actual overtime hours in the survey week. tkelley on DSK3SPTVN1PROD with PROPOSALS2 The Department considered two types of overtime workers in this analysis: regular overtime workers and occasional overtime workers.89 Regular overtime workers typically worked more than 40 hours per week. Occasional overtime workers typically worked 40 hours or less per week, but they worked more than 40 hours in the week they were surveyed. The Department considers these two populations separately in the analysis because labor market responses to overtime pay requirements may differ for these two types of workers. An estimated 181,000 occasional overtime workers will be affected by either the standard salary level or the HCE total annual compensation level increase in any given week (3.9 percent of all affected EAP workers). They averaged 8.0 hours of overtime per week. This group represents the number 87 That is, workers who report they usually work 40 hours or less per week (identified with variable PEHRUSL1 in CPS MORG). 88 A small proportion (0.3 percent) of affected EAP workers earns implicit hourly wages that are less than the applicable minimum wage (the higher of the state or federal minimum wage). The implicit VerDate Sep<11>2014 20:30 Jul 02, 2015 Jkt 235001 of workers with occasional overtime hours in the week the CPS MORG survey was conducted. In other weeks, these specific individuals may not work overtime but other workers, who did not work overtime in the survey week, may work overtime. Because the survey week is a representative week, the Department believes the prevalence of occasional overtime in the survey week, and the characteristics of these workers, is representative of other weeks (even though a different group of workers would be identified as occasional overtime workers in a different week).90 2. Characteristics of Affected EAP Workers In this section the Department examines the characteristics of affected EAP workers. Table 16 presents the distribution of affected workers across hourly wage is calculated as an affected EAP employee’s total weekly earnings divided by total weekly hours worked. 89 Regular overtime workers were identified in the CPS MORG with variable PEHRUSL1. Occasional overtime workers were identified in the CPS MORG with variables PEHRUSL1 and PEHRACT1. PO 00000 Frm 00050 Fmt 4701 Sfmt 4702 industries, occupations, and MSA status. The industry with the largest number of affected EAP workers was education and health services (1.0 million). The management, business, and financial occupation category accounted for the most affected EAP workers by occupation (2.1 million). A substantial majority of affected EAP workers resided in MSAs (4.1 million). Employers in non-MSAs and low-wage industries may perceive a greater impact due to the lower wages and salaries typically paid in those areas and industries. However, because the vast majority of potentially affected workers reside in MSAs and do not work in lowwage industries, the Department believes that the proposed salary level is appropriate. 90 The Department can estimate the average number of occasional overtime workers in any given week but cannot estimate the total number of individuals working occasional overtime in the year since the Department does not know how many weeks in a year a specific worker works overtime. E:\FR\FM\06JYP2.SGM 06JYP2 38565 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 16—ESTIMATED NUMBER OF EXEMPT WORKERS WITH THE CURRENT AND PROPOSED SALARY LEVELS, BY INDUSTRY, OCCUPATION, AND MSA STATUS, 2013 Potentially affected EAP workers (millions) a With updated standard and HCE levels Industry, occupation, and MSA status At current salary levels Number b Total ............................................................................................................................................. Reduction (affected workers) c 21.38 16.70 4.68 0.03 0.18 0.76 3.27 2.42 0.80 0.90 3.30 4.20 3.41 0.75 0.55 0.83 0.03 0.16 0.61 2.86 1.76 0.64 0.71 2.61 3.46 2.41 0.49 0.36 0.59 0.01 0.02 0.16 0.41 0.66 0.16 0.18 0.68 0.73 0.99 0.26 0.18 0.24 10.79 7.04 0.19 2.19 0.97 0.00 0.02 0.05 0.10 0.04 8.69 5.63 0.11 1.57 0.53 0.00 0.02 0.04 0.08 0.03 2.10 1.40 0.08 0.62 0.44 0.00 0.01 0.01 0.02 0.01 19.67 1.62 0.09 15.53 1.11 0.06 4.14 0.52 0.02 By Industry Agriculture, forestry, fishing, & hunting ....................................................................................... Mining .......................................................................................................................................... Construction ................................................................................................................................. Manufacturing .............................................................................................................................. Wholesale & retail trade .............................................................................................................. Transportation & utilities .............................................................................................................. Information ................................................................................................................................... Financial activities ........................................................................................................................ Professional & business services ................................................................................................ Education & health services ........................................................................................................ Leisure & hospitality .................................................................................................................... Other services .............................................................................................................................. Public administration .................................................................................................................... By Occupation Management, business, & financial ............................................................................................. Professional & related ................................................................................................................. Services ....................................................................................................................................... Sales & related ............................................................................................................................ Office & administrative support ................................................................................................... Farming, fishing, & forestry ......................................................................................................... Construction & extraction ............................................................................................................ Installation, maintenance, & repair .............................................................................................. Production .................................................................................................................................... Transportation & material moving ............................................................................................... By MSA Status MSA ............................................................................................................................................. Non-MSA ..................................................................................................................................... Not identified ................................................................................................................................ Note: Pooled data for 2011–2013. a Workers who are white collar, salaried, not eligible for another (non-EAP) overtime exemption, and not in a named occupation. b Workers who continue to be exempt after the proposed increases in the salary levels (assuming affected workers’ weekly earnings do not increase to the proposed salary level). c Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels). iv. Costs tkelley on DSK3SPTVN1PROD with PROPOSALS2 1. Summary Three direct costs to employers were quantified in this analysis: (1) Regulatory familiarization costs; (2) adjustment costs; and (3) managerial costs. Regulatory familiarization costs are costs to learn about the change in the regulation and only occur in Year 1. Adjustment costs are costs incurred by firms to determine workers’ exemption statuses, notify employees of policy changes, and update payroll systems. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 Managerial costs associated with this proposed rulemaking occur because employers may spend more time scheduling newly nonexempt employees and more closely monitor their hours to minimize or avoid paying the overtime premium. The Department estimated costs in Year 1 assuming the first year of the analysis was 2013. The Department estimated that in Year 1 regulatory familiarization costs would equal $254.5 million, Year 1 adjustment costs would sum to $160.1 million, and Year 1 PO 00000 Frm 00051 Fmt 4701 Sfmt 4702 managerial costs would total $178.1 million (Table 17). Total direct employer costs in Year 1 were estimated to equal $592.7 million. Adjustment costs and management costs are ongoing and will need to be projected for future years (section VII.D.x.). Costs that are not quantified are discussed in section VII.D.iv.5. Adjustment costs and managerial costs associated with automatically updating the standard salary level are discussed in section VII.E.iii. E:\FR\FM\06JYP2.SGM 06JYP2 38566 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 17—SUMMARY OF YEAR 1 DIRECT EMPLOYER COSTS OF THIS PROPOSED RULE (MILLIONS) Direct employer costs Standard salary level HCE Compensation level Regulatory familiarization a .......................................................................................................... Adjustment ................................................................................................................................... Managerial ................................................................................................................................... ........................ 158.8 176.0 ........................ $1.2 2.1 $254.5 160.1 178.1 Total direct costs .................................................................................................................. 334.8 3.3 592.7 tkelley on DSK3SPTVN1PROD with PROPOSALS2 a Regulatory familiarization costs are assessed jointly for the change in the standard salary level and the HCE compensation level. 2. Regulatory Familiarization Costs A change in the standard EAP weekly salary test to the proposed level would impose direct costs on businesses by requiring them to review the regulation. It is not clear whether regulatory familiarization costs are a function of the number of establishments or the number of firms. The Department believes that generally the headquarters of a firm will conduct the regulatory review for the entire company; however, some firms provide more autonomy to their establishments, and in such cases regulatory familiarization may occur at the establishment level. To be conservative, the Department uses the number of establishments in its cost estimate because this provides a larger cost estimate. The Department believes that all establishments will incur regulatory familiarization costs, even if they do not employ exempt workers, because all establishments will need to confirm whether this proposed rulemaking includes any provisions that may impact their workers. Firms with more affected EAP workers will likely spend more time reviewing the regulation than firms with fewer or no affected EAP workers (since a careful reading of the regulations will probably follow the initial decision that the firm is affected). However, the Department does not know the distribution of affected EAP workers across firms and so an average cost per establishment is used. No data were identified from which to estimate the amount of time required to review the regulation. The Department requests that commenters provide data if possible. For this NPRM, the Department estimated establishments will use on average one hour of time because the proposed regulation is narrowly focused on the salary level tests. To estimate the total regulatory familiarization costs, three pieces of information must be estimated: (1) A wage level for the employees reviewing the rule; (2) the number of hours each employee spends reviewing the rule; and (3) the number of establishments VerDate Sep<11>2014 Total 19:25 Jul 02, 2015 Jkt 235001 employing workers. The Department’s analysis assumed that mid-level human resource workers with a median wage of $23.63 per hour will review the proposed rule.91 Assuming benefits are paid at a rate of 45 percent of the base wage and one hour of time is required for regulatory familiarization, the average cost per establishment is $34.19.92 The number of establishments with paid employees in 2011 was 7.44 million.93 Regulatory familiarization costs in Year 1 were estimated to be $254.5 million ($34.19 per establishment × 1 hour × 7.44 million establishments).94 In future years, new firms will be formed and may incur regulatory familiarization costs. However, the Department believes the incremental cost of this regulation will be zero since new firms will only need to familiarize themselves with the updated law, instead of the old law. 3. Adjustment Costs A change in the EAP salary test to the proposed level will impose direct costs on firms by requiring them to redetermine the exemption status of employees, update and adapt overtime 91 Calculated as the median wage in the CPS for workers with the occupation ‘‘human resources, training, and labor relations specialists’’ (0620) in 2013. The Department determined this occupation includes most of the workers who would conduct these tasks. Bureau of Labor Statistics, U.S. Department of Labor, Occupational Outlook Handbook, 2014–15 Edition, Human Resources Specialists and Labor Relations Specialists, available at: https://www.bls.gov/ooh/business-andfinancial/human-resources-specialists-and-laborrelations-specialists.htm. 92 The benefits-earnings ratio is derived from the BLS’s Employer Costs for Employee Compensation (ECEC) data. 93 Data for 2011 was the most recent available at the time of writing. Survey of U.S. Businesses 2011. Available at: https://www.census.gov/econ/susb/. Also included in the number of establishments incurring regulatory familiarization costs are the 90,106 state and local governments reported in the 2012 Census of Governments: Employment Summary Report. Available at: https:// www2.census.gov/govs/cog/g12_org.pdf. 94 As previously noted, the Department chose to use the number of establishments rather than the number of firms to provide a more conservative estimate of the regulatory familiarization cost. Using the number of firms, 5.8 million, would result in a reduced regulatory familiarization cost estimate of $197.4 million in Year 1. PO 00000 Frm 00052 Fmt 4701 Sfmt 4702 policies, notify employees of policy changes, and adjust their payroll systems. The Department believes the size of these costs will depend on the number of affected EAP workers and will occur in any year when the salary level is raised and exemption status is changed for some workers. To estimate adjustment costs three pieces of information must be estimated: (1) A wage level for the employees making the adjustments; (2) the amount of time spent making the adjustments; and (3) the estimated number of newly affected EAP workers. The Department again estimated that the average wage with benefits for human resources, training, and labor relations specialists is $34.19 per hour (as explained above). No applicable data were identified from which to estimate the amount of time required to make these adjustments.95 The Department requests that commenters provide any applicable data. For this NPRM, the Department chose to use one hour of time per affected worker. The estimated number of affected EAP workers in Year 1 is 4.682 million (as discussed in section VII.D.iii.). Therefore, total Year 1 adjustment costs were estimated to equal $160.1 million ($34.19 × 1 hour × 4.682 million workers). Adjustment costs may be partially offset by a reduction in the cost to employers of determining employees’ exempt status. Currently, to determine whether an employee is exempt firms must apply the duties test to salaried workers who earn at least $455 per week. Following this rulemaking, firms will no longer be required to apply the potentially time consuming duties test to employees earning less than the proposed salary level. This will be a clear cost savings to employers for employees who do not pass the duties test and earn at least $455 per week but less than the proposed salary level. The Department did not estimate the potential size of this cost savings. 95 Costs in the 2004 Final Rule were considered but because that revision included changes to the duties test the cost estimates are not directly applicable. E:\FR\FM\06JYP2.SGM 06JYP2 38567 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 4. Managerial Costs If employers reclassify employees as overtime eligible due to the changes in the salary levels, then firms may incur ongoing managerial costs associated with this proposed rulemaking because the employer may schedule and more closely monitor an employee’s hours to minimize or avoid paying the overtime premium. These costs are in addition to the one-time regulatory familiarization and adjustment costs described above. For example, when scheduling hours the manager may have to assess whether the marginal benefit of scheduling the worker for more than 40 hours exceeds the marginal cost of paying the overtime premium. Additionally, the manager may have to spend more time monitoring the employee’s work and productivity since the marginal cost of employing the worker per hour has increased. Because there was little precedent or data to aid in evaluating these costs, the Department examined several sources to estimate costs. First, prior part 541 rulemakings were reviewed to determine whether managerial costs were estimated. No estimates were found. This cost was not quantified for the 2004 rulemaking. Second, a literature review was conducted in an effort to identify information to help guide the cost estimates; again, no estimates were found. The Department requests data from the public applicable to this cost estimate. Despite a lack of available data, the Department chose to include estimated managerial costs to produce as full and accurate a cost estimate to employers as possible. To provide a sense of the potential magnitude of these costs, the Department estimated these costs assuming that management spends an additional five minutes per week scheduling and monitoring each affected worker expected to be reclassified as overtime eligible as a result of this NPRM, and whose hours are adjusted (1,022,000 affected EAP workers as calculated in section VII.D.vi.). As will be discussed in detail below, most affected workers do not currently work overtime, and there is no reason to expect their hours worked to change when their status changes from exempt to nonexempt. Similarly, employers are likely to find that it is less costly to give some workers a raise in order to maintain their exempt status. For both these groups of workers, management will have little or no need to increase their monitoring of hours worked. Under these assumptions, the additional managerial hours worked per week were estimated to be 85,200 hours rounded ((5 minutes/60 minutes) × 1,022,000 workers). The median hourly wage in 2013 for a manager was $27.78 and benefits were paid at a rate of 45 percent of the base wage, which totaled $40.20 per hour.96 97 Multiplying the additional 85,200 weekly managerial hours by the hourly wage of $40.20 and 52 weeks per year, the Year 1 costs were estimated to total $178.1 million for the proposed standard salary level. Although the exact magnitude would vary with the number of affected EAP workers each year, these costs would be incurred annually. 5. Other Potential Costs In addition to the costs discussed above, there may be additional costs that have not been quantified. Other categories of unquantified costs are discussed in section VII.D.vii and immediately below. Reduced Profits The increase in worker earnings’ resulting from the revised salary level is a transfer of income from firms to workers, not a cost, and is thus neutral concerning its primary effect on welfare and gross domestic product (GDP). However, there are potential secondary effects (both costs and benefits) of the transfer due to the potential difference in the marginal utility of income and the marginal propensity to consume between workers and business owners. The transfer may result in societal gain during periods when the economy is operating below potential to the extent that transferring income to workers with a relatively high marginal propensity to consume results in a larger multiplier effect and impact on GDP. Conversely, this transfer may also reduce the profits available to firms for business investment. Hiring Costs One of Congress’ goals in enacting the FLSA in 1938 was to spread employment to a greater number of workers by effectively raising the wages of employees working more than 40 hours per week. To the extent that firms respond to an update to the salary level test by reducing overtime, they may do so by spreading hours to other workers, including: Current workers employed for less than 40 hours per week by that employer, current workers who retain their exempt status, and newly hired workers. If new workers are hired to absorb these transferred hours, then the associated hiring costs are a cost of this proposed rulemaking. The reduction in hours is considered in more detail in section VII.D.v. v. Transfers 1. Overview Transfer payments occur when income is redistributed from one party to another. The Department has quantified two possible transfers likely to result from this proposed update to the salary level tests: (1) Transfers to employees from employers to ensure compliance with the FLSA minimum wage provision; and (2) transfers to employees from employers to ensure compliance with the FLSA overtime pay provision. Transfers in Year 1 to workers from employers due to the minimum wage provision were estimated to equal $46.7 million. The proposed increase in the HCE exemption compensation level does not affect minimum wage transfers because workers eligible for the HCE exemption earn well above the minimum wage. Transfers to employees from employers due to the overtime pay provision were estimated to be $1,435.8 million, $1,394.2 million of which is from the increased standard salary level, while the remainder is attributable to the increased HCE compensation level. Total Year 1 transfers were estimated to be $1,482.5 million (Table 18). TABLE 18—SUMMARY OF YEAR 1 REGULATORY TRANSFERS tkelley on DSK3SPTVN1PROD with PROPOSALS2 (Millions) Standard salary level Transfer from employers to workers Due to minimum wage ................................................................................................................. 96 Calculated as the median wage in the CPS for workers in management occupations (excluding chief executives) in 2013. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 97 The adjustment ratio is derived from the BLS’s Employer Costs for Employee Compensation (ECEC) PO 00000 Frm 00053 Fmt 4701 Sfmt 4702 $46.7 HCE Compensation level $0.0 Total $46.7 data using variables CMU1020000000000D and CMU1030000000000D. E:\FR\FM\06JYP2.SGM 06JYP2 38568 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 18—SUMMARY OF YEAR 1 REGULATORY TRANSFERS—Continued (Millions) Standard salary level Transfer from employers to workers HCE Compensation level Total Due to overtime pay .................................................................................................................... 1,394.2 41.7 1,435.8 Total transfers ....................................................................................................................... 1,440.8 41.7 1,482.5 Because the overtime premium depends on the base wage, the estimates of minimum wage transfers and overtime transfers are linked. This can be considered a two-step approach. The Department first identified affected EAP workers with an implicit regular hourly wage lower than the minimum wage, and then calculated the wage increase necessary to reach the minimum wage. The implicit regular rate of pay is calculated as usual weekly earnings divided by usual weekly hours worked. For those employees whose implicit regular rate of pay is below the minimum wage, the overtime premium was based on the minimum wage as the regular rate of pay. 2. Transfers Due to the Minimum Wage Provision Transfers from employers to workers to ensure compliance with the federal minimum wage are small compared to the transfers attributed to overtime pay and are only associated with the change in the standard salary level (workers currently eligible for the HCE test earn well above the minimum wage). For purposes of this analysis, the hourly rate of pay is calculated as usual weekly earnings divided by usual weekly hours worked. In addition to earning low wages, this set of workers earns an hourly rate below the federal minimum wage and also works many hours per week. To demonstrate, in order to earn less than the federal minimum wage of $7.25 per hour, but at least $455 per week, these workers must regularly work significant amounts of overtime (since $455/$7.25 = 62.8 hours). The applicable minimum wage is the higher of the federal minimum wage and the state minimum wage. Most affected EAP workers already receive at least the minimum wage; an estimated 12,000 affected EAP workers (less than 0.3 percent of all affected EAP workers) currently earn an implicit hourly rate of pay less than the minimum wage. The Department estimated transfers due to payment of the minimum wage by calculating the change in earnings if wages rose to the minimum wage for workers who become nonexempt and thus would have to be paid the minimum wage.98 In response to an increase in the regular rate of pay to the minimum wage, employers may reduce the workers’ hours, which must be considered when estimating transfers attributed to payment of the minimum wage to newly overtime-eligible workers. In theory, because the quantity of labor hours demanded is inversely related to wages, a higher mandated wage could result in fewer hours of labor demanded. However, the weight of the empirical evidence finds that increases in the minimum wage have caused little or no significant job loss.99 Thus, in the case of this proposed regulation, the Department believes that any disemployment effect due to the effect of the minimum wage provision would be negligible. This is partially due to the small number of workers affected by this provision. The Department estimated the potential disemployment effects (i.e., the estimated reduction in hours) of the transfer attributed to the minimum wage by multiplying the percent change in the regular rate of pay by a labor demand elasticity of ¥0.075.100 At the proposed salary level ($921 per week), the Department estimated that 12,000 affected EAP workers will on average see an hourly wage increase of $0.98, work 1.0 fewer hour per week, and receive an increase in weekly earnings of $74.0 as a result of coverage by the minimum wage provisions (Table 19). Thus, the total change in weekly earnings due to the payment of the minimum wage was estimated to be $897,300 per week ($74.0 × 12,000) or $46.7 million in Year 1. TABLE 19—MINIMUM WAGE ONLY: MEAN HOURLY WAGES, USUAL OVERTIME HOURS, AND WEEKLY EARNINGS FOR AFFECTED EAP WORKERS, 2013 Hourly wage a Before proposed regulation ............................................................................. After proposed regulation ................................................................................ Change ............................................................................................................ Usual weekly hours $7.09 8.07 0.98 76.4 75.4 ¥1.0 Usual weekly earnings $529.1 603.1 74.0 Total weekly transfer (1,000s) b — — $897.3 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Note: Pooled data for 2011–2013. aThe applicable minimum wage is the higher of the federal minimum wage and the state minimum wage. bUsual weekly earnings multiplied by the 12,000 exempt workers with an implicit regular rate of pay below the minimum wage who would lose their exemption status under the proposed rulemaking if weekly earnings did not change. 98 Because these workers’ hourly wages will be set at the minimum wage after the proposed rule, their employers will not be able to adjust their wages downward to offset part of the cost of paying the overtime pay premium (which will be discussed in the following section). Therefore, these workers will generally receive larger transfers attributed to the overtime pay provision than other workers. 99 Belman, D., and P.J. Wolfson (2014). What Does the Minimum Wage Do? Kalamazoo, MI: W.E. Upjohn Institute for Employment Research. Dube, VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 A., T.W. Lester, and M. Reich. (2010). Minimum Wage Effects Across State Borders: Estimates Using Contiguous Counties. IRLE Working Paper No. 157– 07. https://irle.berkeley.edu/workingpapers/15707.pdf. Schmitt, J. (2013). Why Does the Minimum Wage Have No Discernible Effect on Employment? Center for Economic and Policy Research. 100 This is based on the estimated impact of a change in the minimum wage from $7.25 to $9.00 per hour on the employment of teenagers from Congressional Budget Office. (2014). The Effects of PO 00000 Frm 00054 Fmt 4701 Sfmt 4702 a Minimum-Wage Increase on Employment and Family Income. While an elasticity estimate for adult workers would be more appropriate, the report stated that the elasticity for adults was ‘‘about one-third of the elasticity’’ for teenagers, without providing a specific value. In addition, the literature for adults is more limited. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 3. Transfers Due to the Overtime Pay Provision The proposed rule will also transfer income to affected EAP workers working in excess of 40 hours per week through payment of overtime to workers earning between the current and proposed salary levels. The size of the transfers will depend largely on how employers respond to the proposed salary level for affected EAP workers who work overtime. Employers may respond by: (1) Paying the required overtime premium to affected workers for the same number of overtime hours at the same implicit regular rate of pay; (2) reducing the regular rate of pay for workers working overtime; (3) eliminating overtime hours and potentially transferring some of these hours to other workers; (4) increasing workers’ salary to the proposed salary level; or (5) using some combination of these responses. How employers will respond depends on the relative costs of each of these alternatives; in turn, the relative costs of each of these alternatives are a function of workers’ earnings and hours worked. The simplest approach to estimating these transfer payments would be to multiply an employee’s regular rate of pay (after compliance with the minimum wage) by 1.5 for all overtime hours; this is referred to as the ‘‘full overtime premium’’ model.101 However, due to expected wage and hour adjustments by employers, this would likely overestimate the size of the transfer. Therefore, the Department used a methodology that allows for employer adjustments, such as changes in the regular rate of pay or hours worked. The size of these adjustments is likely to vary depending on the affected worker’s salary and work patterns. tkelley on DSK3SPTVN1PROD with PROPOSALS2 Employer Adjustments to the Regular Rate of Pay This section focuses on evaluating employers’ responses to affected EAP workers who work regular overtime (usually work more than 40 hours in a week). The requirement that employers pay newly nonexempt employees in accordance with minimum wage and overtime requirements may result in changes in employment conditions; requiring an overtime premium increases the marginal cost of labor, 101 The implicit regular rate of pay is calculated as usual weekly earnings divided by usual weekly hours worked. For example, the regular rate of pay for an employee previously ineligible for overtime whose usual weekly earnings was $600 and usual weekly hours was 50 would be $12. Under the full overtime premium model, this employee would receive $660 (40 hours × $12) + (10 hours × $12 × 1.5). VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 which employers will likely try to offset by adjusting wages or hours. How employers respond to a new salary level and the ensuing changes in employment conditions will depend on the demand for labor, current wages, employer and employee bargaining power, and other factors. To model employer responses, the Department used a method that reflects the average response among all employers for all affected workers. However, individual employer responses will vary. Two conceptual models are useful for thinking about how employers may respond to reclassifying certain employees as overtime eligible: The ‘‘full overtime premium’’ model and the ‘‘employment contract’’ model.102 These models make different assumptions about the demand for overtime hours and the structure of the employment agreement which result in different implications for predicting employer responses. The full overtime premium model is based on the traditional ‘‘labor demand’’ model of determining wage and hour conditions. In the labor demand model, employers and employees negotiate fixed hourly wages and then subsequently negotiate hours worked, rather than determining both hours and pay simultaneously. This model assumes employees are aware of the hourly wage rate they negotiated and may be more reluctant to accept downward adjustments. The labor demand model would apply if employees had a contract to be paid at an hourly rate, meaning that employers could not reduce the regular rate of pay in response to the requirement to pay a 50 percent premium on hours worked beyond 40 in a week. However, the increase in the cost of labor would lead to a reduction in the hours of labor demanded as long as labor demand is not completely inelastic. The full overtime premium model is a particular scenario of the labor demand model in which the demand for labor is completely inelastic, that is employers will demand the same number of hours worked regardless of the cost. In the employment contract model, employers and employees negotiate total pay and hours simultaneously, rather than negotiating a fixed hourly wage and then determining hours. Under this model, when employers are 102 The employment contract model is also known as the fixed-job model. See Trejo, S.J. (1991). The Effects of Overtime Pay Regulation on Worker Compensation. American Economic Review, 81(4), 719–740 and Barkume, A. (2010). The Structure of Labor Costs with Overtime Work in U.S. Jobs. Industrial and Labor Relations Review, 64(1), 128– 142. PO 00000 Frm 00055 Fmt 4701 Sfmt 4702 38569 required to pay employees an overtime premium, they adjust the employees’ implicit hourly rate of pay downward so that when the overtime premium is paid total employee earnings (and thus total employer cost) remain constant, along with the employees’ hours. The employer does not experience a change in cost and the employee does not experience a change in earnings or hours. The employment contract model would hold if the workers who are reclassified as overtime protected had an employment agreement specifying set total earnings and hours of work. The employment contract model tends to be more applicable to salaried workers while the labor demand model is generally more applicable to workers paid hourly. Since all affected EAP workers in this analysis are salaried, the Department believes the employment contract model may be more appropriate for estimating employer response to the proposed salary increase. However, the employment contract model may not always hold true due to market constraints, employer incentives, or workers’ bargaining power. Four examples are provided. • Employers are constrained because they cannot reduce an employee’s implicit hourly rate of pay below the minimum wage. If the employee’s implicit hourly rate of pay before the change is at or below the minimum wage, then employers will not be able to reduce the rate of pay to offset the cost of paying the overtime premium. • Employees generally have some, albeit limited, bargaining power which may prevent employers from reducing the employee’s implicit hourly rate of pay to fully offset increased costs. • Employers may be hesitant to reduce the employee’s implicit hourly rate of pay by the entire amount predicted by the employment contract model because it may hurt employee morale and consequently productivity. • Employers are often limited in their ability to pay different regular rates of pay to different employees who perform the same work and have the same qualifications. In order to keep wages constant across employees and reduce wages for overtime workers, employers would need to reduce the implicit hourly rate of pay for employees who do not work overtime as well as those who do work overtime. This would reduce total earnings for these non-overtime employees (potentially causing retention problems, productivity losses, and morale concerns). Therefore, the likely outcome will fall somewhere between the conditions predicted by the full overtime premium and employment contract models. For E:\FR\FM\06JYP2.SGM 06JYP2 38570 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 example, the implicit hourly rate of pay may fall, but not all the way to the wage predicted by the employment contract model, and overtime hours may fall but not be eliminated since the implicit hourly rate of pay has fallen. The Department conducted a literature review to evaluate how the market would adjust to a change in the requirement to pay overtime. Barkume (2010) and Trejo (1991) empirically tested for evidence of these two competing models by measuring labor market responses to the application of FLSA overtime pay regulations.103 Both concluded that wages partially adjust toward the level consistent with the employment contract model in response to the overtime pay provision.104 Barkume found that employee wage rates were adjusted downward by 40 to 80 percent of the amount the employment contract model predicted, depending on modeling assumptions. Earlier research had demonstrated that in the absence of regulation some employers may voluntarily pay workers some overtime premium to entice them to work longer hours, to compensate workers for unexpected changes in their schedules, or as a result of collective bargaining.105 Thus Barkume assumed that workers would receive an average voluntary overtime pay premium of 28 percent in the absence of an overtime pay regulation. Including this voluntary overtime pay from employers, he estimated that in response to overtime pay regulation, the wage adjusted downward by 80 percent of the amount that would occur with the employment contract model. Conversely, when Barkume assumed workers would receive no voluntary overtime pay premium in the absence of an overtime 103 Barkume, A. (2010). The Structure of Labor Costs with Overtime Work in U.S. Jobs. Industrial and Labor Relations Review, 64(1), 128–142. Trejo, S.J. (1991). The Effects of Overtime Pay Regulation on Worker Compensation. American Economic Review, 81(4), 719–740. 104 Since both papers were based on crosssectional data, findings were assumed to be at the final equilibrium wages. Studies showing wage contracts are likely to be stickier in the short run than in the long run have limited applicability here since this analysis deals exclusively with salaried workers who are less likely to be aware of their implicit hourly wage rate. The Department has modeled a sticky adjustment process by assuming the wage elasticity of demand for labor is smaller in Year 1 than in subsequent years. 105 Barzel, Y. (1973). The Determination of Daily Hours and Wages. The Quarterly Journal of Economics, 87(2), 220–238 demonstrated that modest fluctuations in labor demand could justify substantial overtime premiums in the employment contract model. Hart, R.A. and Yue, M. (2000). Why Do Firms Pay an Overtime Premium? IZA Discussion Paper No. 163, showed that establishing an overtime premium in an employment contract can reduce inefficiencies. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 pay regulation, wages adjusted downward 40 percent of the amount the employment contract model predicted.106 107 However, while it seemed reasonable that some premium was paid for overtime in the absence of regulation, Barkume’s assumption of a 28 percent initial overtime premium is likely too high for the salaried workers potentially affected by a change in the salary and compensation level requirements for the EAP exemptions.108 Modeling Employer Adjustments to the Hourly Rate of Pay and Overtime Hours In practice, employers do not seem to adjust wages of regular overtime workers to the full extent indicated by the employment contract model, and thus employees appear to get a small but significant increase in weekly earnings due to coverage by overtime pay regulations. Barkume and Trejo found evidence partially supporting both the employment contract model and the full overtime premium model in response to a 50 percent overtime premium requirement: A decrease in the regular rate of pay for workers with overtime (but not the full decrease to the employment contract model level) and a decrease in the probability of working overtime. Therefore, when modeling employer responses with respect to the adjustment to the regular rate of pay, the Department used a method that falls 106 Barkume’s estimates are consistent with Trejo’s 1991 finding that the wage adjustment when there is no overtime premium was only about 40 percent of the full employment contract model adjustment. Trejo’s estimates range from 25 percent to 49 percent and average 40 percent. Trejo, S.J. (1991). The Effects of Overtime Pay Regulation on Worker Compensation. American Economic Review, 81(4), 719–740. 107 Consider a worker earning $500 and working 50 hours per week. Assuming no overtime premium is paid the imputed hourly rate of pay is $10. Assuming a 28 percent overtime premium, the hourly rate of pay is $9.47 (($9.47 × 40) + (($9.47 × 1.28) × 10)) = $500. If the hourly rate of pay was fully adjusted to the employment contract model level when overtime pay is newly required, the hourly rate of pay would be $9.09 (($9.09 × 40) + (($9.09 × 1.5) × 10)) = $500. Forty percent of the adjustment from $10 to $9.09 results in an adjusted regular rate of pay of $9.64. Eighty percent of the adjustment from $9.47 to $9.09 results in an adjusted hourly rate of pay of $9.17. The Department took the average of these two adjusted wages to estimate that the resulting hourly rate of pay would be $9.40. 108 Barkume (2010) based this assumption on the findings of Bell, D. and Hart, R. (2003). Wages, Hours, and Overtime Premia: Evidence from the British Labor Market. Industrial and Labor Relations Review, 56(3), 470–480. This study used 1998 data on male, non-managerial full-time workers in Britain. British workers were likely paid a larger voluntary overtime premium than American workers because Britain did not have a required overtime pay regulation and so collective bargaining played a larger role in implementing overtime pay. PO 00000 Frm 00056 Fmt 4701 Sfmt 4702 somewhere between the employment contract model and the full overtime premium model (i.e., the partial employment contract model). Barkume reported two methods to estimate this partial employment contract wage, depending on the amount of overtime pay assumed to be paid in the absence of regulation. As noted above, the Department believes both the model assuming a voluntary 28 percent overtime premium and the model assuming no voluntary overtime premium are unrealistic for the affected population. Therefore, lacking more information, the Department determined that an appropriate estimate of the impact on the implicit hourly rate of pay for regular overtime workers after the proposed rule should be determined using the average of Barkume’s two estimates of partial employment contract model adjustments: A wage change that is 40 percent of the wage change assuming an initial zero overtime pay premium, and a wage change that is 80 percent of the wage change assuming an initial 28 percent overtime pay premium.109 This is approximately equivalent to assuming that overtime workers received a 14 percent overtime premium in the absence of regulation (the mid-point between 0 and 28 percent). How employers adjust workers’ wages and hours depends on employment conditions. The discussion begins with a description of how employment conditions affect employers’ wage adjustments depending on the differing work characteristics of their employees. However, changing employees’ earnings is also likely to result in adjustments to hours worked. Thus, after estimating wage adjustments the Department calculated the adjustments to hours worked as a function of the new wage. Finally, transfers from employers to employees were estimated as a function of the changes in wages and the changes in hours. The Department identified four types of workers whose work characteristics impact how employers were modeled to respond to the proposed changes in both the standard and HCE salary levels: 109 Both studies considered a population that included hourly workers. Evidence is not available on how the adjustment towards the employment contract model differs between salaried and hourly workers. The employment contract model may be more likely to hold for salaried workers than for hourly workers since salaried workers directly observe their weekly total earnings, not their implicit equivalent hourly wage. Thus, applying the partial adjustment to the employment contract model as estimated by these studies may overestimate the transfers from employers to workers who are salaried. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 • Type 1: Workers who do not work overtime. These workers will not experience any adjustment in their hourly rate of pay. • Type 2: Workers who do not regularly work overtime but occasionally work overtime.110 Some of these workers’ implicit hourly rate of pay will fall.111 Others will have no change in their hourly rate of pay. • Type 3: Workers who regularly work overtime. These workers’ implicit hourly rate of pay falls to reflect the partial employment contract model adjustment.112 • Type 4: Workers who regularly work overtime. These workers differ from the Type 3 workers in that once wages and hours are adjusted, weekly earnings are greater than the proposed salary level, so employers will increase these workers’ earnings to the proposed salary level so they can continue to claim the EAP exemption for them.113 Type 1 affected EAP workers will become overtime eligible, but since they do not work overtime, they will see no change in their weekly earnings. Type 2 and Type 3 affected EAP workers will become overtime eligible and must be paid the overtime premium for any overtime hours worked and may see changes in their regular rate of pay, and/ or hours, and thus weekly earnings. As explained in more detail below, Type 2 and Type 3 affected workers were modeled differently due to the difference in the nature of the overtime hours worked. Type 3 workers receive wages adjusted for partial compliance with the employment contract model and their hours adjust in response. Type 4 workers are those who regularly work overtime, but will remain exempt because their weekly earnings will be raised to the proposed EAP salary level (either the standard salary level or HCE compensation level depending on 110 Type 2 workers are those who worked overtime in the survey week (the week referred to in the CPS MORG questionnaire). If a different week was chosen as the survey week then likely some of these workers would not have worked overtime. However, because the data are representative of both the population and all twelve months in a year, the Department believes the share of Type 2 workers in the given week is representative of an average week in the year. 111 The Department assumes that Type 2 workers are currently paid additional wages for overtime hours worked at the usual hourly wage rate. Specifically, Type 2 workers’ actual earnings for the week are calculated as (usual weekly earnings/usual hours worked) x (actual hours worked last week). 112 The reduction in the regular hourly wage is restricted by the minimum wage; the wage cannot fall below the minimum wage. 113 It is possible that employers will increase the salaries paid to some ‘‘occasional’’ overtime workers to maintain the exemption for the worker, but the Department has no way of identifying these workers. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 which test the worker passed). How employers respond to workers who work overtime hours is described in more detail in the following paragraphs for Type 2 and Type 3 workers. The Department distinguishes those who regularly work overtime (Type 3 workers) from those who occasionally, or irregularly, work overtime (Type 2 workers) because employer adjustment to the proposed rule may differ accordingly. The Department believes that employers are more likely to adjust hours worked and wages for regular overtime workers because their hours are predictable. Conversely, it may be more difficult to adjust hours and wages for occasional overtime workers because employers may be responding to a transient, perhaps unpredicted, shift in market demand for the good or service they provide. In this case it is likely advantageous for the employer to pay for this occasional overtime rather than to adjust permanent staffing. Additionally, the transient and possibly unpredicted nature of the change may make it difficult to adjust wages for these workers. The Department treats Type 2 affected workers in two ways due to the uncertainty of the nature of these occasional overtime hours worked. If these workers work extra hours on an unforeseen, short-term, as-needed basis (e.g., to adjust to unanticipated increases in demand), then there may be less opportunity for employers to adjust straight-time wages downward.114 However, if these workers work extra hours on a foreseen, periodic basis (e.g., work a few extra hours one week each month, but workers do not consider it ‘‘regular overtime’’ because they do not work overtime during three weeks each month), then there may be some opportunity for employers to adjust straight-time wages downward (e.g., so pre- and post-revision monthly income is more similar). That this overtime is periodic and predictable is what makes it much more similar to that worked by Type 3 workers, and provides employers with more opportunity to adjust hours and wages. Since in reality there is likely a mix of these two 114 Employers may be reluctant to reset hourly wage rates to respond to unexpected changes to the need for overtime because the negative impact on worker morale may outweigh the gains from adjusting wages to unexpected shifts in demand. Of relevance is the well-established literature that shows employers do not quickly adjust wages downward in regard to downturns in the economy; the same logic applies to our approach to unexpected changes in demand. See, for example: Bewley, T. (1999). Why Wages Don’t Fall During a Recession. Cambridge, MA: Harvard University Press. See also Barzel, Y. (1973). The Determination of Daily Hours and Wages. The Quarterly Journal of Economics, 87(2), 220–238. PO 00000 Frm 00057 Fmt 4701 Sfmt 4702 38571 occasional overtime scenarios, the Department combines models representing these two scenarios when estimating impacts.115 Our estimate for how Type 2 workers are affected is based on the assumption that 50 percent of these workers who worked occasional overtime worked expected overtime hours and the other 50 percent worked unexpected overtime. Workers were randomly assigned to these two groups. Workers with expected occasional overtime hours were treated like Type 3 affected workers (partial employment contract model adjustments). Workers with unexpected occasional overtime hours were assumed to receive a 50 percent pay premium for the overtime hours worked (full overtime premium model). Since affected Type 2 and Type 3 EAP workers work more than 40 hours per week, whether routinely or occasionally, they will now be overtime protected. These workers will receive an overtime premium based on their implicit hourly wage adjusted as described above. Because employers must now pay more for the same number of labor hours, they will seek to reduce those hours; in economics, this is described as a decrease in the quantity of labor hours demanded (a movement to the left along the labor demand curve). It is the net effect of these two changes that will determine the final weekly earnings for affected EAP workers. Next we describe how these workers’ hours adjust in response to the change in their implicit hourly wage and the requirement to pay an overtime premium on that wage for each hour worked in excess of 40 hours per week. The reduction in hours is calculated using the elasticity of labor demand with respect to wages. The Department used a short-run demand elasticity of 115 Trejo and Barkume’s adjustments are averages; excluding some workers (i.e., half of Type 2 workers) from these adjustments could potentially bias the size of the adjustment for the workers who continue to receive the adjustment. This bias would exist if Barkume and Trejo estimated the average adjustment for a sample of workers including irregular overtime workers and the size of the adjustment for these workers differs from other workers. It is not clear whether Trejo’s and Barkume’s samples include both occasional and regular overtime workers; however, the Department’s interpretation is that Trejo includes only workers who usually work overtime and Barkume includes both. If these assumptions are correct, the magnitude of this RIA’s adjustment made for the workers whose wages and hours are adjusted would be appropriate if it were applying Trejo’s results but may, due to applying Barkume’s, result in an underestimate of the average fall in base wages. We believe the magnitude of any potential bias will be small because the half of Type 2 workers who are occasional overtime workers (and thus treated differently) compose only 8 percent of Type 2 and Type 3 workers. E:\FR\FM\06JYP2.SGM 06JYP2 38572 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules ¥0.20 to estimate the percentage decrease in hours worked resulting from the increase in average hourly wages in Year 1 calculated using the adjusted base wage and the overtime wage premium.116 The interpretation of the short run demand elasticity in this context is that a 10 percent increase in wages will result in a 2 percent decrease in hours worked. Transfers projected for years 2 through 10 used a long-run tkelley on DSK3SPTVN1PROD with PROPOSALS2 116 This elasticity estimate is based on the Department’s analysis of Lichter, A., Peichl, A. & Siegloch, A. (2014). The Own-Wage Elasticity of Labor Demand: A Meta-Regression Analysis. IZA DP No. 7958. Some researchers have estimated larger impacts from own wage changes on the number of overtime hours worked (Hamermesh, D. and S. Trejo. (2000). The Demand for Hours of Labor: Direct Evidence from California. The Review of Economics and Statistics, 82(1), 38–47 conclude the price elasticity of demand for overtime hours is at least ¥0.5). The Department decided to use a general measure of elasticity applied to the average change in wages since the increase in the overtime wage is somewhat offset by a decrease in the nonovertime wage as indicated in the employment contract model, and welcomes comments on the appropriate elasticity to be used in this analysis. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 elasticity; this will be discussed in section VII.D.x.1.117 The Department calculates the percent increase in hourly wages since it must be used with the elasticity of labor demand to determine the change in hours. This is equal to workers’ new average hourly wage (including overtime pay) divided by their original implicit hourly wage. For Type 3 affected workers, and the 50 percent of Type 2 affected workers who worked expected overtime, we estimate adjusted total hours worked after making wage adjustments using the partial employment contract model. To estimate adjusted hours worked, we set the percent change in total hours worked equal to the percent change in average wages multiplied by the wage elasticity of labor demand.118 The wage 117 In the short-run not all factors of production can be changed and so the change in hours demanded is smaller than in the long run, when all factors are flexible. 118 In this equation, the only unknown is adjusted total hours worked. Since adjusted total hours worked is in the denominator of the left side of the equation and is also in the numerator of the right PO 00000 Frm 00058 Fmt 4701 Sfmt 4702 elasticity of labor demand was determined from a review of published econometric studies. The percent change in average wages is equal to the adjusted implicit average hourly wage minus the original implicit average hourly wage divided by the original implicit average hourly wage. The original implicit average hourly wage is equal to original weekly earnings divided by original hours worked. The adjusted implicit average hourly wage is equal to adjusted weekly earnings divided by adjusted total hours worked. Adjusted weekly earnings equals the adjusted hourly wage (i.e., after the partial employment contract model adjustment) multiplied by 40 hours plus adjusted hours worked in excess of 40 multiplied by 1.5 times the adjusted hourly wage. Figure 4 is a flow chart summarizing the four types of affected EAP workers. Also shown are the impacts on exempt status, weekly earnings, and hours worked for each type of affected worker. side of the equation, solving for adjusted total hours worked requires solving a quadratic equation. E:\FR\FM\06JYP2.SGM 06JYP2 a Affected EAP workers are those who are exempt under the current EAP exemptions and would gain minimum wage and overtime protection or receive a raise to the proposed increased salary level. b Depending on how employers respond to this rule, some workers may experience adverse consequences due to a reduction in their hours of work, potentially necessitating a second job to maintain their pre-rule earnings level. c Occasional overtime workers are those who responded that they (1) do not usually work overtime and (2) worked overtime in the survey week. In any given week different workers may be working occasional overtime but the Department assumes the total number of occasional overtime workers and VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 occasional overtime hours are similar across weeks. d The amount wages are adjusted downwards depends on whether the employment contract model or the labor demand model holds. The Department’s preferred method uses a combination of the two. Employers reduce the regular hourly wage rate somewhat in response to overtime pay requirements, but the wage is not reduced enough to keep total compensation constant. e Based on hourly wage and weekly hours it is more cost efficient for the employer to increase the worker’s weekly salary to the updated salary level than to pay overtime pay. f The Department assumed hours would not change due to lack of data and relevant PO 00000 Frm 00059 Fmt 4701 Sfmt 4702 38573 literature; however, it is possible employers will increase these workers’ hours in response to paying them a higher salary. Estimates of the Number of and Impacts on Affected EAP workers The Department projects 4.7 million workers will be affected by either (1) an increase in the standard salary level to the 40th earnings percentile because they earn salaries between $455 per week and $921 per week or (2) an increase in the HCE compensation level to the 90th earnings percentile, $122,148 annually. These workers are categorized into the four ‘‘types’’ identified previously. There are 3.5 E:\FR\FM\06JYP2.SGM 06JYP2 EP06JY15.003</GPH> tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38574 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules million Type 1 workers (74.7 percent of all affected EAP workers), those who work 40 hours per week or less and thus will not be paid an overtime premium despite their expected change in status to overtime protected (Table 20). Type 2 workers, those who are expected to become overtime eligible and do not usually work overtime but did work overtime in the survey week (i.e., occasional overtime workers), total 181,000 workers (3.9 percent of all affected EAP workers). Type 3 workers, those who are expected to become overtime eligible and be paid the overtime premium, are composed of an estimated 931,000 workers (19.9 percent of all affected EAP workers). The number of affected Type 4 workers was estimated to be 71,000 workers (1.5 percent of all affected workers); these are workers who the Department believes will remain exempt because firms will have a financial incentive to increase their weekly salaries to the proposed salary level so that they remain exempt, rather than pay a premium for overtime hours.119 TABLE 20—AFFECTED EAP WORKERS BY TYPE (1,000S), 2013 Regular OT Total a Standard salary level ................................................. HCE compensation level ........................................... No overtime worked (T1) Occasional OT (T2) Newly nonexempt (T3) Remain exempt (T4) 4,646 36.2 180 1.0 920 11.1 67 3.4 4,682 Total .................................................................... 3,478 20.7 3,499 181 931 71 Note: Pooled data for 2011–2013. a Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels). *Type 1: Workers without regular OT and without occasional OT. *Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly earnings increase, but regular rate of pay and hours fall for 50 percent of workers. *Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours increase, but regular rate of pay and hours fall. *Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level). tkelley on DSK3SPTVN1PROD with PROPOSALS2 The proposed rulemaking will likely impact affected workers’ wages, hours, and earnings. How these will change depends on the type of worker. Predicted changes in implicit wage rates are outlined in Table 21; changes in hours in Table 22; and changes in weekly earnings in Table 23. Type 1 workers will have no change in wages, hours, or earnings.120 Estimating changes in the regular rate of pay for Type 3 workers and the 50 percent of Type 2 workers who regularly work occasional overtime requires application of the partial employment contract model, which predicts a decrease in their average regular rates of pay. The Department estimates that employers would decrease these workers’ regular hourly rates of pay to the amount predicted by the partial employment contract model adjustment. Employers would not be able to adjust the regular rate of pay for the occasional overtime workers whose overtime is irregularly scheduled and unpredictable (the remaining 50 percent of Type 2 workers). As a group, Type 2 workers currently exempt under the standard test would see a decrease in their average regular hourly wage (i.e., excluding the overtime premium) from $18.30 to $17.88, a decrease of 2.3 percent. Type 2 workers paid between $100,000 and the proposed HCE compensation level would see an average decrease in their regular hourly wage from $52.99 to $50.85, a decrease of 4.0 percent. However, because workers will now receive a 50 percent premium on their regular hourly wage for each hour worked in excess of 40 hours per week, average weekly earnings for Type 2 workers would increase. Type 3 workers will also receive decreases in their regular hourly wage as predicted by the partial employment contract model. Type 3 affected workers paid below the proposed standard salary level would have their regular hourly rate of pay decrease on average from $14.71 to $13.93 per hour, a decrease of 5.3 percent. Type 3 workers paid between $100,000 and the proposed HCE compensation level would have their regular rate of pay decrease on average from $39.23 to $36.66 per hour, a decrease of 6.5 percent. Again, although regular hourly rates decline, weekly earnings will increase on average because these workers are now eligible for the overtime premium. Type 4 workers’ implicit hourly rates of pay would increase in order for their earnings to meet the proposed standard salary level ($921 per week) or the proposed HCE annual compensation level (122,148 annually). The implicit hourly rate for Type 4 affected EAP workers who had earned between $455 and $921 per week would increase on average from $16.40 to $16.72 (a 2.0 percent increase) (Table 21). The implicit hourly rate of pay for Type 4 workers who had earned between $100,000 and $122,148 annually would increase on average from $41.87 to $42.32 (a 1.1 percent increase). 119 As previously described, the Department calculated a wage and hour adjustment for all regular overtime workers. Consider, by way of example, a worker who initially earned $900 and worked 70 hours per week. Suppose the partial employment contract adjustment results in a regular rate of pay of $11.94 and 69.5 hours worked per week. After the partial employment contract adjustments, this worker would receive approximately $1,006 per week ((40 × $11.94) + (29.5 × ($11.94 × 1.5)). Since this is greater than the proposed standard salary level, the Department estimated that this worker would have his salary increased to $921 and remain exempt at that threshold. 120 It is possible that these workers may experience an increase in hours and weekly earnings because of transfers of hours from overtime workers. Due to the high level of uncertainty in employers’ responses regarding the transfer of hours, the Department did not have credible evidence to support an estimation of the number of hours transferred to other workers. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00060 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38575 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 21—AVERAGE REGULAR RATE OF PAY BY TYPE OF AFFECTED EAP WORKER, 2013 Regular OT No overtime worked (T1) Total Occasional OT (T2) Newly nonexempt (T3) Remain exempt (T4) Standard Salary Level Before proposed rule ........................................................... After proposed rule .............................................................. Change ................................................................................. $18.38 18.21 ¥0.17 $19.39 19.39 0.00 $18.30 17.88 ¥0.42 $14.71 13.93 ¥0.78 $16.40 16.72 0.33 $52.18 52.18 0.00 $52.99 50.85 ¥2.14 $39.23 36.66 ¥2.57 $41.87 42.32 0.45 HCE Compensation Level Before proposed rule ........................................................... After proposed rule .............................................................. Change ................................................................................. $47.26 46.46 ¥0.80 Note: Pooled data for 2011–2013. *Type 1: Workers without regular OT and without occasional OT. *Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly earnings increase, but regular rate of pay and hours fall for 50 percent of workers. *Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours increase, but regular rate of pay and hours fall. *Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level). Type 1 and Type 4 workers would have no change in hours. Type 1 workers’ hours would not change because they do not work overtime and thus the requirement to pay an overtime premium does not affect them. Type 4 workers’ hours would not change because they continue to be exempt, and therefore are not paid a premium for overtime hours. Type 2 and Type 3 workers would see a small decrease in their hours of overtime worked. This reduction in hours is relatively small and is due to the effect on labor demand of the increase in the average hourly base wage as predicted by the employment contract model.121 Type 2 workers who would be newly overtime eligible would see a decrease in average weekly hours in weeks where occasional overtime is worked, from 48.0 to 47.9 hours (0.3 percent) (Table 22).122 Type 2 workers who would no longer earn the HCE compensation level would see a decrease in average weekly hours in applicable weeks from 46.5 to 46.3 (0.4 percent). Type 3 workers affected by the increase in the standard salary level would see a decrease in hours worked from 50.7 to 50.3 hours per week (0.8 percent). Type 3 workers affected by the increase in the HCE compensation level would see an average decrease from 53.7 to 53.3 hours per week (0.8 percent). TABLE 22—AVERAGE WEEKLY HOURS FOR AFFECTED EAP WORKERS BY TYPE, 2013 Regular OT No overtime worked (T1) Total Occasional OT (T2) Newly nonexempt (T3) Remain exempt (T4) Standard Salary Level a Before proposed rule ........................................................... After proposed rule .............................................................. Change ................................................................................. 41.6 41.5 ¥0.1 38.6 38.6 0.0 48.0 47.9 ¥0.1 50.7 50.3 ¥0.4 56.9 56.9 0.0 39.8 39.8 0.0 46.5 46.3 ¥0.2 53.7 53.3 ¥0.4 56.4 56.4 0.0 HCE Compensation Level a tkelley on DSK3SPTVN1PROD with PROPOSALS2 Before proposed rule ........................................................... After proposed rule .............................................................. Change ................................................................................. 45.8 45.7 ¥0.1 Note: Pooled data for 2011–2013. a Usual hours for Types 1, 3, and 4 but actual hours for Type 2. *Type 1: Workers without regular OT and without occasional OT. *Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly earnings increase, but regular rate of pay and hours fall for 50 percent of workers. *Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours increase, but regular rate of pay and hours fall. *Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level). 121 The Department estimates that half of Type 2 workers will not see a reduction in their hours; however as a group, Type 2 workers are expected to experience a reduction in their hours of work. VerDate Sep<11>2014 20:30 Jul 02, 2015 Jkt 235001 122 Type 2 workers do not see increases in regular earnings to the new salary level (as Type 4 workers do) even if their new earnings exceed that new level. This is because the estimated new earnings PO 00000 Frm 00061 Fmt 4701 Sfmt 4702 only reflect their earnings in that week; their earnings for the entire year do not necessarily exceed the salary level. E:\FR\FM\06JYP2.SGM 06JYP2 38576 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules Because Type 1 workers do not experience a change in their regular rate of pay or hours they would have no change in earnings due to the proposed rule (Table 23). While their hours are not expected to change, Type 4 workers’ salaries would increase to the proposed standard salary level or HCE compensation level (depending on which test they pass). Thus, Type 4 workers’ average weekly earnings would increase by $20.47 (2.3 percent) for those affected by the change in the standard salary level and by $27.36 per week (1.2 percent) for those affected by the HCE compensation level. Although both Type 2 and Type 3 workers on average experience a decrease in both their regular rate of pay and hours worked, their weekly earnings are expected to increase as a result of the overtime premium. Based on a standard salary level of $921 per week, Type 2 workers’ average weekly earnings increase from $879.35 to $925.33, a 5.2 percent increase.123 The average weekly earnings of Type 2 workers affected by the change in the HCE compensation level were estimated to increase from $2,470.77 to $2,514.22, a 1.8 percent increase. Average weekly earnings of Type 3 workers also increase. For Type 3 workers affected by the standard salary level, average weekly earnings would increase from $731.54 to $751.13, an increase of 2.7 percent. Type 3 workers affected by the change in the HCE compensation level have an increase in average weekly earnings from $2,057.41 to $2,117.56, an increase of 2.9 percent. TABLE 23—AVERAGE WEEKLY EARNINGS FOR AFFECTED EAP WORKERS BY TYPE, 2013 Regular OT No overtime worked (T1) Total Occasional OT (T2) Newly Nonexempt (T3) Remain exempt (T4) Standard Salary Level a b Before proposed rule ........................................................... After proposed rule .............................................................. Change ................................................................................. $730.58 736.54 5.96 $719.31 719.31 0.00 $879.35 925.33 45.97 $731.54 751.13 19.60 $900.53 921.00 20.47 2,470.77 2,514.22 43.45 2,057.41 2,117.56 60.15 2,321.64 2,349.00 27.36 HCE Compensation Level a b Before proposed rule ........................................................... After proposed rule .............................................................. Change ................................................................................. 2,103.26 2,125.42 22.16 2,075.18 2,075.18 0.00 Note: Pooled data for 2011–2013. a The mean of the hourly wage multiplied by the mean of the hours does not necessarily equal the mean of the weekly earnings because the product of two averages is not necessarily equal to the average of the product. b Weekly earnings for weeks where overtime is worked. Thus for Type 3 and 4 workers weekly earnings is derived by multiplying the wage by usual hours worked but for Type 2 workers weekly earnings is derived by multiplying the wage by actual hours worked in the survey week. * Type 1: Workers without regular OT and without occasional OT. * Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly earnings increase, but regular rate of pay and hours fall for 50 percent of workers. * Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours increase, but regular rate of pay and hours fall. * Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level). tkelley on DSK3SPTVN1PROD with PROPOSALS2 Weekly earnings after an increase to the proposed standard salary level were estimated using the new wage (i.e., the partial employment contract model wage) and the reduced number of overtime hours worked. At the proposed standard salary level, the average weekly earnings of all affected workers will increase from $730.58 to $736.54, a change of $5.96 (0.8 percent). However, these figures mask the impact on workers whose hours and earnings will change because Type 1 workers make up more than 70 percent of the pool of affected workers. If Type 1 workers, who do not work overtime, are excluded the average increase in weekly earnings is $23.72. At the proposed standard salary level, multiplying the average change of $5.96 by the 4.6 million affected standard EAP workers equals an increase in earnings of $27.7 million per week or $1,441 million in the first year (Table 24). Of the weekly total, $897,000 is due to the minimum wage provision and $26.8 million stems from the overtime pay provision. For workers affected by the change in the HCE compensation level, average weekly earnings increase by $22.16 ($51.91 if Type 1 workers, who do not work overtime, are excluded). When multiplied by 36,000 affected workers, the national increase in weekly earnings will be $801,000 per week, or $41.7 million in the first year. Thus, Year 1 transfer payments attributable to this proposed rule total $1,482.5 million. If the Department assumed Type 2 workers received no additional pay for occasional overtime hours prior to the rulemaking (as discussed above), then Year 1 transfers would instead be $1,499.1 million. 123 For these calculations, the Department assumed Type 2 workers are paid their regular rate of pay for all occasional overtime hours. For example, if a Type 2 worker earned $700 per week and normally worked a 40 hour workweek then his or her regular rate of pay would be $17.50 per hour. If that person worked 10 hours of overtime in some week, he or she would earn $875 ($700 + $17.50 × 10) in that week. This is why baseline average weekly earnings are higher than for other types of workers. These workers do not see increases in regular earnings to the new salary level since their earnings only exceed the salary level in some weeks. If instead, the Department assumed Type 2 workers received no additional pay for occasional overtime hours, but merely received their usual weekly salary, then estimated baseline earnings would be smaller, and estimated transfers would be larger for these workers. VerDate Sep<11>2014 20:30 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00062 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38577 TABLE 24—TOTAL CHANGE IN WEEKLY AND ANNUAL EARNINGS FOR AFFECTED EAP WORKERS BY PROVISION, 2013 Total change in earnings (1,000s) Provision Weekly Total a ....................................................................................................................................................................... Standard salary level Total ...................................................................................................................................... Minimum wage only .......................................................................................................................................... Overtime pay only b .......................................................................................................................................... HCE compensation level Total ................................................................................................................................ Minimum wage only .......................................................................................................................................... Overtime pay only b .......................................................................................................................................... Annual $28,509 27,708 897 26,811 801 ........................ 801 $1,482,490 1,440,825 46,662 1,394,163 41,665 ........................ 41,665 a Due to both the minimum wage and overtime pay provisions and proposed changes in both the standard salary level and the HCE compensation level. b Estimated by subtracting the minimum wage transfer from the total transfer. 4. Potential Transfers Not Quantified There may be additional transfers attributable to this proposed rulemaking; however, the magnitude of these other transfers could not be quantified. These transfers are discussed in this section, as well as in section VII.D.vii, below. tkelley on DSK3SPTVN1PROD with PROPOSALS2 Converted to Hourly Status From Salaried Status Changing the EAP salary and HCE compensation level tests may impact whether a worker is classified as overtime ineligible or overtime eligible. Some evidence suggests that it is more costly for an employer to employ a salaried worker than an hourly worker. If true, employers may choose to accompany the change in exemption status with a change to the employee’s method of pay, from salary to an hourly basis, since there is no longer an incentive to classify the worker as salaried.124 Several employer stakeholders noted that salaried workers may perceive such a change as a loss of status. If the worker prefers to be salaried rather than hourly, then this change may impact the worker. The likelihood of this impact occurring depends on the costs to employers and benefits to employees of being salaried. Research has shown that salaried workers (who are not synonymous with exempt workers, but whose status is correlated with exempt status) are more likely than hourly workers to receive benefits such as paid vacation time and health insurance 125 and are more satisfied 124 There is no requirement that overtime eligible employees be paid on an hourly basis. Paying such employees on a salary basis is appropriate so long as the employee receives overtime pay for working more than 40 hours in the workweek. See 29 CFR 778.113. 125 Lambert, S. J. (2007). Making a Difference for Hourly Employees. In A. Booth, & A. C. Crouter, Work-Life Policies that Make a Real Difference for Individuals, Families, and Communities. Washington, DC: Urban Institute Press. VerDate Sep<11>2014 20:30 Jul 02, 2015 Jkt 235001 with their benefits,126 and that when employer demand for labor decreases hourly workers tend to see their hours cut before salaried workers, making earnings for hourly workers less predictable.127 However, this literature generally does not control for differences between salaried and hourly workers such as education, job title, or earnings; therefore, this correlation is not necessarily attributable to hourly status. Additionally, even if a worker’s salaried status is not officially changed, a salaried worker may effectively become an hourly worker if managers have to monitor hours more closely, so the worker may have less flexibility in work schedule.128 Reduced earnings for some workers Holding regular rate of pay and work hours constant, payment of an overtime premium will increase weekly earnings for workers who work overtime. However, as discussed previously, employers may try to mitigate cost increases by reducing the number of overtime hours worked, either by transferring these hours to other workers or monitoring hours more closely. Depending on how hours are adjusted, a specific worker may earn less pay after this proposed rulemaking. For example, assume an exempt worker is paid for overtime hours at his regular rate of pay (not paid the overtime premium but still acquires a benefit from each additional 126 Balkin, D. B., & Griffeth, R. W. (1993). The Determinants of Employee Benefits Satisfaction. Journal of Business and Psychology, 7(3), 323–339. 127 Lambert, S. J., & Henly, J. R. (2009). Scheduling in Hourly Jobs: Promising Practices for the Twenty-First Century Economy. The Mobility Agenda. Lambert, S. J. (2007). Making a Difference for Hourly Employees. 128 Swanberg, J. E., Pitt-Catsouphes, M., & Drescher-Burke, K. (2005). A Question of Justice: Disparities in Employees’ Access to Flexible Schedule Arrangements. Journal of Family Issues, 26 (6), 866–895. WorldatWork Research. (2009). Flexible Work Arrangements for Nonexempt Employees. WorldatWork Research. PO 00000 Frm 00063 Fmt 4701 Sfmt 4702 hour worked over 40 in a week). If the employer does not raise the worker’s salary to the new level, requiring the overtime premium may cause the employer to reduce the worker’s hours to 40 per week. If the worker’s regular rate of pay does not increase, the worker will earn less due to the lost hours of work. vi. Deadweight Loss Deadweight loss (DWL) occurs when a market operates at less than optimal equilibrium output. This typically results from an intervention that sets, in the case of a labor market, wages above their equilibrium level. While the higher wage results in transfers from employers to workers, it also causes a decrease in the total number of labor hours that are being purchased on the market. DWL is a function of the difference between the wage the employers were willing to pay for the hours lost and the wage workers were willing to take for those hours. In other words, DWL represents the total loss in economic surplus resulting from a ‘‘wedge’’ between the employer’s willingness to pay and the worker’s willingness to accept work arising from the proposed change. DWL may vary in magnitude depending on market parameters, but is typically small when wage changes are small or when labor supply and labor demand are relatively price (wage) inelastic. The DWL resulting from this proposed rulemaking was estimated based on the average decrease in hours worked and increase in hourly wages calculated in section VII.D.v. As the cost of labor rises due to the requirement to pay the overtime premium, the demand for overtime hours decreases, which results in fewer hours of overtime worked. To calculate the DWL, the following values must be estimated: • The increase in average hourly wages for affected EAP workers, • the decrease in average hours per worker, and E:\FR\FM\06JYP2.SGM 06JYP2 38578 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules • the number of affected EAP workers. Only 50 percent of Type 2 workers (those who work regular or predictable occasional overtime) and Type 3 workers are included in the DWL calculation because the other workers either do not work overtime (Type 1), continue to work the same number of overtime hours (Type 4), or their employers are unable to adjust their hourly wage because their overtime hours worked are unpredictable (the other 50 percent of Type 2 workers). As described above, after taking into account a variety of potential responses by employers, the Department estimated the average wage change for EAP affected workers whose hours change. Workers impacted by the change in the standard salary level are considered separately from workers impacted by the change in the HCE compensation level. For workers affected by the revised standard salary level, and who experience a change in hours, average wages (including overtime) will increase by $0.68 per hour. Average hours will fall by 0.40 per week. These changes result in an average DWL of $0.14 per week per Type 2 (the 50 percent who work foreseeable overtime) and Type 3 worker. An estimated 1.01 million workers will be eligible for the overtime premium on some of their hours worked after employer adjustments are taken into account. Multiplying the $0.14 per worker estimate by the number of affected workers results in a total DWL of $7.2 million in the first year of this proposed rulemaking attributable to the revised standard salary level (1.01 million workers in DWL analysis x $0.14 per worker per week x 52 weeks). For workers affected by the revised HCE compensation level and who experience a change in hours, the average hourly wage will increase by $2.14 and average hours worked will fall by 0.41 per week. This results in an average DWL of $0.44 per week for each of the estimated 12,000 workers affected by the compensation level who will see their hours fall. Multiplying this per worker estimate by the number of affected workers results in a DWL of $273,000 in the first year attributable to the HCE component of this proposed rulemaking (12,000 workers in DWL analysis x $0.44 per worker x 52 weeks). Thus, total DWL attributed to the proposed rulemaking is estimated to be $7.4 million in Year 1, which is small in comparison to the size of the costs and transfers associated with this proposal. TABLE 25—SUMMARY OF DEADWEIGHT LOSS COMPONENT VALUES Standard salary level Component Average hourly wages: Pre ................................................................................................................................................................ Post ............................................................................................................................................................... Change ......................................................................................................................................................... Average overtime hours: Pre ................................................................................................................................................................ Post ............................................................................................................................................................... Change ......................................................................................................................................................... Affected EAP workers .......................................................................................................................................... DWL: DWL per worker per week ........................................................................................................................... Total annual DWL (millions) ......................................................................................................................... HCE compensation level $15.01 $15.70 $0.68 $40.31 $42.45 $2.14 10.45 10.05 ¥0.40 1,010,433 13.14 12.73 ¥0.41 12,042 $0.14 $7.15 $0.44 $0.27 Note: DWL analysis is limited to Type 2 (50%) and Type 3 workers who experience hour adjustments. vii. Other Benefits, Costs and Transfers tkelley on DSK3SPTVN1PROD with PROPOSALS2 1. Benefits, Costs and Transfers Due to Strengthening Overtime Protection for Other Workers In addition to the 4.7 million affected EAP workers who will be newly eligible for overtime protection (absent employer response to increase the salary level to retain the exemption), overtime protection will be strengthened for an additional 10.0 million salaried workers who earn between the current salary level of $455 per week and the proposed salary level of $921 per week. These workers, who were previously vulnerable to misclassification through misapplication of the duties test, will now be automatically overtime protected because their salary falls below the new salary level and therefore they will not be subject to the duties test. These 10.0 million workers include: VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 • 6.3 million salaried white collar workers who are at particular risk of being misclassified because they currently pass the salary level test but do not satisfy the duties test; and • 3.7 million salaried workers in blue collar occupations whose overtime protection will be strengthened because their salary will fall below the proposed salary threshold.129 (Identification of blue collar workers is explained in section VII.B.iv). Although these workers are currently entitled to minimum wage and overtime protection, their protection is better assured with the proposed salary level. The salary level test is considered a bright-line test because it is clear to employers and employees alike whether or not a worker passes. The duties test (which is the reason employers cannot claim the EAP exemption for the above workers in this group may be overtime ineligible due to another non-EAP exemption. PO 00000 129 Some Frm 00064 Fmt 4701 Sfmt 4702 workers) is more discretionary and therefore harder to apply. An outdated salary level reduces the effectiveness of this bright-line test. At the proposed salary level, the number of overtimeeligible white collar workers earning at or above the salary level will decrease by 6 million, and an estimated 806,562 (13.5 percent) of these workers are currently misclassified as exempt. Therefore, increasing the salary level is expected to result in less worker misclassification. Employers will be able to more readily determine their legal obligations and comply with the law, thus leading to benefits, costs and transfers that are qualitatively similar to the impacts discussed elsewhere in this analysis but the magnitudes of which have not been estimated. 2. Cost Savings: Reduction in Litigation Reducing the number of white collar employees for whom a duties analysis must be performed in order to E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules determine entitlement to overtime will also reduce litigation related to the EAP exemption. As previously discussed, employer uncertainty about which workers should be classified as EAP exempt has contributed to a sharp increase in FLSA lawsuits over the past decade. Much of this litigation has involved whether employees who satisfy the salary level test also meet the duties test for exemption. See, e.g, Soehnle v. Hess Corp., 399 F. App’x 749 (3d Cir. 2010) (gas station manager earning approximately $654 per week satisfied duties test for executive employee); Morgan v. Family Dollar Stores, Inc., 551 F.3d 1233 (11th Cir. 2008) (store managers earning an average weekly salary of up to $706 did not satisfy duties test for executive exemption). Setting an appropriate salary level for the standard duties test and maintaining the salary level with automatic updates will restore the test’s effectiveness as a bright-line method for determining exempt status, and in turn decrease the litigation risk created when employers must apply the duties test to employees who generally are not performing bona fide EAP work. This will eliminate legal challenges regarding the duties test involving employees earning between the current salary level ($455) and the proposed level ($921). See, e.g., Little v. Belle Tire Distribs., Inc., 588 F. App’x 424 (6th Cir. 2014) (applicability of administrative or executive exemption to tire store assistant manager earning $1,100 semi-monthly); Taylor v. Autozone, Inc., 572 F. App’x 515 (9th Cir. 2014) (applicability of executive exemption to store managers earning as little as $800 per week); Diaz v. Team Oney, Inc., 291 F. App’x. 947 (11th Cir. 2008) (applicability of executive duties test to pizza restaurant assistant manager earning $525 per week). Setting the salary level test at the proposed level will alleviate the need for employers to apply the duties test in these types of cases, which is expected to result in decreased litigation as employers will be able to determine employee exemption status through application of the salary level test without the need to perform a duties analysis. See Weiss Report at 8 (The salary tests ‘‘have amply proved their effectiveness in preventing the misclassification by employers of obviously nonexempt employees, thus tending to reduce litigation. They have simplified enforcement by providing a ready method of screening out the obviously nonexempt employees, making an analysis of duties in such cases unnecessary.’’) VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 3. Benefits and Costs: Reduction in Uncertainty about Future Overtime Hours and Pay The proposed rule may have an impact on employees who are not currently working any overtime, but will now be entitled to minimum wage and overtime pay protections. These workers may face a lower risk of being asked to work overtime in the future, because they are now entitled to an overtime premium, which could reduce their uncertainty and improve their welfare if they do not desire to work overtime. Additionally, if they are asked to work overtime, they are compensated for the inconvenience with an overtime premium. Economic theory suggests that workers tend to assign monetary values to risk or undesirable job characteristics, as evidenced by the presence of compensating wage differentials for undesirable jobs, relative to other jobs the worker can perform in the marketplace. To the extent a compensating wage differential exists, compensation may decrease with the reduction in uncertainty.130 For this reason, overall compensation would be expected to decrease for workers whose uncertainty decreases. Employees who prefer the reduced uncertainty to the wage premium would experience a net benefit of the rule, and employees who prefer the wage premium to the reduced uncertainty would experience a net harm as a result of the rule. The Department believes that attempting to model the net monetary value of reduced uncertainty is not feasible due to its heavy reliance on data that are not readily available, and the potentially questionable nature of the resulting estimates.131 4. Benefits and Costs: Work-Life Balance Due to the increase in marginal cost for overtime hours, employers will demand fewer hours from some of the workers affected by this rule.132 The estimated transfer payment does not take into account the benefit to these workers of working fewer hours in exchange for more (or equal) pay. 130 In this case, the size of the compensating wage differential is a function of the likelihood of working overtime and the amount of overtime worked. If the probability of working overtime is small then the wage differential may not exist. 131 For a discussion of compensating wage differentials, see Gronberg, T. J., & Reed, W. R. (1994). Estimating Workers’ Marginal Willingness to Pay for Job Attributes using Duration Data. Journal of Human Resources, 29(3), 911–931. 132 The Department recognizes that not all workers would prefer to work fewer hours and thus some of these workers might experience an adverse impact. The Department has no basis for estimating this potential impact. PO 00000 Frm 00065 Fmt 4701 Sfmt 4702 38579 Therefore, an additional benefit of this proposed rulemaking is the increase in time off for affected EAP workers. On average, affected EAP workers were estimated to work 5.2 minutes less per week after the proposed rulemaking. The effect is much more pronounced when limited to just those workers whose hours are adjusted (50 percent of Type 2 and all Type 3 workers); they would on average work 23.9 minutes less per week after the proposed rulemaking. The additional time off may help these workers better balance worklife commitments, thus potentially making them better off. Empirical evidence shows that workers in the United States typically work more than workers in other comparatively wealthy countries.133 Although estimates of the actual level of overwork vary considerably, executive, administrative, and professional occupations have the highest percentage of workers who would prefer to work fewer hours compared to other occupational categories.134 Therefore, the Department believes that the proposed rule may result in increased time off for a group of workers who may prefer such an outcome. However, the empirical evidence does not allow us to estimate how many workers would prefer fewer hours or how much workers value this additional time off so it is difficult to monetize the benefit they may receive. However, if we use affected workers’ average wage to approximate the value they place on this time ($15.31), then the benefit of this additional time off would total $6.2 million weekly (0.40 hours x 1.02 million workers × $15.31). This would result in an estimated total benefit of $324.4 million per year. This is likely an overestimate to the extent that not all workers would prefer to work fewer hours and thus some of these workers might experience an adverse impact. In addition, the estimated work loss represents an average over all affected workers, and some workers may experience a larger reduction in hours.135 133 For more information, see OECD series, average annual hours actually worked per worker, available at: https://stats.oecd.org/ index.aspx?DataSetCode=ANHRS. 134 Hamermesh, D.S., Kawaguchi, D., Lee, J. (2014). Does Labor Legislation Benefit Workers? Well-Being after an Hours Reduction. IZA DP No. 8077. Golden, L., & Gebreselassie, T. (2007). Overemployment mismatches: the preference for fewer work hours. Monthly Labor Review, 130(4), 18–37. Hamermesh, D.S. (2014). ‘‘Not enough time?’’ American Economist, 59(2). 135 It is possible that some employers may choose to eliminate all overtime for affected workers and E:\FR\FM\06JYP2.SGM Continued 06JYP2 38580 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 5. Additional Benefits and Costs not Quantified The largest benefit to workers from the proposed rule is the transfer of income from employers (as discussed in the transfer section of the analysis); but, to the extent that the benefits to workers outweigh the costs to employers, there may be a societal welfare increase due to this transfer. The channels through which societal welfare may increase and other secondary benefits may occur are discussed below and include increased productivity and improved worker outcomes, such as improved health. The discussion references the potential magnitude of these benefits where possible; however, due to data limitations and mixed evidence on the significance of such effects, the Department was not able to quantify the size of these potential benefits. Health Working long hours is correlated with an increased risk of injury or health problems.136 Therefore, by reducing overtime hours, some affected EAP workers’ health may improve. This would benefit the worker’s welfare, their family’s welfare, and society since fewer resources would need to be spent on health. Health has also been shown to be highly correlated with productivity.137 These beneficial effects, and how they compare with other potential responses by employers, especially regarding workers who pass the duties test and whose salaries are either already above the proposed threshold or would be adjusted to be so, have not been quantified. tkelley on DSK3SPTVN1PROD with PROPOSALS2 Increased productivity This proposed rule is expected to increase the marginal cost of some workers’ labor, predominately due to the overtime pay requirement since almost all affected EAP workers already earn the federal minimum wage. hire additional workers or spread the work to existing employees to replace the lost hours. The potential for this adjustment is uncertain, and the Department has found no studies that estimate the potential magnitude of this effect. In addition, an employer may be limited in his or her ability to make such adjustments; many affected employees work only a few hours of overtime each week; affected employees’ tasks may not be easily divisible; and hiring new workers and/or managing different work flows will impose additional costs on the employer that will offset the savings from avoiding paying the overtime premium. 136 Keller, S. M. (2009). Effects of Extended Work Shifts and Shift Work on Patient Safety, Productivity, and Employee Health. AAOHN Journal, 57(12), 497–502. 137 Loeppke, R., Taitel, M., Richling, D., Parry, T., Kessler, R., Hymel, P., et al. (2007). Health and Productivity as a Business Strategy. Journal of Occupational and Environmental Medicine, 49(7), 712–721. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 However, some of the cost to employers of paying the overtime premium may be offset by increased worker productivity. This may occur through a variety of channels, including: increased marginal productivity as fewer hours are worked, reduction in turnover, efficiency wages, and worker health. Reduction in turnover: Research demonstrates a positive correlation between earnings and employee turnover: as earnings increase, employee turnover decreases.138 139 Reducing turnover may increase productivity, at least partially because new employees have less firm-specific capital (i.e., skills and knowledge that have productive value in only one particular company) and thus are less productive and require additional supervision and training.140 In short, replacing experienced workers with new workers decreases productivity, and avoiding that will increase productivity. Reduced turnover should also reduce firms’ hiring and training costs. As a result, even though marginal labor costs rise, they may rise by less than the amount of the wage change because the higher wages may be offset by lower turnover rates, increased productivity, and reduced hiring costs for firms. It is difficult to estimate the impact of reduced turnover on worker productivity and firm hiring costs. The potential reduction in turnover is a function of several variables: the current wage, hours worked, turnover rate, industry, and occupation. Additionally, estimates of the cost of replacing a worker who quits vary significantly. Therefore, quantifying the potential benefit associated with a decrease in turnover attributed to this proposed rule is difficult. Efficiency wages: By increasing earnings this proposed rulemaking may increase a worker’s productivity by incentivizing the worker to work harder. Thus the additional cost to firms may be partially offset by higher productivity. 138 Howes, Candace, (2005). Living Wages and Retention of Homecare Workers in San Francisco. Industrial Relations, 44(1), 139–163. Dube, A., Lester, T.W., & Reich, M. (2014). Minimum Wage Shocks, Employment Flows and Labor Market Frictions. IRLE Working Paper #149–13. 139 Note that this literature tends to focus on changes in earnings for a specific sector or subset of the labor force. The impact on turnover when earnings increase across sectors (as would be the case with this regulation) may be smaller. 140 Argote, L., Insko, C. A., Yovetich, N., & Romero, A. A. (1995). Group Learning Curves: The Effects of Turnover and Task Complexity on Group Performance. Journal of Applied Social Psychology, 25(6), 512–529. Shaw, J. D. (2011). Turnover Rates and Organizational Performance: Review, Critique, and Research Agenda. Organizational Psychology Review, 1(3), 187–213. PO 00000 Frm 00066 Fmt 4701 Sfmt 4702 In particular, the estimated managerial costs associated with greater monitoring effort may be offset due to this effect. A strand of economic research, commonly referred to as ‘‘efficiency wages,’’ considers how an increase in wages may be met with greater productivity.141 However, this literature tends to focus on firms voluntarily paying higher wages, and thus distinguishing themselves from other firms. Since this rulemaking mandates wage increases, extrapolating from efficiency wage theory may not provide a reliable guide to the likely effects of the rule. Conversely, there are channels through which mandating overtime pay may reduce productivity. For example, some overtime hours may be spread to other workers. If the work requires significant project-specific knowledge or skills, then the new worker receiving these transferred hours may be less productive than the first worker, especially if there is a steep learning curve. Additionally, having an additional worker versed in the project may be beneficial to the firm if the first worker leaves the firm or is temporarily away (e.g., sick) or by providing benefits of teamwork (e.g., facilitating information exchange). 6. Transfers: Reduction in Social Assistance Expenditures The transfer of income resulting from this proposed rule may result in reduced need for social assistance (and by extension reduced social assistance expenditures by the government). A worker earning the current salary level of $455 per week earns $23,660 annually. If this worker resides in a family of four and is the sole earner, then the family will be considered impoverished. This makes the family eligible for many social assistance programs. Thus, transferring income to these workers may reduce eligibility for government social assistance programs and government expenditures. A social welfare improvement will result from the reduced resource needs for making those transfer payments. Benefits for which currently EAP exempt workers may qualify include Medicaid, the Supplemental Nutrition Assistance Program (SNAP), the Temporary Assistance for Needy Families (TANF) program, the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), and school breakfasts and lunches. Quantifying the impact of this proposed rulemaking on government expenditures 141 Akerlof, G. A. (1982). Labor Contracts as Partial Gift Exchange. The Quarterly Journal of Economics, 97(4), 543–569. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules is complex and thus not estimated here. In order to conduct such an analysis, the Department would need estimates of the transfer per worker, his or her current income level, other sources of family income, number of family members, state of residence, and receipt of aid. viii. Bounds on Transfer Payments Because the Department cannot predict the precise reaction of employers to the proposed rule, the Department also calculated bounds to the size of the estimated transfers from employers to workers using a variety of assumptions. Since transfer payments are the largest component of this proposed rulemaking the scenarios considered here are bounds around the transfer estimate. Based on the assumptions made, these bounds do not generate bounded estimates for costs or DWL. The maximum potential upper limit occurs with the assumption that the demand for labor is completely inelastic, and therefore neither the implicit regular hourly rate of pay nor hours worked adjust in response to the changes in the EAP standard salary level and HCE annual compensation level test. Employers then pay workers one and a half times their current implicit hourly rate of pay for all overtime hours currently worked (i.e., the full overtime premium). The minimum potential lower bound occurs when wages adjust completely and weekly earnings are unchanged as predicted by the employment contract model. The Department believes that both the maximum upper bound scenario and the minimum lower bound scenario are unrealistic; therefore, we constructed more credible bounds. For a more realistic upper bound on transfer payments, the Department assumed that all occasional overtime workers and half of regular overtime workers would receive the full overtime premium, as it was computed in the maximum upper bound methodology (i.e., such workers would work the same number of hours but be paid 1.5 times their implicit initial hourly wage for all overtime hours). Conversely, in the preferred model we assumed that only 50 percent of occasional overtime workers and no regular overtime workers would receive the full overtime premium. It was assumed that employers could not instantaneously 38581 adjust earnings for the 50 percent of affected EAP workers who regularly work overtime. However, for the other half of regular overtime workers, the Department assumed they would have their implicit hourly wage adjusted as predicted by the partial employment contract model (wage rates fall and hours are reduced but total earnings continue to increase, as in the preferred method). Table 26 summarizes the assumptions described above. The plausible lower transfer bound also depends on whether employees work regular overtime or occasional overtime. For those who regularly work overtime hours and half of those who work occasional overtime, the Department assumes the employees’ wages will fully adjust as predicted by the employment contract model, whenever possible (in the preferred method their wages adjust based on the partial employment contract model).142 For the other half of employees with occasional overtime hours, the lower bound assumes they will be paid one and one-half times their implicit hourly wage for overtime hours worked (full overtime premium). TABLE 26—SUMMARY OF THE ASSUMPTIONS USED TO CALCULATE THE LOWER ESTIMATE, PREFERRED ESTIMATE, AND UPPER ESTIMATE OF TRANSFERS Lower transfer estimate Preferred estimate Upper transfer estimate Occasional Overtime Workers (Type 2) 50% full employment contract model adj .......... 50% partial employment contract model adj .... 50% full overtime premium ................................ 100% full overtime premium. 50% full overtime premium. Regular Overtime Workers (Type 3) 100% full employment contract model adj ........ 100% partial employment contract model adj .. 50% partial employment contract model adj 50% full overtime premium. tkelley on DSK3SPTVN1PROD with PROPOSALS2 Legend: * Full overtime premium: Regular rate of pay equals the implicit hourly wage prior to the proposed regulation (with no adjustments); workers are paid 1.5 times this base wage for the same number of overtime hours worked prior to the regulation (assuming the worker was paid the minimum wage, otherwise the wage increases to the minimum wage and overtime hours may decrease). * Full employment contract model adjustment: Base wages are set at the higher of: (1) a rate such that total earnings and hours remain the same before and after the proposed regulation; thus the base wage falls, and workers are paid 1.5 times the new base wage for overtime hours (the employment contract model) or (2) the minimum wage. * Partial employment contract model adjustment: Regular rates of pay are partially adjusted to the wage implied by the employment contract model. The resulting regular rate of pay is the midpoint of: (1) a base wage that adjusts 40 percent of the way to the employment contract model wage level, assuming no overtime premium was initially paid and (2) a base wage that adjusts 80 percent of the way to the employment contract model wage level, assuming the workers initially received a 28 percent premium for overtime hours worked. The cost and transfer payment estimates associated with the bounds are presented in Table 27. Regulatory familiarization costs and adjustment costs do not vary across the scenarios. These employer costs are a function of the number of affected firms or affected workers, human resource personnel hourly wages, and time estimates. None of these vary based on the assumptions made above. Conversely, managerial costs are lower under these alternative employer response assumptions because fewer workers’ hours are adjusted by employers and thus managerial costs, which depend on the number of workers whose hours change, will be smaller. Managerial costs vary according to employers’ response to the proposed rule. Depending on how employers adjust the implicit regular hourly wage, the estimated transfer may range from $543.7 million to $2,851.2 million, with the preferred estimate equal to $1,482.5 million. The DWL associated with the preferred estimate is $7.4 million. The 142 The straight-time wage adjusts to a level that keeps weekly earnings constant when overtime hours are paid at 1.5 times the straight-time wage. In cases where adjusting the straight-time results in a wage less than the minimum wage, the straighttime wage is set to the minimum wage. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00067 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38582 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules upper transfer estimate of DWL is smaller than the preferred estimate because the assumptions made for this upper bound scenario result in fewer hours lost. For the lower transfer estimate DWL was estimated to be less than $400,000; for the upper transfer estimate scenario the DWL was estimated to be $3.7 million. TABLE 27—BOUNDS ON ANNUAL COST AND TRANSFER PAYMENT ESTIMATES, 2013 (MILLIONS) Lower transfer estimate a Cost/transfer Direct employer costs. Reg. familiarization ............................................................................................................... Adjustment costs .................................................................................................................. Managerial costs .................................................................................................................. Total direct employer costs ......................................................................................................... Transfers a .................................................................................................................................... DWL ............................................................................................................................................. $254.5 160.1 1.7 416.3 543.7 0.4 Preferred estimate $254.5 160.1 178.1 592.7 1,482.5 7.4 Upper transfer estimate $254.5 160.1 82.8 497.4 2,851.2 3.7 Note: Pooled data for 2011–2013. a Due to both the minimum wage and overtime pay provisions and changes in both the standard salary level and the HCE compensation level. tkelley on DSK3SPTVN1PROD with PROPOSALS2 ix. Regulatory Alternatives The Department proposes in this NPRM to update the standard salary level to the 40th percentile of weekly earnings for all full-time salaried workers ($921 per week). The Department considered a range of alternatives before deciding on this level. Seven alternatives are presented here. Two of these (alternatives 1 and 7) inflate the value of earlier salary levels to take into account inflation in the intervening years. Three others (alternatives 2, 3, and 5) adapt the 2004 method or the Kantor method to set the salary level. Alternatives 4 and 6 set the salary level to the median weekly salary for either all full-time hourly and salaried workers or full-time salaried workers, respectively. Table 28 presents the alternative salary levels considered and the number of workers estimated to be affected under these salary levels. Alternative 1 increases the 2004 salary level of $455 per week by the rate of inflation between 2004 and 2013 as measured by the CPI–U. This results in a salary level of $561 per week. At this salary level 576,000 workers would be affected in Year 1, imposing direct adjustment and managerial costs of $36.1 million, transferring $127.9 million in earnings from employers to employees, and resulting in DWL of $0.5 million. Alternative 2 sets the salary level using the 2004 method resulting in a salary level of $577 per week. At this salary level 734,000 workers would be affected, Year 1 adjustment and managerial costs would equal $44.5 million, with transfers of $151.5 VerDate Sep<11>2014 20:32 Jul 02, 2015 Jkt 235001 million, while DWL would equal $0.7 million. Alternative 3 sets the salary level using the Kantor method. This results in a salary level of $657 per week. At this salary level, 1.4 million workers are affected, Year 1 adjustment and managerial costs are $91.8 million, Year 1 transfers are $318.6 million, and Year 1 DWL is $1.7 million. Alternative 4 sets the salary level equal to the 50th percentile, or median, of weekly earnings for full-time hourly and salaried workers. This results in a salary level of $776 per week. At this salary level, 2.7 million workers would be affected in Year 1, employer costs would total $179.8 million with transfers of $686.6 million, and DWL would be $3.6 million. Alternative 5 is based on the Kantor method but, whereas alternative 3 generates the salary level associated with the long duties test, alternative 5 generates a level more appropriate to the short duties test (as explained in section VII.C) and results in a salary level of $979 per week. At this salary level, 5.6 million workers would be affected in Year 1, with adjustment and managerial costs of $404.2 million, transfers of $1.8 billion, and DWL equal to $10.3 million. As previously noted, while this alternative uses the average difference between the Kantor long and short tests, the ratio of the short to long salary tests ranged between approximately 130 percent and 180, which would result in a salary between $854 and $1,183. Alternative 6 sets the standard salary equal to the 50th percentile, or median, of weekly earnings for full-time salaried workers. This approach is similar to the PO 00000 Frm 00068 Fmt 4701 Sfmt 4702 proposed method but uses a higher weekly earnings percentile: 50th instead of the 40th. This results in a salary level of $1,065 per week. At this salary level, 6.9 million workers would be affected in Year 1, employer costs would total $522.1 million with transfers of $2.5 billion, and DWL would be $14.8 million. Alternative 7 increases the 1975 short test salary level of $250 per week by the rate of inflation from 1975 to 2013. This results in a salary level of $1,083 per week. At this salary level, 7.1 million workers would be affected in Year 1, employer costs would total $543.0 million with transfers of $2.7 billion, and DWL would be $15.5 million. The Department also examined alternatives to the proposed HCE compensation level. HCE alternative 1 left the current $100,000 annual compensation level unchanged. Therefore, no employer costs, transfers, or DWL are associated with this alternative. HCE alternative 2 sets the HCE annual compensation level at $150,000 per year. This compensation level would affect 52,000 workers in Year 1 (compared to 36,000 at the proposed compensation level), impose adjustment and managerial costs on employers of $5.5 million, transfer $71.2 million in earnings from employers to employees, and generate $600,000 in DWL. Because regulatory familiarization costs cannot realistically be differentiated into those relevant to the standard salary level, and those relevant to the HCE compensation level, the Department does not separately estimate those costs for the HCE alternatives. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38583 TABLE 28—PROPOSED STANDARD SALARY AND HCE COMPENSATION LEVELS AND ALTERNATIVES, AFFECTED EAP WORKERS, COSTS, AND TRANSFERS, 2013 Alternative Salary level Year 1 impacts (Millions) Affected EAP workers (1,000s) Adj. & managerial costs a Transfers DWL b Standard Salary Level (Weekly) Proposed .............................................................................. Alt. #1: Inflate 2004 levels ................................................... Alt. #2: 2004 method ........................................................... Alt. #3: Kantor method ......................................................... Alt. #4: Median full-time hourly and salaried workers ......... Alt. #5: Kantor short test ...................................................... Alt. #6: Median full-time salaried ......................................... Alt. #7: Inflate 1975 short test level ..................................... $921 561 577 657 776 979 1,065 1,083 4,646 576 734 1,390 2,704 5,632 6,855 7,128 $334.8 36.1 44.5 91.8 179.8 404.2 522.1 543.0 $1,440.8 127.9 151.5 318.6 686.6 1,821.3 2,525.8 2,666.1 $7.2 0.5 0.7 1.7 3.6 10.3 14.8 15.5 $3.3 ........................ 5.5 $41.7 ........................ 71.2 $0.0 ........................ 0.6 HCE Compensation Level (Annually) Proposed .............................................................................. Alt. #1: No change ............................................................... Alt. #2: 2004 percentile ........................................................ $122,148 100,000 150,000 36 0 52 Note: Pooled data for 2011–2013. a Regulatory familiarization costs are excluded because they are a one-time cost that do not vary based on the proposed salary levels. b DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. Since the transfer associated with the minimum wage is negligible compared to the transfer associated with overtime pay, the vast majority of this cost is attributed to the overtime pay provision. x. Projections tkelley on DSK3SPTVN1PROD with PROPOSALS2 1. Methodology In addition to estimating Year 1 costs and transfers, the Department projected costs and transfers forward for ten years. To project costs and transfers, the Department used several pieces of data, specifically the median wage growth rate and the employment growth rate. These calculations are described below, after which the ten-year projected costs and transfers are presented. The projections presented in this section assume the proposed salary level remains constant over ten years. Thus, the number and percent of affected EAP workers decline over time as real earnings increase.143 The section on automatic updating of the salary level will present the estimated ten-year impacts based on how real earnings change relative to an automatically updated salary level because the selection of the salary level is conceptually separate from the decision to update (and how to update) the salary level. Thus, the costs and impacts of each are considered and presented separately. In order to identify workers whose projected salaries fall between the current salary level ($455 per week) and the proposed salary level based on the 40th earnings percentile ($921 per week), a wage growth rate must be 143 As described in the following paragraphs, the Department used historical wage growth rates to project wage growth rates. VerDate Sep<11>2014 20:32 Jul 02, 2015 Jkt 235001 applied to current earnings. The Department applied an annual real growth rate based on the average annual growth rate in median wages from 2005 to 2012.144 The wage growth rate is calculated as the geometric growth rate in median wages using the historical CPS MORG data for exempt workers by occupation-industry categories. The geometric growth rate is the constant annual growth rate that when compounded (applied to the first year’s wage, then to the resulting second year’s wage, etc.) yields the last historical year’s wage. This method only depends on the value of the wage in the first available year and the last available year, and may be a flawed measure if either or both of those years were atypical; however, in this instance these values seem typical. An alternative method would be to use the time series of median wage data to estimate the linear trend in the values and continue this to project future median wages. This method may be preferred if either or both of the endpoint years are outliers, since the trend will be less influenced by them. The linear trend may be flawed if there are outliers in the interim years (because 144 In order to maximize the number of observations used in calculating the median wage for each occupation-industry group, three years of data were pooled for each of the endpoint years. Specifically, data from 2004, 2005, and 2006 (converted to 2005 dollars) were used to calculate the 2005 median wage and data from 2011, 2012, and 2013 (converted to 2012 dollars) were used to calculate the 2012 median wage. PO 00000 Frm 00069 Fmt 4701 Sfmt 4702 these have no impact on the geometric mean but will influence the estimate of a linear trend). The Department chose to use the geometric mean because individual year fluctuations are difficult to predict and applying the geometric growth rate to each year provides a better estimate of the long-term growth in wages. Using this method is also consistent with the estimation of the employment growth rate as described below. The geometric wage growth rate was also calculated from the BLS’ Occupational Employment Statistics (OES) and used as a validity check.145 Additionally, in occupation-industry categories where the CPS MORG data had an insufficient number of observations to reliably calculate median wages, the Department used the growth rate in median wages calculated from the OES data.146 Any remaining occupation-industry combinations without estimated median growth rates were assigned the median of the growth rates in median wages from the CPS MORG data for EAP exempt workers. 145 The difference between the OES and CPS growth measures averaged ¥0.0002 percent, but ranged from ¥7.2 to 5.8 percent, depending on the occupation-industry category. The CPS growth estimates were used as the primary source because the sample could be restricted to EAP exempt workers (the relevant population). 146 To lessen small sample bias in the estimation of the median growth rate, this rate was only calculated using CPS MORG data when these data contained at least 10 observations in each time period. E:\FR\FM\06JYP2.SGM 06JYP2 38584 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 The Department calculated projected earnings for each worker in the sample by applying the annual projected wage growth rate to current earnings for each projected year. In each projected year, affected EAP workers were identified as those who are exempt in the current year (prior to the rule change) but have projected earnings in the projected year that are less than the proposed salary level. The employment growth rate is the geometric annual growth rate based on the ten-year employment projection from BLS’ National Employment Matrix (NEM) within an occupation-industry category. This is the constant annual growth rate that when compounded yields the NEM ten-year projection. An alternative method is to spread the total change in the level of employment over the ten years evenly across years (constant change in the number employed). The Department believes that on average employment is more likely to grow at a constant percentage rate rather than by a constant level (a decreasing percentage rate). To account for population growth, the Department applied the growth rates to the sample weights of the workers. This is because the Department cannot introduce new observations to the CPS MORG data to represent the newly employed. Affected EAP workers may experience a reduction in hours since the wage they receive for overtime hours is higher after the proposed rulemaking. The reduction in hours is calculated as described in section VII.D.v. The only difference is that for projections the long-run elasticity of labor demand is used instead of the short-run elasticity. The Department used a long-run elasticity of ¥0.4.147 2. Estimated Projections Projected costs and transfers both depend on the projected number of affected EAP workers. The Department estimated that in Year 1 4.7 million EAP workers will be affected, with about 36,000 of these attributable to the revised HCE compensation level. In Year 10, if the salary levels are not updated, the number of affected EAP workers was estimated to equal 2.7 million, with fewer than 8,000 attributed to the HCE exemption. The projected number of affected EAP workers accounts for projected employment growth by increasing the number of workers represented by the affected EAP workers (i.e., increasing 147 This elasticity estimate is based on the Department’s analysis of the following paper: Lichter, A., Peichl, A. & Siegloch, A. (2014). The Own-Wage Elasticity of Labor Demand: A MetaRegression Analysis. IZA DP No. 7958. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 sampling weights). However, with no additional changes in the salary level and most workers experiencing positive wage growth, workers affected in Year 1 become less likely to still be affected in each future year. That is, some of these workers return to exempt status over time as their growing salaries eventually exceed the proposed standard salary level of $921 per week. The net impact is a decrease in the number of affected EAP workers in each subsequent year. The projected number of affected workers only includes workers who were originally determined to be exempt in 2013. However, additional workers may be affected in future years who were not EAP exempt in the base year but would have become exempt in the absence of this proposed rule. For example, a worker may earn less than $455 in 2013 but at least $455 (and less than the proposed salary level) in subsequent years; such a worker would not be counted as an affected worker in the projections above. In the absence of this proposed rule he or she would likely have become exempt at some point in the 10-year projections period; however, as a result of the proposed rule, this worker remains nonexempt, and is thus affected by the proposed rule. Therefore, the Department estimated the number of workers who were: Paid on a salary basis, pass the duties test, and earn less than $455 per week in 2013, but are projected to earn at least $455 (but less than the proposed salary level) per week at some point in the following nine years. The Department found that in Year 10, an additional 398,000 workers meet these criteria and therefore are also affected workers. Similarly, the Department estimated the number of workers who are paid on a salary basis, meet the HCE duties test, and currently earn less than $100,000 annually but are projected to earn more than $100,000 per year at some point in the following nine years. The Department estimated that, in Year 10, an additional 115,000 workers meet these criteria and therefore are also affected workers. The Department did not estimate costs, transfers, or DWL for these workers because it would be necessary to make additional assumptions such as how employers respond by adjusting workers’ wages and hours. The Department quantified three types of direct employer costs in the ten-year projections: (1) Regulatory familiarization costs; (2) adjustment costs; and (3) managerial costs. Regulatory familiarization costs are onetime costs and only occur in Year 1. Although start-up firms must still PO 00000 Frm 00070 Fmt 4701 Sfmt 4702 become familiar with the FLSA following Year 1, the difference between the time necessary for familiarization with the current part 541 exemptions and those exemptions as modified by the proposed rule is essentially zero. Therefore, projected regulatory familiarization costs over the next nine years are zero. Similarly, adjustment costs are only incurred when workers’ status changes from exempt to nonexempt, so adjustment costs are incurred predominately in Year 1 (some very minor adjustment costs may exist in projected years because some workers’ earnings decrease and thus these workers may transition from exempt to nonexempt). However, managerial costs recur for all affected EAP workers whose hours are adjusted and were projected through Year 10. The Department estimated that Year 1 managerial costs would be $178.1 million (section VII.D.iv.4.); by Year 10 these costs would fall to $93.1 million (Table 29). Over 97 percent of this amount ($176.0 million) in Year 1, and roughly 99 percent ($92.6 million) in Year 10 is attributable to the revised standard salary level. The projected reduction in managerial costs over the years is due to the reduction in the number of affected EAP workers over time as workers’ earnings increase relative to the constant salary and compensation levels. The Department also projected two transfers associated with workers affected by the proposed regulation: (1) Transfers to workers from employers due to the minimum wage provision and (2) transfers to workers from employers due to the overtime pay provision. Transfers to workers from employers due to the minimum wage provision, estimated to be $46.7 million in Year 1, are projected to decline to $9.9 million in Year 10 as increased earnings over time move workers’ regular rate of pay above the minimum wage.148 Transfers to workers from employers due to the overtime pay provision decrease from $1,435.8 million in Year 1 to $490.2 million in Year 10. Workers affected by the revised standard salary level account for 97 percent of overtime transfers in Year 1, and 99 percent in Year 10. Again, the decrease in transfers is primarily due to the reduction in the number of affected workers over time. Table 29 also summarizes average annualized costs and transfers over the ten-year projection period, using 3 148 In states with higher minimum wages, then effective state minimum wages were used in 2013 and 2014 and minimum wages on December 31, 2014 were used for projected years. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules attributed to the revised standard salary level. The remaining $138.9 million in average annualized direct costs were accounted for by managerial costs, of which 99 percent were associated with the revised standard salary level. The average annualized value of total transfers was estimated to equal $872.9 million. The largest component of this was the average annualized transfer from employers to workers due to overtime pay, which was $843.6 million per year, while average annualized transfers due to the minimum wage totaled $29.3 million per year. None of percent and 7 percent real discount rates. The Department estimated that total direct employer costs have an average annualized value of $194.2 million per year over ten years when using a 7 percent real discount rate. Of this total, average annualized regulatory familiarization costs were estimated to be $33.9 million; the Department does not apportion these out between the revised standard salary and HCE annual compensation levels. Average annualized adjustment costs were estimated to be $21.5 million; roughly 99 percent of adjustment costs were 38585 the transfer associated with the minimum wage was attributed to the revised HCE compensation level. Although composing less than one percent of affected workers, those receiving overtime due to the revised HCE compensation level account for 2.2 percent of total average annualized transfers ($19.5 of $872.9 million) because of their high implicit regular hourly rate of pay. The remaining $853.4 million in transfers accrue to those affected by the revised standard salary level. TABLE 29—PROJECTED COSTS AND TRANSFERS WITHOUT AUTOMATIC UPDATING, STANDARD AND HCE SALARY LEVELS Year (Year #) Affected EAP workers (Millions) Costs Transfers DWL a Reg. Fam. Adjustment Managerial Due to MW Due to OT (Millions 2013$) Year 2013 (1) 2014 (2) 2015 (3) 2016 (4) 2017 (5) 2018 (6) 2019 (7) 2020 (8) 2021 (9) 2022 (10) Average Annualized 3% real rate 7% real rate 4.7 4.5 4.2 4.0 3.8 3.6 3.4 3.1 2.9 2.7 $254.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 $160.1 1.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 $178.1 169.0 155.8 146.1 137.5 128.5 118.4 108.9 100.6 93.1 $46.7 44.0 39.5 33.0 27.4 22.6 18.3 14.9 11.8 9.9 $1,435.8 1,017.1 923.9 843.1 771.4 702.0 640.1 582.0 539.2 490.2 $7.4 9.8 8.9 8.1 7.5 6.8 6.0 5.3 4.9 4.3 .......................... .......................... 29.0 33.9 18.4 21.5 135.9 138.9 27.9 29.3 815.7 843.6 7.0 7.2 a DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. Since the transfer associated with the minimum wage is negligible compared to the transfer associated with overtime pay, the vast majority of this cost is attributed to the overtime pay provision. tkelley on DSK3SPTVN1PROD with PROPOSALS2 The cost to society of lower employment expressed as DWL was estimated to be $7.4 million in Year 1. After year 2, DWL falls over time; in Year 10 it is projected to equal $4.3 million. DWL increases sharply between Year 1 and Year 2 because the Department assumes the market has had time to fully adjust to the revised standard salary and HCE annual compensation levels by Year 2. In Year 1 employers may not be able to fully adjust wages and hours in response to the rulemaking, so the Department used a short run wage elasticity of labor demand to reflect this constrained VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 response; in Year 2 employers have sufficient time to fully adjust, and a long run wage elasticity is used. Therefore, the decrease in hours worked is larger in Year 2 than Year 1, and the DWL is also larger. Finally, the Department estimated that average annualized DWL was $7.2 million per year; about $200,000 of DWL (2.7 percent) was attributed to affected HCE workers, and the remaining $7.0 million was attributed to workers affected by the revised standard salary level. A summary of the estimates used in calculating DWL for years 1, 2 and 10 is presented in Table 30. The size of the PO 00000 Frm 00071 Fmt 4701 Sfmt 4702 DWL depends on the change in average hourly wages, the change in average hours, and the number of affected EAP workers. While the change in average hourly wages generally tends to increase over time in the projected years, the number of affected EAP workers decreases over time; because the relative decrease in workers is larger than the relative increase in wages after Year 2, there is a net decrease in annual DWL over time. E:\FR\FM\06JYP2.SGM 06JYP2 38586 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 30—SUMMARY OF PROJECTED DEADWEIGHT LOSS COMPONENT VALUES Year 1 Component Future years Year 2 Year 10 Standard Average hourly wages Pre ........................................................................................................................................ Post a .................................................................................................................................... Change ................................................................................................................................. Change in average overtime hours ............................................................................................. Affected EAP workers (1,000s) ................................................................................................... DWL Per worker per week ............................................................................................................ Nominal annual (millions) ..................................................................................................... Real annual (millions of 2013$) ........................................................................................... $15.01 $15.70 $0.68 ¥0.40 1,010 $15.09 $15.59 $0.50 ¥0.77 959 $15.47 $15.98 $0.51 ¥0.75 532 $0.14 $7.2 $7.2 $0.19 $9.7 $9.4 $0.19 $5.3 $4.3 HCE Average hourly wages Pre ........................................................................................................................................ Post [a] ................................................................................................................................. Change ................................................................................................................................. Change in average overtime hours ............................................................................................. Affected EAP workers (1,000s) ................................................................................................... DWL Per worker per week ............................................................................................................ Nominal annual (millions) ..................................................................................................... Real annual (millions of 2013$) ........................................................................................... $40.31 $42.45 $2.14 ¥0.41 12 $40.48 $41.96 $1.48 ¥0.80 11 $46.19 $47.67 $1.47 ¥0.62 3 $0.44 $0.27 $0.27 $0.59 $0.34 $0.34 $0.46 $0.07 $0.07 Note: DWL analysis is limited to workers in Types 2 and 3 who experience hour adjustments. a Despite general growth in wages, the average wage may fall slightly from Year 1 to Year 2 because the population has changed. In conclusion, because the number of affected EAP workers and consequently all costs and transfers diminish over time, the economic impact of the regulation will decrease over time as the real value of the salary levels fall. This occurs because real wages increase over time while the proposed salary levels would remain constant without automatic updating. However, if the salary levels are annually updated, the projected costs and transfers would increase over time. Cost and transfer projections based on the proposed standard salary level with annual updates are examined in section VII.E.iii. E. Automatic Updates tkelley on DSK3SPTVN1PROD with PROPOSALS2 i. Background Between periodic updates to the salary level, nominal wages typically increase, resulting in an increase in the number of workers qualifying for the EAP exemption even if there has been no change in their duties or real earnings. Thus, workers whom Congress intended to be covered by the minimum wage and overtime pay provisions of the FLSA lose that protection. Automatically updating the salary level would allow the level to keep pace with changes in either prices or earnings, keeping the real value of the salary level constant over time. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 The Department proposes to include in the regulations a mechanism for automatically updating the proposed standard salary level and proposed HCE annual compensation level annually either by maintaining a fixed percentile of earnings (40th and 90th percentile of weekly wages for full-time salaried workers, respectively) or by updating the salary and compensation levels based on changes in the CPI–U. Automatically updating the EAP standard salary level and HCE compensation level would allow these levels to continue to serve as an effective dividing line between potentially exempt and nonexempt workers. Furthermore, automatically updating the standard salary level and HCE compensation level would provide employers more certainty in knowing that the salary and compensation levels would change by a small amount each year, rather than the more disruptive increases caused by much larger changes after longer, uncertain increments of time. This would allow firms to better predict short- and longterm costs and employment needs. ii. Automatic Updating Methods 1. Introduction There are many indices that could be used to adjust the salary levels. In general, these indices are classified into PO 00000 Frm 00072 Fmt 4701 Sfmt 4702 two groups: Price indices and earnings indices. Price indices are normalized averages of prices used to measure the change in the average level of prices in an economy over time. The general growth rate of prices, also known as the inflation rate, is calculated as the annual percentage increase in the average price level. A price index is intended to measure the cost of achieving a given level of economic well-being or utility.149 Because one cannot directly observe utility or well-being, a ‘‘market basket’’ of goods and services is selected to represent a given level of utility. By keeping the contents of this basket constant, one can approximate the cost of obtaining the same level of utility at different points in time. In order to keep utility or the cost-of-living constant, incomes must rise by the same amount as the price index. An alternative to indexing the salary level to a price level is to update the salary level based upon an earnings measure. Price indices are intended to keep a consumer’s utility constant by adjusting for changes in the cost of living due to inflation. However, while price indices account for changes to the price of products in the market basket, they may not reflect the real growth in 149 Nordhaus, W.D. (1998). Quality Change in Price Indexes. Journal of Economic Perspectives, 12(1), 59–68. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules wages, growth that might result in the ability to purchase a larger ‘‘market basket.’’ Updating the salary level by maintaining it at a fixed percentile of earnings would reflect real growth in wages and keep the percentage of workers exempt roughly constant over time, but may not fully account for inflation in all circumstances. 2. Updating Methods Considered This section details the price and earnings indices that were considered as methods to update the salary levels. The Department assessed each method’s strengths, weaknesses, and current use. The methods considered include: • Consumer Price Index for all urban consumers (CPI–U) • Chained CPI (C–CPI–U) • Earnings percentiles (fixed percentiles of the distribution of weekly earnings for full-time salaried workers) tkelley on DSK3SPTVN1PROD with PROPOSALS2 The CPI–U The CPI–U is the most commonly used price index in the U.S. and is calculated monthly by BLS. The CPI–U holds quantities constant at base levels while allowing prices to change. The quantities are fixed to represent a ‘‘basket of goods and services’’ bought by the average consumer. However, most economists believe that the CPI–U overestimates the rate of inflation, although there are a broad range of views as to the sources and size of the overestimate. CPI–U estimates are generally not subject to revision. The CPI–U is the primary index used by the government to index benefit payments, program eligibility levels, and tax payments, including: • Federal income tax brackets, personal exemptions, and standard deductions.150 • Both eligibility for and benefits under the Earned Income Tax Credit (EITC).151 • Funding allocated to some government grants, such as funding to the Nutrition Education and Obesity Prevention Grant Program.152 • Treasury inflation-indexed debt securities’ interest rates.153 • Many government programs’ income eligibility requirements, including school meal programs.154 • Federal poverty levels, which determine eligibility for many government social assistance programs.155 150 26 U.S.C. 1(f). 151 Id. U.S.C. 2036a(d)(1)(F). CFR part 456, appendix D. 154 42 U.S.C. 1758(b)(1). 155 42 U.S.C. 9902(2). 153 31 19:25 Jul 02, 2015 3. Comparison of Indices The Department proposes to automatically update the standard salary level and the HCE annual compensation level annually either by maintaining them at a fixed percentile of earnings (the 40th and 90th percentiles of weekly wages for full-time salaried workers, respectively) or by updating the levels based on changes in the CPI–U. Updating salary and 156 Congressional Budget Office. (2010). Using a Different Measure of Inflation for Indexing Federal Programs and the Tax Code. https://www.cbo.gov/ publication/25036. 157 See https://www.bls.gov/cpi/cpisupqa.htm. 152 7 VerDate Sep<11>2014 The Chained CPI–U (C–CPI–U) The C–CPI–U is a variation of the CPI–U. The C–CPI–U is an index that accounts for changes in the market basket of goods from one year to the next. The C–CPI–U results in inflation estimates roughly 0.3 percentage points lower than the CPI–U.156 Although the C–CPI–U is viewed by some as a more accurate measure of inflation than the CPI–U, it has shortcomings as an indexation method. ‘‘The C–CPI–U requires data on changes in consumers’ spending patterns. Since those data are not available for several years the BLS releases preliminary estimates of the C–CPI–U and revises them over the following two years.’’ 157 Thus any measure using the C–CPI–U would have to be either (1) indexed to a preliminary estimate of the C–CPI–U that is subject to estimation error and revision or (2) indexed to changes in prices from a few years prior. Earnings percentiles (fixed percentiles of the distribution of weekly earnings for full-time salaried workers) Updating the salary levels based upon the growth rate of earnings at a specified percentile of the earnings distribution is consistent with the Department’s historical practice of using salary level as a key criterion for the exemption. The growth rate of earnings reflecting labor market conditions is an appropriate measure of the relative status, responsibility, and independence that characterize exempt workers. While earnings and prices generally mirror one another over time, they do not change in tandem. A price index maintains a constant level of utility or economic well-being; an earnings index reflects real gains in the standard of living. Accordingly, if earnings grow more quickly than prices an earnings index will increase the salary levels by more than a price index. Conversely, if prices grow more quickly than earnings a price index will increase the salary levels more than an earnings index. Jkt 235001 PO 00000 Frm 00073 Fmt 4701 Sfmt 4702 38587 compensation levels based on earnings would keep the share of workers who are exempt fairly constant over time, while updating based on prices will keep the earnings power of the levels constant over time. The Department is seeking detailed comments on both methods of updating the standard salary and HCE compensation levels. The CPI–U is based on a tremendous amount of data that represents average prices paid by a majority of Americans and is by far the best-known and most widely-used index. While earnings percentiles are less familiar, BLS publishes the deciles of weekly earnings for full-time salaried workers on a quarterly basis. In recent years the CPI–U has grown at a rate very closely aligned with the 40th percentile of earnings for full-time salaried workers; between 2003 and 2013 the average annual growth rates for the 40th percentile and CPI–U have been: 2.6 percent and 2.4 percent respectively. The Department therefore expects that both methods would produce similar standard salary levels in future years. Growth in CPI–U in recent years has been smaller than growth at the 90th percentile of earnings, however, so the HCE total annual compensation levels generated by these two methods may vary in the future. iii. Estimated Impacts of Automatically Updating the EAP Salary and HCE Compensation Levels In section VII.D.x. the Department projected ten years of costs and transfers due to a one-time increase in the standard salary and HCE compensation levels. Updating these salary levels annually will increase the number of affected workers because more workers will earn below the higher indexed salary levels than the fixed salary levels. Consequently, the projected costs and transfers of the proposed rule will increase with indexation. In this section, the Department describes and quantifies the annual costs and transfer payments associated with automatically updating the salary levels under both methods (fixed percentile and CPI–U). To predict the salary and compensation levels in 2014 through 2022 using the fixed percentile method, the Department estimated the salary levels using data from 2003 through 2013, calculated the geometric average annual growth rate, and applied it to the future years. For example, between 2003 and 2013 the 40th percentile of earnings for full-time salaried workers increased by an average of 2.6 percent annually; therefore, the projected salary level for Year 2 is $945 ($921 × 1.026). For the E:\FR\FM\06JYP2.SGM 06JYP2 38588 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules CPI–U method, the Department used the predicted annual CPI–U values for 2014 through 2022 from the Congressional Budget Office.158 For example, CPI–U for 2014 is predicted to be 1.5 percent; therefore, the projected salary level for Year 2 is $935 ($921 × 1.015). In other years, predicted CPI–U ranges from 1.9 percent to 2.4 percent. As the required salary levels are updated in Year 2 through Year 10 of the analysis, more workers will potentially be affected with automatic updating than without. With automatic updating of the salary levels, the number of affected EAP workers is projected to increase from 4.7 million to between 5.1 and 5.6 million over 10 years, depending on the updating methodology used. Conversely, in the absence of automatic updating, the number of affected EAP workers is projected to decline from 4.7 to 2.7 million (Table 31). The relatively constant number of affected workers over the years with updating validates the choice of indexing methods. Starting in Year 1 and running through Year 10 the population of affected workers as a percent of potentially affected workers (defined using the current salary level) increases modestly from 21.9 to 23.4 percent using the fixed percentile method, but declines modestly to 21.2 percent using the CPI–U method. The three costs to employers previously considered are (1) regulatory familiarization costs, (2) adjustment costs, and (3) managerial costs. Regulatory familiarization costs only occur in Year 1 and thus do not vary with automatic updating. Adjustment costs and managerial costs are a function of the number of affected EAP workers and so will be higher with automatic updating. Adjustment costs will occur in projected years when workers are newly affected (which— while relatively rare—will be more common with automatic updating than without). Management costs recur each year for all affected EAP workers whose hours are adjusted. Therefore, managerial costs fall significantly over time without updating (since the number of affected EAP workers decreases over time) but increase modestly over time with annual updating (where the number of affected EAP workers increases over time because of the higher salary level). Similarly, transfers and DWL will both be higher with automatic updating than without because the number of affected workers will increase, rather than decrease, over time. Table 31 presents the projected estimated costs, transfer payments, and DWL with and without automatic updating. Total direct costs were projected to decrease from $592.7 million in Year 1 to $225.3 million in Year 10 with fixed percentile updating and to $198.6 million in Year 10 with CPI–U updating. In the absence of automatic updating, costs were projected to decrease to $93.1 million in Year 10. Transfers from employers to employees were projected to decrease from $1,482.5 million to $1,339.6 million using the fixed percentile method, and to $1,191.4 million using the CPI–U method. Without updating, transfers were projected to decrease to $500.1 million in Year 10. DWL increases over time with automatic updating, but decreases over time without it. With updating, DWL was estimated to increase from $7.4 million to $11.2 million (fixed percentile method) or to $9.7 million (CPI–U method), but decline from $7.4 million to $4.3 million without updating. TABLE 31—PROJECTED COSTS AND TRANSFERS; STANDARD AND HCE SALARY LEVELS, WITH AND WITHOUT AUTOMATIC UPDATING Year Automatic updating method a 1 2 3 . . . 8 9 10 Affected Workers (Millions) Without ..................................................... Percentile ................................................. CPI–U ....................................................... 4.7 4.7 4.7 4.5 4.8 4.7 4.2 4.9 4.7 . . . . . . . . . 3.1 5.4 4.9 2.9 5.5 5.0 2.7 5.6 5.1 . . . . . . . . . $108.9 214.8 191.6 $100.6 220.1 195.2 $93.1 225.3 198.6 . . . . . . . . . $596.9 1,315.2 1,150.6 $551.0 1,320.6 1,192.7 $500.1 1,339.6 1,191.4 . . . . . . . . . $5.3 11.0 9.7 $4.9 11.1 9.7 $4.3 11.2 9.7 Total Direct Employer Costs (Millions 2013$) Without ..................................................... Percentile ................................................. CPI–U ....................................................... $592.7 592.7 592.7 $170.0 188.8 181.1 $155.8 191.9 178.6 Total Transfers (Millions 2013$) Without ..................................................... Percentile ................................................. CPI–U ....................................................... $1,482.5 1,482.5 1,482.5 $1,061.2 1,160.2 1,126.4 $963.4 1,162.4 1,104.3 DWL (Millions 2013$) tkelley on DSK3SPTVN1PROD with PROPOSALS2 Without ..................................................... Percentile ................................................. CPI–U ....................................................... $7.4 7.4 7.4 $9.8 10.8 10.3 $8.9 10.9 10.1 Note: For the purposes of projecting costs, transfers, and DWL, Year 1 corresponds to 2013 and Year 10 corresponds to 2022. a The percentile method sets the standard salary level at the 40th percentile of weekly earnings for full-time salaried workers and the HCE compensation level at the 90th percentile. The CPI–U method adjusts both salary levels based on the annual percent change in the CPI–U. 158 Congressional Budget Office. (2014). The Budget and Economic Outlook: 2014 to 2024. Pub. No. 4869. Table G–2. VerDate Sep<11>2014 20:32 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00074 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38589 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules In Years 1 through 10, using a 7 percent real discount rate, total annualized adjustment and managerial costs were estimated to average between $205.7 and $221.4 million per year with automatic updating (using CPI–U or fixed percentile, respectively) and $160.3 million without updating (Table 32). Therefore, the incremental average annualized direct employer costs of automatic updating is between $45.4 automatic updating (using CPI–U or fixed percentile, respectively) and $7.2 million per year without updating. Thus, automatic updating increases DWL by between $2.3 and $3.3 million per year on average. Benefits were not monetized for either Year 1 or Years 2 through 10; therefore this section does not repeat the previous discussion on potential benefits. and $61.1 million per year. Average annualized total transfers were estimated to be between $1,178.0 and $1,271.4 million with automatic updating (using CPI–U or fixed percentile, respectively) and $872.9 million without updating, resulting in incremental transfers of between $305.2 and $398.5 million per year. Projected average annualized DWL totals between $9.5 and $10.5 million per year with TABLE 32—SUMMARY OF TEN-YEAR AVERAGE ANNUALIZED REGULATORY COSTS AND TRANSFERS, STANDARD AND HCE SALARY LEVELS, WITH AND WITHOUT AUTOMATIC UPDATING Average annualized values (Millions 2013$) a Cost/transfer Without updating Fixed percentile Values CPI–U Difference Values Difference Regulatory Familiarization Costs Regulatory familiarization b ...................................................................... $33.9 $33.9 $0.0 $33.9 $0.0 $218.6 1,232.4 10.0 $59.9 379.1 3.0 $203.3 1,144.2 9.2 $44.6 290.8 2.2 $1.7 19.5 0.2 $2.9 39.0 0.5 $1.2 19.5 0.3 $2.4 33.8 0.3 $0.8 14.3 0.2 $160.3 872.9 7.2 $221.4 1,271.4 10.5 $61.1 398.5 3.3 $205.7 1,178.0 9.5 $45.4 305.2 2.3 Standard Salary Level Adj. & managerial costs ........................................................................... Transfers .................................................................................................. DWL ......................................................................................................... $158.7 853.4 7.0 HCE Compensation Level Adj. & managerial costs ........................................................................... Transfers .................................................................................................. DWL ......................................................................................................... Total Adj. & managerial costs ........................................................................... Transfers .................................................................................................. DWL ......................................................................................................... a Over ten years, using a discount rate of 7 percent. familiarization costs are a one-time cost that do not vary based on the proposed salary levels or automatic updating. b Regulatory tkelley on DSK3SPTVN1PROD with PROPOSALS2 The above table demonstrates that the two updating methods yield similar costs and transfers estimates. However, this does not imply these indices will necessarily result in similar salary levels over time. The Department compared the standard salary levels that would VerDate Sep<11>2014 20:32 Jul 02, 2015 Jkt 235001 have resulted from 1998 to 2013 if (1) the standard salary level was set each year to the 40th percentile of weekly earnings for full-time salaried workers, and (2) the standard salary level was set using the growth in the CPI–U (and setting the level in 2013 to match the PO 00000 Frm 00075 Fmt 4701 Sfmt 4702 40th percentile earnings level, i.e., $921 per week) (Figure 5). While not identical, the data show that during this sixteen year period these two methods produced similar results. E:\FR\FM\06JYP2.SGM 06JYP2 38590 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules The Department has not proposed specific revisions to the standard duties tests; however, as mentioned in section III., we received significant input regarding this issue from both employer and employee representatives during the Department’s stakeholder listening sessions. If changes were made to the standard duties tests, the Department would need to consider whether any of the probabilities of exemption for specific occupations used in the analysis would need to be revised since the new duties test would potentially result in workers in some occupations being more or less likely to meet the duties tests. The Department has begun to consider whether O*NET can be used to identify any occupations for which the Department may need to adjust its assumptions of the likelihood of exemption should the Department revise the duties test. The O*NET database contains information on hundreds of standardized and occupation-specific descriptors. The database, which is available to the public, is continually updated by surveying a broad range of workers from each occupation. The database of occupational requirements and worker attributes describes occupations in terms of the skills and knowledge required, how the work is performed, and typical work settings.159 For each occupation, O*NET includes a list of tasks performed, and rates the 159 See https://www.onetcenter.org/overview.html. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 tasks’ frequency, importance, relevance, and whether it is a core or supplemental task. O*NET also includes data on work activities, including the importance, relevancy, and frequency of specified tasks performed in each occupation. This information could inform the Department in determining whether the task is indicative of exempt duties. The Department believes it could use O*NET data to construct a model to identify occupations for which the probability of exemption would be impacted by any changes to the duties tests. The Department also could look to O*NET data to determine changes to the probability codes for those identified occupations. Therefore, if there are any changes to the duties test, the Department would likely update its estimate of the impact of the rule based on its analysis of the O*NET data for any occupations for which the probability codes were modified. The Department invites detailed comments on this proposed methodology and alternative data sources for determining the impact of any changes to the standard duties tests. Appendix A: Methodology for Estimating Exemption Status The number of workers exempt under the FLSA’s part 541 regulations is unknown. It is neither reported by employers to any central agency nor asked in either an employee or establishment survey. The Department estimated the number of exempt workers using the following methodology. This methodology is based largely on the approach used PO 00000 Frm 00076 Fmt 4701 Sfmt 4702 during the 2004 revisions.160 This appendix expands on the methodology description in this NPRM. The methodology explained there is not repeated here unless additional details are provided. A.1 The Duties Tests Probability Codes The CPS MORG data do not include information about job duties. To determine whether a worker meets the duties test the Department again employs the methodology it used in the 2004 Final Rule. Each occupation is assigned a probability representing the odds that a worker in that occupation would pass the duties test. For the EAP duties test, the five probability intervals are: • Category 0: Occupations not likely to include any workers eligible for the EAP exemptions. • Category 1: Occupations with probabilities between 90 and 100 percent. • Category 2: Occupations with probabilities between 50 and 90 percent. • Category 3: Occupations with probabilities between 10 and 50 percent. • Category 4: Occupations with probabilities between 0 and 10 percent.161 The occupations identified in this classification system represent an earlier occupational classification scheme (the 1990 Census Codes). Therefore, an occupational crosswalk was used to map the previous occupational codes to 160 69 FR 22196–22209 (Apr. 23, 2004). A2 lists the probability codes by occupation used to estimate exemption status. 161 Table E:\FR\FM\06JYP2.SGM 06JYP2 EP06JY15.004</GPH> tkelley on DSK3SPTVN1PROD with PROPOSALS2 F. Duties Test Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules the 2002 Census Occupational Codes which are used in the CPS MORG 2002 through 2010 data.162 163 When the new occupational category was comprised of more than one previous occupation, the Department assigned a probability category using the weighted average of the previous occupations’ probabilities, rounded to the closest category code. Next, the Department must determine which workers to classify as exempt.164 For example, the probability codes indicate that out of every ten public relation managers between five and nine are exempt; however, the Department does not know which five to nine workers are exempt. Exemption status could be randomly assigned but this would bias the earnings of exempt workers downward, since higher paid workers are more likely to perform the required duties. Therefore, the probability of being classified as exempt should increase with earnings. First, the Department assigned the upper bound of the probability range in each exemption category to workers with topcoded weekly earnings. For all other white collar salaried workers earning at least $455 per week in each exemption category, the Department estimated the probability of exemption for each worker in the data based on both occupation and earnings using a gamma distribution.165 166 For the gamma distribution, the shape parameter alpha was set to the squared quotient of the sample mean divided by the sample standard deviation, and the scale parameter beta was set to the sample 38591 variance divided by the sample mean. These parameter calculations are based on the method described in the 2004 rulemaking, except for the use of the standard deviation instead of the standard error.167 Table A1 shows that the expected number of workers exempt using a gamma distribution method is similar to the expected number exempt when assigning the midpoint of each probability code range to all workers in that probability code. After determining the probabilities of exemption for each worker in the data (dependent on both occupation and earnings), the Department randomly assigns exemption status to each worker, conditional on the worker’s probability of exemption. TABLE A1—COMPARISON OF PART 541—EXEMPT WORKER ESTIMATES a Midpoint probability estimate Probability code category High probability of exemption (1) ................................................................................................................ Probably exempt (2) .................................................................................................................................... Probably not exempt (3) .............................................................................................................................. Low or no probability of exemption (4) ........................................................................................................ Total ...................................................................................................................................................... a Numbers tkelley on DSK3SPTVN1PROD with PROPOSALS2 22,014,576 4,573,895 1,605,096 297,336 28,490,903 shown are the expected value of the number of workers exempt in each of the four probability code categories. The 2004 Final Rule assigned probabilities for whether workers in each occupation would pass the HCE abbreviated duties test if they earned $100,000 or more in total annual compensation; these probabilities are: • Category 0: Occupations not likely to include any workers eligible for the HCE exemption. • Category 1: Occupations with a probability of 100 percent. • Category 2: Occupations with probabilities between 94 and 96 percent. • Category 3: Occupations with probabilities between 58.4 and 60 percent. • Category 4: Occupations with a probability 15 percent. Like under the standard test, there is a positive relationship between earnings and exemption status; however, unlike the standard test, the relationship for the HCE analysis can be represented well with a linear function. Once individual probabilities are determined, workers are randomly assigned to exemption status. 162 To match 1990 Census Codes to the corresponding 2000 Census Codes see: https:// www.census.gov/people/io/methodology/. To translate the 2000 Census Codes into the 2002 Census Codes each code is multiplied by 10. 163 Beginning January 2011, the MORG data use the 2010 Census Codes. The Department translates these codes into the equivalent 2002 Census Codes to create continuity. The crosswalk is available at: https://www.census.gov/people/io/methodology/. 164 These probabilities are applied to the population of workers who are either (1) in occupational categories associated with named occupations or (2) white collar, earn $455 or more per week, and are salaried. 165 The gamma distribution was chosen because during the 2004 revision it fit the data the best of the non-linear distributions considered, which included normal, lognormal, and gamma. 69 FR 22204–08. VerDate Sep<11>2014 21,947,066 4,557,146 1,617,632 281,382 28,403,227 Gamma distribution model estimate 20:32 Jul 02, 2015 Jkt 235001 A.2 Other Exemptions There are many other exemptions to the minimum wage and overtime pay provisions of the FLSA. Accordingly, in the 2004 Final Rule, the Department excluded workers in agriculture and certain transportation occupations from the analysis. The Department now is, in addition, estimating those workers who fall under one of the other exemptions in section 13(a) of the FLSA, because such workers are exempt from both minimum wage and overtime pay under the relevant section and would remain exempt regardless of any changes to the EAP exemption. In fact, many of the workers estimated below as falling within one of the section 13(a) exemptions will already have been excluded from the analysis because they are paid on an hourly basis or are in a blue collar occupation. The PO 00000 Frm 00077 Fmt 4701 Sfmt 4702 methodology for identifying the workers who fall under the section 13(a) exemptions is explained here and is based generally on the methodology the Department used in 1998 when it issued its last report under section 4(d) of the FLSA. Section 4(d) previously required the Department to submit a report to Congress every two years regarding coverage under the FLSA. A.2.1 Section 13(a)(1) Outside Sales Workers Outside sales workers are a subset of the section 13(a)(1) exemptions, but since they are not affected by the salary regulations they are not discussed in detail in the preamble. Outside sales workers are included in occupational category ‘‘door-to-door sales workers, news and street vendors, and related workers’’ (Census code 4950). This category is composed of workers who both would and would not qualify for the outside sales worker exemption; for 166 A gamma distribution is a general type of statistical distribution that is based on two parameters, in this case alpha and beta. 167 Since the standard error is much smaller than the sample standard deviation, using the standard error to calculate the shape and location parameters resulted in probabilities that vary less with earnings. E:\FR\FM\06JYP2.SGM 06JYP2 38592 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules example, street vendors would not qualify. Therefore, the percentage of these workers that qualify for the exemption was estimated. The Department believes that, under the 1990 Census Codes system, outside sales workers were more or less uniquely identified with occupational category ‘‘street & door-to-door sales workers’’ (277). Therefore, the Department exempts the share of workers in category 4950 who under the old classification system would have been classified as code 277 (43 percent). A.2.2 Agricultural Workers Similar to the 2004 analysis, the Department excluded agricultural workers from the universe of affected employees. Agricultural workers were identified by occupational-industry combination. However, in the 2004 Final Rule all workers in agricultural industries were excluded; here only workers also in select occupations were excluded since not all workers in agricultural industries qualify for the agricultural overtime pay exemptions. This method better approximates the true number of exempt agricultural workers and provides a more conservative estimate of the number of affected workers. Industry categories include: ‘‘crop production’’ (0170), ‘‘animal production’’ (0180), and ‘‘support activities for agriculture and forestry’’ (0290). Occupational categories include all blue collar occupations (identified with the probability codes), ‘‘farm, ranch, and other agricultural managers’’ (0200), ‘‘general and operations managers’’ (0020), and ‘‘first-line supervisors/ managers of farming, fishing, and forestry workers’’ (6000). tkelley on DSK3SPTVN1PROD with PROPOSALS2 A.2.3 Other Section 13(a) Exemptions The following methodology relies mainly on CPS MORG data but also incorporates alternative data sources when necessary. Section 13(a)(3): Seasonal amusement and recreational establishment Any employee of an amusement or recreational establishment may be exempt from minimum wage and overtime pay if the establishment meets either of the following tests: (a) It operates for seven months or less during any calendar year, or (b) its revenue for the six lowest months of the year is less than one-third of the other six months of such year. Amusement and recreational establishments are defined as ‘‘establishments frequented by the public for its amusement or recreation,’’ and ‘‘typical examples of such are the concessionaires at amusement parks and VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 beaches.’’ 168 In the CPS MORG data the Department identifies general amusement and recreation in the following industry categories: • ‘‘independent artists, performing arts, spectator sports, and related industries’’ (8560), • ‘‘museums, art galleries, historical sites, and similar institutions’’ (8570), • ‘‘bowling centers’’ (8580), • ‘‘other amusement, gambling, and recreation industries’’ (8590), and • ‘‘recreational vehicle parks and camps, and rooming and boarding houses’’ (8670).169 The CPS MORG data does not provide information on employers’ operating information or revenue. Using Business Employment Dynamics (BED) data, the Department estimated the share of leisure and hospitality employees working for establishments that are closed for at least one quarter a year.170 Although not technically the same as the FLSA definition of ‘‘seasonal,’’ this is the best available approximation of ‘‘seasonal’’ employees. The Department estimated that 3 percent of amusement and recreational workers will be exempt. The 1998 section 4(d) report estimated the number of exempt workers by applying an estimate determined in 1987 by a detailed report from the Employment Standards Administration. The Department chose not to use this estimate because it is outdated. Section 13(a)(3) also exempts employees of seasonal religious or nonprofit educational centers, but many of these workers have already been excluded from the analysis either as religious workers (not covered by the FLSA) or as teachers (professional exemption) and so are not estimated. CFR 779.385. Department does not believe that all employees in this industry category would qualify for this exemption. However, we had no way to segregate in the data employees who would and would not qualify for exemption. 170 Seasonal employment was calculated by taking the difference in employment between establishment openings (all establishments that are either opening for the first time or reopening) and establishment births (establishments that are opening for the first time)—resulting in employment in only establishments reopening. Similarly, seasonal employment was estimated by taking the difference in employment between establishment closings and establishment deaths. These two estimates were then averaged. The analysis is limited to the leisure and hospitality industry. Since the exemption is limited to workers in ‘‘establishments frequented by the public for its amusement or recreation’’ the Department must assume the rate of employment in seasonal establishments, relative to all establishments, is equivalent across these amusement or recreation establishments and all leisure and hospitality establishments. PO 00000 168 29 169 The Frm 00078 Fmt 4701 Sfmt 4702 Section 13(a)(5): Fishermen Any employee, such as a fisherman, employed in the catching, harvesting, or farming of fish or other aquatic life forms, is exempt from minimum wage and overtime pay. Fishermen are identified in occupational categories ‘‘fishers and related fishing workers’’ (6100) and ‘‘ship and boat captains and operators’’ (9310) and the industry category ‘‘fishing, hunting, and trapping’’ (0280). Workers identified in both these occupational and industry categories are considered exempt. Section 13(a)(8): Small, local newspapers This exemption from minimum wage and overtime pay applies to any employee employed by a newspaper with circulation of less than 4,000 and circulated mainly within the county where published. Newspaper employees are identified in the following occupational categories: • ‘‘news analysts, reporters and correspondents’’ (2810), • ‘‘editors’’ (2830), • ‘‘technical writers’’ (2840), • ‘‘writers and authors’’ (2850), and • ‘‘miscellaneous media and communication workers’’ (2860). The exemption is limited to the industry category ‘‘newspaper publishers’’ (6470). To limit the exemption to small, local papers, the Department limits the exemption to employees in rural areas. Although employment in a rural area is not synonymous with employment at a small newspaper, this is the best approach currently available. Alternatively, the Department could use data from Dun and Bradstreet (D&B) as was done in the 1998 section 4(d) report. This data would provide information on which establishments are in rural areas; from this the Department could estimate the share of employment in rural areas. This approach would be much more time intensive but would not necessarily provide a better result. Section 13(a)(10): Switchboard operators An independently owned public telephone company that has not more than 750 stations may claim the minimum wage and overtime pay exemption for its switchboard operators. ‘‘Switchboard operators, including answering service’’, are exempt under occupation code 5010 and industry classifications ‘‘wired telecommunications carriers’’ (6680) and ‘‘other telecommunications carriers’’ (6690). Using the 2007 Economic Census, the Department E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 estimated that 0.84 percent of employees in the relevant telecommunication sub-industries are employed by firms with fewer than ten employees (the estimated level of employment necessary to service seven hundred and fifty stations). According to the 1998 section 4(d) report, fewer than 10,000 workers were exempt in 1987 and so the Department did not develop a methodology for estimating the number exempt. Section 13(a)(12): Seamen on foreign vessels Any employee employed as a seaman on a vessel other than an American vessel is exempt from minimum wage and overtime pay. Seamen are identified by occupational categories: • ‘‘sailors and marine oilers’’ (9300), • ‘‘ship and boat captains and operators’’ (9310), and • ‘‘ship engineers’’ (9330). The CPS MORG data does not identify whether the vessel is foreign or domestic. The best approach the Department has devised is to assume that the number of workers in the occupation ‘‘deep sea foreign transportation of freight’’ (SIC 441) in 2000 is roughly equivalent to the number of workers on foreign vessels. The 2000 Occupational Employment Statistics estimates there were 14,210 workers in this occupation and thus that number of seamen are assigned exempt status on a random basis.171 Section 13(a)(15): Companions Domestic service workers employed to provide ‘‘companionship services’’ for an elderly person or a person with an illness, injury, or disability are not required to be paid the minimum wage or overtime pay. Companions are classified under occupational categories: • ‘‘nursing, psychiatric, and home health aides’’ (3600) and • ‘‘personal and home care aides’’ (4610). And industry categories: • ‘‘home health care services’’ (8170), • ‘‘individual and family services’’ (8370), and • ‘‘private households’’ (9290). All the workers who fall within these occupational and industry categories were previously excluded from the analysis because they are paid on an hourly basis and/or are in an occupation where workers have no likelihood of qualifying for the section 13(a)(1) exemption. Section 13(a)(16): Criminal investigators 171 Revisions to the SIC classification system since 2000 have eliminated this category; thus, more recent data are not available. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 The criminal investigator must be employed by the federal government and paid ‘‘availability pay.’’ 172 Criminal investigators are identified in occupational categories: • ‘‘detectives and criminal investigators’’ (3820), • ‘‘fish and game wardens’’ (3830), and • ‘‘private detectives and investigators’’ (3910). This exemption was not mentioned in the 1998 section 4(d) report. The Department exempts all workers in the occupations identified above and employed by the federal government. Section 13(a)(17): Computer workers Computer workers who meet the duties test are exempt under two sections of the FLSA. Salaried computer workers who earn a weekly salary of not less than $455 are exempt under section 13(a)(1) and computer workers who are paid hourly are exempt under section 13(a)(17) if they earn at least $27.63 an hour. Occupations that may be considered exempt include: ‘‘computer and information systems managers’’ (110), ‘‘computer scientists and systems analysts’’ (1000), ‘‘computer programmers’’ (1010), ‘‘computer software engineers’’ (1020), ‘‘computer support specialists’’ (1040), ‘‘database administrators’’ (1060), ‘‘network and computer systems administrators’’ (1100), ‘‘network systems and data communications analysts’’ (1110), ‘‘computer operators’’ (5800), and ‘‘computer control programmers and operators’’ (7900). To identify computer workers exempt under section 13(a)(17), we restrict the population to workers who are paid on an hourly basis and who earn at least $27.63 per hour. To determine which of these workers pass the computer duties test, we use the probabilities of exemption assigned to these occupations by the Department and assume a linear relationship between earnings and exemption status. A.2.4 Section 13(b) Exemptions Section 13(b)(1): Motor carrier employees This exemption eliminated overtime pay for ‘‘any employee with respect to whom the Secretary of Transportation has power to establish qualifications and maximum hours of service pursuant to the provisions of Section 31502 of Title 49[.]’’ In essence, these are motor 172 Availability pay is compensation for hours when the agent must be available to perform work over and above the standard 40 hours per week. See https://www.opm.gov/oca/pay/HTML/AP.HTM. PO 00000 Frm 00079 Fmt 4701 Sfmt 4702 38593 carrier workers,173 identified by industry category ‘‘truck transportation’’ (6170). To be exempt, these workers must engage in ‘‘safety affecting activities’’. Examples of exempt occupations include: ‘‘driver, driver’s helper, loader, or mechanic’’.174 The relevant occupational categories are: • ‘‘electronic equipment installers and repairers, motor vehicles’’ (7110), • ‘‘automotive service technicians and mechanics’’ (7200), • ‘‘bus and truck mechanics and diesel engine specialists’’ (7210), • ‘‘heavy vehicle and mobile equipment service technicians and mechanics’’ (7220), and • ‘‘driver/sales workers and truck drivers’’ (9130).175 Section 13(b)(2): Rail carrier employees Section 13(b)(2) exempts ‘‘any employee of an employer engaged in the operation of a rail carrier subject to part A of subtitle IV of Title 49.’’ 176 This includes industrial category ‘‘rail transportation’’ (6080). The 1998 methodology did not include occupational requirements but the 2004 methodology did, so this restriction was included. Occupations are limited to: • ‘‘locomotive engineers and operators’’ (9200), • ‘‘railroad brake, signal, and switch operators’’ (9230), • ‘‘railroad conductors and yardmasters’’ (9240), and • ‘‘subway, streetcar, and other rail transportation workers’’ (9260). Section 13(b)(3): Air carrier employees This section exempts employees subject to the ‘‘provisions of title II of the Railway Labor Act.’’ 177 In essence, this exempts air carrier employees, identified by industry category ‘‘air transportation’’ (6070). The 1998 methodology did not include occupational requirements but the 2004 methodology did, so this restriction was included. Occupations are limited to ‘‘aircraft pilots and flight engineers’’ 173 49 U.S.C. 31502. The text of the law is available at: https://www.gpo.gov/fdsys/pkg/ USCODE–2011-title49/html/USCODE–2011-title49subtitleVI-partB-chap315-sec31502.htm. 174 Fact Sheet #19: The Motor Carrier Exemption under the Fair Labor Standards Act (FLSA). 175 The 2004 methodology used 1990 Census codes 505, 507, and 804 which crosswalk to these occupations. However, occupations 605, 613, and 914 (included in the 1990 Census code 804 crosswalk) were excluded because under the new classification system they were deemed irrelevant. 176 49 U.S.C. 10101–11908. Text of the law is available at: https://www.gpo.gov/fdsys/pkg/ USCODE-2013-title49/pdf/USCODE-2013-title49subtitleIV-partA.pdf. 177 45 U.S.C. 181 et seq. Available at: https:// www.gpo.gov/fdsys/pkg/USCODE–2013-title45/ html/USCODE–2013-title45-chap8-subchapII.htm. E:\FR\FM\06JYP2.SGM 06JYP2 38594 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules (9030) and ‘‘aircraft mechanics and service technicians’’ (7140). Section 13(b)(6): Seamen Occupational categories include ‘‘sailors and marine oilers’’ (9300), ‘‘ship and boat captains and operators’’ (9310), and ‘‘ship engineers’’ (9330).178 The exemption is limited to the ‘‘water transportation’’ industry (6090). Section 13(b)(10): Salesmen, partsmen, or mechanics The Department limited this exemption to workers employed in a ‘‘nonmanufacturing establishment primarily engaged in the business of selling such vehicles or implements to ultimate purchasers.’’ Industry classifications include: ‘‘automobile dealers’’ (4670) and ‘‘other motor vehicle dealers’’ (4680). In the 2004 Final Rule, the industry was limited to 1990 Census code 612 which became Census code ‘‘automobile dealers’’ (4670). Category 4680 (‘‘other motor vehicle dealers’’) is also included here in keeping with the 1998 section 4(d) report methodology. The 1998 methodology did not include an occupational restriction; however, the 2004 methodology limited the exemption to automobiles, trucks, or farm implement sales workers and mechanics. Automobiles, trucks, or farm implement sales workers include: • ‘‘parts salespersons’’ (4750), and • ‘‘retail salespersons’’ (4760).179 Mechanics include: • ‘‘electronic equipment installers and repairers, motor vehicles’’ (7110), • ‘‘automotive body and related repairers’’ (7150), • ‘‘automotive glass installers and repairers’’ (7160), • ‘‘automotive service technicians and mechanics’’ (7200), • ‘‘bus and truck mechanics and diesel engine specialists’’ (7210), • ‘‘heavy vehicle and mobile equipment service technicians and mechanics’’ (7220), • ‘‘small engine mechanics’’ (7240), and • ‘‘miscellaneous vehicle and mobile equipment mechanics, installers, and repairers’’ (7260).180 TABLE A2—PROBABILITY CODES BY OCCUPATION tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 10 20 40 50 60 100 110 120 130 140 150 160 200 210 220 230 300 310 320 330 340 350 360 400 410 420 430 500 510 520 530 540 560 600 620 700 710 720 730 800 810 820 Probability code Occupation Chief executives .................................................................................................................................................................... General and operations managers ........................................................................................................................................ Advertising and promotions managers .................................................................................................................................. Marketing and sales managers ............................................................................................................................................. Public relations managers ..................................................................................................................................................... Administrative services managers ......................................................................................................................................... Computer and information systems managers ..................................................................................................................... Financial managers ............................................................................................................................................................... Human resources managers ................................................................................................................................................. Industrial production managers ............................................................................................................................................. Purchasing managers ............................................................................................................................................................ Transportation, storage, and distribution managers .............................................................................................................. Farm, ranch, and other agricultural managers ...................................................................................................................... Farmers and ranchers ........................................................................................................................................................... Construction managers .......................................................................................................................................................... Education administrators ....................................................................................................................................................... Engineering managers ........................................................................................................................................................... Food service managers ......................................................................................................................................................... Funeral directors .................................................................................................................................................................... Gaming managers ................................................................................................................................................................. Lodging managers ................................................................................................................................................................. Medical and health services managers ................................................................................................................................. Natural sciences managers ................................................................................................................................................... Postmasters and mail superintendents ................................................................................................................................. Property, real estate, and community association managers ............................................................................................... Social and community service managers .............................................................................................................................. Managers, all other ................................................................................................................................................................ Agents and business managers of artists, performers, and athletes ................................................................................... Purchasing agents and buyers, farm products ..................................................................................................................... Wholesale and retail buyers, except farm products .............................................................................................................. Purchasing agents, except wholesale, retail, and farm products ......................................................................................... Claims adjusters, appraisers, examiners, and investigators ................................................................................................. Compliance officers, except agriculture, construction, health and safety, and transportation ............................................. Cost estimators ...................................................................................................................................................................... Human resources, training, and labor relations specialists .................................................................................................. Logisticians ............................................................................................................................................................................ Management analysts ............................................................................................................................................................ Meeting and convention planners ......................................................................................................................................... Other business operations specialists ................................................................................................................................... Accountants and auditors ...................................................................................................................................................... Appraisers and assessors of real estate ............................................................................................................................... Budget analysts ..................................................................................................................................................................... 178 The 2004 methodology used 1990 Census codes 828, 829, and 833 which crosswalk to these occupations. However, occupation 952 (dredge, excavating, and loading machine operators) was excluded because under the new classification system they were deemed irrelevant. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 179 The 2004 methodology used codes 263 and 269 which crosswalk to these codes plus a few others which have been deemed irrelevant and excluded (4700, 4740, and 4850). 180 The 2004 methodology used codes 505, 506, 507, and 514 which generally crosswalk to these PO 00000 Frm 00080 Fmt 4701 Sfmt 4702 codes. A few additional codes were added which were deemed relevant (7240 and 7260). E:\FR\FM\06JYP2.SGM 06JYP2 1 1 1 1 2 1 1 1 1 1 1 1 3 0 1 1 1 3 2 2 3 1 1 0 3 1 1 2 2 2 2 2 3 1 2 1 2 2 2 1 3 2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38595 TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 830 840 850 860 900 910 930 940 950 1000 1010 1020 1040 1060 1100 1110 1200 1210 1220 1230 1240 1300 1310 1320 1330 1340 1350 1360 1400 1410 1420 1430 1440 1450 1460 1500 1510 1520 1530 1540 1550 1560 1600 1610 1640 1650 1700 1710 1720 1740 1760 1800 1810 1820 1830 1840 1860 1900 1910 1920 1930 1940 1960 2000 2010 2020 2040 2050 2060 2100 Probability code Occupation Credit analysts ....................................................................................................................................................................... Financial analysts .................................................................................................................................................................. Personal financial advisors .................................................................................................................................................... Insurance underwriters .......................................................................................................................................................... Financial examiners ............................................................................................................................................................... Loan counselors and officers ................................................................................................................................................ Tax examiners, collectors, and revenue agents ................................................................................................................... Tax preparers ........................................................................................................................................................................ Financial specialists, all other ................................................................................................................................................ Computer scientists and systems analysts ........................................................................................................................... Computer programmers ......................................................................................................................................................... Computer software engineers ............................................................................................................................................... Computer support specialists ................................................................................................................................................ Database administrators ........................................................................................................................................................ Network and computer systems administrators .................................................................................................................... Network systems and data communications analysts .......................................................................................................... Actuaries ................................................................................................................................................................................ Mathematicians ...................................................................................................................................................................... Operations research analysts ................................................................................................................................................ Statisticians ............................................................................................................................................................................ Miscellaneous mathematical science occupations ................................................................................................................ Architects, except naval ......................................................................................................................................................... Surveyors, cartographers, and photogrammetrists ............................................................................................................... Aerospace engineers ............................................................................................................................................................. Agricultural engineers ............................................................................................................................................................ Biomedical engineers ............................................................................................................................................................ Chemical engineers ............................................................................................................................................................... Civil engineers ....................................................................................................................................................................... Computer hardware engineers .............................................................................................................................................. Electrical and electronic engineers ........................................................................................................................................ Environmental engineers ....................................................................................................................................................... Industrial engineers, including health and safety .................................................................................................................. Marine engineers and naval architects ................................................................................................................................. Materials engineers ............................................................................................................................................................... Mechanical engineers ............................................................................................................................................................ Mining and geological engineers, including mining safety engineers ................................................................................... Nuclear engineers .................................................................................................................................................................. Petroleum engineers .............................................................................................................................................................. Engineers, all other ................................................................................................................................................................ Drafters .................................................................................................................................................................................. Engineering technicians, except drafters .............................................................................................................................. Surveying and mapping technicians ...................................................................................................................................... Agricultural and food scientists .............................................................................................................................................. Biological scientists ................................................................................................................................................................ Conservation scientists and foresters ................................................................................................................................... Medical scientists ................................................................................................................................................................... Astronomers and physicists ................................................................................................................................................... Atmospheric and space scientists ......................................................................................................................................... Chemists and materials scientists ......................................................................................................................................... Environmental scientists and geoscientists ........................................................................................................................... Physical scientists, all other .................................................................................................................................................. Economists ............................................................................................................................................................................ Market and survey researchers ............................................................................................................................................. Psychologists ......................................................................................................................................................................... Sociologists ............................................................................................................................................................................ Urban and regional planners ................................................................................................................................................. Miscellaneous social scientists and related workers ............................................................................................................. Agricultural and food science technicians ............................................................................................................................. Biological technicians ............................................................................................................................................................ Chemical technicians ............................................................................................................................................................. Geological and petroleum technicians .................................................................................................................................. Nuclear technicians ............................................................................................................................................................... Other life, physical, and social science technicians .............................................................................................................. Counselors ............................................................................................................................................................................. Social workers ....................................................................................................................................................................... Miscellaneous community and social service specialists ...................................................................................................... Clergy ..................................................................................................................................................................................... Directors, religious activities and education .......................................................................................................................... Religious workers, all other ................................................................................................................................................... Lawyers .................................................................................................................................................................................. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00081 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 2 2 2 1 3 2 1 2 2 1 2 1 1 1 1 1 1 1 1 1 1 1 3 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 4 4 4 1 1 1 1 1 1 1 1 3 2 2 1 2 3 2 4 4 4 4 4 4 2 3 3 0 0 0 1 38596 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 2110 2140 2150 2200 2300 2310 2320 2330 2340 2400 2430 2440 2540 2550 2600 2630 2700 2710 2720 2740 2750 2760 2800 2810 2820 2830 2840 2850 2860 2900 2910 2920 2960 3000 3010 3030 3040 3050 3060 3110 3120 3130 3140 3150 3160 3200 3210 3220 3230 3240 3250 3260 3300 3310 3320 3400 3410 3500 3510 3520 3530 3540 3600 3610 3620 3630 3640 3650 3700 3710 Probability code Occupation Judges, magistrates, and other judicial workers ................................................................................................................... Paralegals and legal assistants ............................................................................................................................................. Miscellaneous legal support workers .................................................................................................................................... Postsecondary teachers ........................................................................................................................................................ Preschool and kindergarten teachers .................................................................................................................................... Elementary and middle school teachers ............................................................................................................................... Secondary school teachers ................................................................................................................................................... Special education teachers ................................................................................................................................................... Other teachers and instructors .............................................................................................................................................. Archivists, curators, and museum technicians ...................................................................................................................... Librarians ............................................................................................................................................................................... Library Technicians ................................................................................................................................................................ Teacher assistants ................................................................................................................................................................. Other education, training, and library workers ...................................................................................................................... Artists and related workers .................................................................................................................................................... Designers ............................................................................................................................................................................... Actors ..................................................................................................................................................................................... Producers and directors ........................................................................................................................................................ Athletes, coaches, umpires, and related workers ................................................................................................................. Dancers and choreographers ................................................................................................................................................ Musicians, singers, and related workers ............................................................................................................................... Entertainers and performers, sports and related workers, all other ..................................................................................... Announcers ............................................................................................................................................................................ News analysts, reporters and correspondents ...................................................................................................................... Public relations specialists ..................................................................................................................................................... Editors .................................................................................................................................................................................... Technical writers .................................................................................................................................................................... Writers and authors ............................................................................................................................................................... Miscellaneous media and communication workers ............................................................................................................... Broadcast and sound engineering technicians and radio operators ..................................................................................... Photographers ....................................................................................................................................................................... Television, video, and motion picture camera operators and editors ................................................................................... Media and communication equipment workers, all other ...................................................................................................... Chiropractors ......................................................................................................................................................................... Dentists .................................................................................................................................................................................. Dietitians and nutritionists ...................................................................................................................................................... Optometrists ........................................................................................................................................................................... Pharmacists ........................................................................................................................................................................... Physicians and surgeons ....................................................................................................................................................... Physician assistants .............................................................................................................................................................. Podiatrists .............................................................................................................................................................................. Registered nurses .................................................................................................................................................................. Audiologists ............................................................................................................................................................................ Occupational therapists ......................................................................................................................................................... Physical therapists ................................................................................................................................................................. Radiation therapists ............................................................................................................................................................... Recreational therapists .......................................................................................................................................................... Respiratory therapists ............................................................................................................................................................ Speech-language pathologists .............................................................................................................................................. Therapists, all other ............................................................................................................................................................... Veterinarians .......................................................................................................................................................................... Health diagnosing and treating practitioners, all other .......................................................................................................... Clinical laboratory technologists and technicians .................................................................................................................. Dental hygienists ................................................................................................................................................................... Diagnostic related technologists and technicians ................................................................................................................. Emergency medical technicians and paramedics ................................................................................................................. Health diagnosing and treating practitioner support technicians .......................................................................................... Licensed practical and licensed vocational nurses ............................................................................................................... Medical records and health information technicians ............................................................................................................. Opticians, dispensing ............................................................................................................................................................. Miscellaneous health technologists and technicians ............................................................................................................. Other healthcare practitioners and technical occupations .................................................................................................... Nursing, psychiatric, and home health aides ........................................................................................................................ Occupational therapist assistants and aides ......................................................................................................................... Physical therapist assistants and aides ................................................................................................................................ Massage therapists ................................................................................................................................................................ Dental assistants ................................................................................................................................................................... Medical assistants and other healthcare support occupations ............................................................................................. First-line supervisors/managers of correctional officers ........................................................................................................ First-line supervisors/managers of police and detectives ..................................................................................................... VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00082 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 1 4 3 1 2 1 1 1 1 1 1 4 4 1 2 1 1 1 2 1 1 1 2 3 3 3 3 2 2 4 1 2 4 1 1 3 1 1 1 2 1 1 2 3 2 3 2 3 2 2 1 1 3 3 3 3 4 4 4 0 2 3 0 0 0 0 0 4 2 3 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38597 TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 3720 3730 3740 3750 3800 3820 3830 3840 3850 3860 3900 3910 3920 3940 3950 4000 4010 4020 4030 4040 4050 4060 4110 4120 4130 4140 4150 4160 4200 4210 4220 4230 4240 4250 4300 4320 4340 4350 4400 4410 4420 4430 4460 4500 4510 4520 4530 4540 4550 4600 4610 4620 4640 4650 4700 4710 4720 4740 4750 4760 4800 4810 4820 4830 4840 4850 4900 4920 4930 4940 Probability code Occupation First-line supervisors/managers of fire fighting and prevention workers .............................................................................. Supervisors, protective service workers, all other ................................................................................................................. Fire fighters ............................................................................................................................................................................ Fire inspectors ....................................................................................................................................................................... Bailiffs, correctional officers, and jailers ................................................................................................................................ Detectives and criminal investigators .................................................................................................................................... Fish and game wardens ........................................................................................................................................................ Parking enforcement workers ................................................................................................................................................ Police and sheriff’s patrol officers ......................................................................................................................................... Transit and railroad police ..................................................................................................................................................... Animal control workers .......................................................................................................................................................... Private detectives and investigators ...................................................................................................................................... Security guards and gaming surveillance officers ................................................................................................................. Crossing guards ..................................................................................................................................................................... Lifeguards and other protective service workers .................................................................................................................. Chefs and head cooks ........................................................................................................................................................... First-line supervisors/managers of food preparation and serving workers ........................................................................... Cooks ..................................................................................................................................................................................... Food preparation workers ...................................................................................................................................................... Bartenders ............................................................................................................................................................................. Combined food preparation and serving workers, including fast food .................................................................................. Counter attendants, cafeteria, food concession, and coffee shop ........................................................................................ Waiters and waitresses ......................................................................................................................................................... Food servers, nonrestaurant ................................................................................................................................................. Dining room and cafeteria attendants and bartender helpers .............................................................................................. Dishwashers .......................................................................................................................................................................... Hosts and hostesses, restaurant, lounge, and coffee shop .................................................................................................. Food preparation and serving related workers, all other ...................................................................................................... First-line supervisors/managers of housekeeping and janitorial workers ............................................................................. First-line supervisors/managers of landscaping, lawn service, and groundskeeping workers ............................................. Janitors and building cleaners ............................................................................................................................................... Maids and housekeeping cleaners ........................................................................................................................................ Pest control workers .............................................................................................................................................................. Grounds maintenance workers .............................................................................................................................................. First-line supervisors/managers of gaming workers .............................................................................................................. First-line supervisors/managers of personal service workers ............................................................................................... Animal trainers ....................................................................................................................................................................... Nonfarm animal caretakers ................................................................................................................................................... Gaming services workers ...................................................................................................................................................... Motion picture projectionists .................................................................................................................................................. Ushers, lobby attendants, and ticket takers .......................................................................................................................... Miscellaneous entertainment attendants and related workers .............................................................................................. Funeral service workers ........................................................................................................................................................ Barbers .................................................................................................................................................................................. Hairdressers, hairstylists, and cosmetologists ...................................................................................................................... Miscellaneous personal appearance workers ....................................................................................................................... Baggage porters, bellhops, and concierges .......................................................................................................................... Tour and travel guides ........................................................................................................................................................... Transportation attendants ...................................................................................................................................................... Child care workers ................................................................................................................................................................. Personal and home care aides ............................................................................................................................................. Recreation and fitness workers ............................................................................................................................................. Residential advisors ............................................................................................................................................................... Personal care and service workers, all other ........................................................................................................................ First-line supervisors/managers of retail sales workers ........................................................................................................ First-line supervisors/managers of non-retail sales workers ................................................................................................. Cashiers ................................................................................................................................................................................. Counter and rental clerks ...................................................................................................................................................... Parts salespersons ................................................................................................................................................................ Retail salespersons ............................................................................................................................................................... Advertising sales agents ........................................................................................................................................................ Insurance sales agents .......................................................................................................................................................... Securities, commodities, and financial services sales agents .............................................................................................. Travel agents ......................................................................................................................................................................... Sales representatives, services, all other .............................................................................................................................. Sales representatives, wholesale and manufacturing ........................................................................................................... Models, demonstrators, and product promoters ................................................................................................................... Real estate brokers and sales agents ................................................................................................................................... Sales engineers ..................................................................................................................................................................... Telemarketers ........................................................................................................................................................................ VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00083 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 3 3 0 0 0 0 0 0 0 0 0 4 0 0 0 0 3 0 0 0 0 0 0 0 0 0 4 0 4 3 0 0 0 0 1 4 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 2 2 4 4 4 4 2 2 2 4 3 3 4 3 3 4 38598 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 4950 4960 5000 5010 5020 5030 5100 5110 5120 5130 5140 5150 5160 5200 5210 5220 5230 5240 5250 5260 5300 5310 5320 5330 5340 5350 5360 5400 5410 5420 5500 5510 5520 5530 5540 5550 5560 5600 5610 5620 5630 5700 5800 5810 5820 5830 5840 5850 5860 5900 5910 5920 5930 6000 6010 6020 6040 6050 6100 6110 6120 6130 6200 6210 6220 6230 6240 6250 6260 6300 Probability code Occupation Door-to-door sales workers, news and street vendors, and related workers ....................................................................... Sales and related workers, all other ...................................................................................................................................... First-line supervisors/managers of office and administrative support workers ..................................................................... Switchboard operators, including answering service ............................................................................................................ Telephone operators .............................................................................................................................................................. Communications equipment operators, all other ................................................................................................................... Bill and account collectors ..................................................................................................................................................... Billing and posting clerks and machine operators ................................................................................................................ Bookkeeping, accounting, and auditing clerks ...................................................................................................................... Gaming cage workers ............................................................................................................................................................ Payroll and timekeeping clerks .............................................................................................................................................. Procurement clerks ................................................................................................................................................................ Tellers .................................................................................................................................................................................... Brokerage clerks .................................................................................................................................................................... Correspondence clerks .......................................................................................................................................................... Court, municipal, and license clerks ...................................................................................................................................... Credit authorizers, checkers, and clerks ............................................................................................................................... Customer service representatives ......................................................................................................................................... Eligibility interviewers, government programs ....................................................................................................................... File Clerks .............................................................................................................................................................................. Hotel, motel, and resort desk clerks ...................................................................................................................................... Interviewers, except eligibility and loan ................................................................................................................................. Library assistants, clerical ..................................................................................................................................................... Loan interviewers and clerks ................................................................................................................................................. New accounts clerks .............................................................................................................................................................. Order clerks ........................................................................................................................................................................... Human resources assistants, except payroll and timekeeping ............................................................................................. Receptionists and information clerks ..................................................................................................................................... Reservation and transportation ticket agents and travel clerks ............................................................................................ Information and record clerks, all other ................................................................................................................................. Cargo and freight agents ....................................................................................................................................................... Couriers and messengers ..................................................................................................................................................... Dispatchers ............................................................................................................................................................................ Meter readers, utilities ........................................................................................................................................................... Postal service clerks .............................................................................................................................................................. Postal service mail carriers ................................................................................................................................................... Postal service mail sorters, processors, and processing machine operators ....................................................................... Production, planning, and expediting clerks .......................................................................................................................... Shipping, receiving, and traffic clerks .................................................................................................................................... Stock clerks and order fillers ................................................................................................................................................. Weighers, measurers, checkers, and samplers, recordkeeping ........................................................................................... Secretaries and administrative assistants ............................................................................................................................. Computer operators ............................................................................................................................................................... Data entry keyers .................................................................................................................................................................. Word processors and typists ................................................................................................................................................. Desktop publishers ................................................................................................................................................................ Insurance claims and policy processing clerks ..................................................................................................................... Mail clerks and mail machine operators, except postal service ........................................................................................... Office clerks, general ............................................................................................................................................................. Office machine operators, except computer ......................................................................................................................... Proofreaders and copy markers ............................................................................................................................................ Statistical assistants .............................................................................................................................................................. Office and administrative support workers, all other ............................................................................................................. First-line supervisors/managers of farming, fishing, and forestry workers ........................................................................... Agricultural inspectors ........................................................................................................................................................... Animal breeders ..................................................................................................................................................................... Graders and sorters, agricultural products ............................................................................................................................ Miscellaneous agricultural workers ........................................................................................................................................ Fishers and related fishing workers ...................................................................................................................................... Hunters and trappers ............................................................................................................................................................. Forest and conservation workers .......................................................................................................................................... Logging workers .................................................................................................................................................................... First-line supervisors/managers of construction trades and extraction workers ................................................................... Boilermakers .......................................................................................................................................................................... Brickmasons, blockmasons, and stonemasons .................................................................................................................... Carpenters ............................................................................................................................................................................. Carpet, floor, and tile installers and finishers ........................................................................................................................ Cement masons, concrete finishers, and terrazzo workers .................................................................................................. Construction laborers ............................................................................................................................................................. Paving, surfacing, and tamping equipment operators ........................................................................................................... VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00084 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 4 3 1 4 4 4 4 4 4 4 4 4 4 4 4 4 3 3 3 4 4 4 4 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 0 4 4 4 4 4 4 3 4 4 4 4 4 4 4 3 3 0 0 0 0 0 0 4 0 0 0 0 0 0 0 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38599 TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 6310 6320 6330 6350 6360 6400 6420 6430 6440 6460 6500 6510 6520 6530 6600 6660 6700 6710 6720 6730 6740 6750 6760 6800 6820 6830 6840 6910 6920 6930 6940 7000 7010 7020 7030 7040 7050 7100 7110 7120 7130 7140 7150 7160 7200 7210 7220 7240 7260 7300 7310 7320 7330 7340 7350 7360 7410 7420 7430 7510 7520 7540 7550 7560 7600 7610 7620 7700 7710 7720 Probability code Occupation Pile-driver operators .............................................................................................................................................................. Operating engineers and other construction equipment operators ....................................................................................... Drywall installers, ceiling tile installers, and tapers ............................................................................................................... Electricians ............................................................................................................................................................................. Glaziers .................................................................................................................................................................................. Insulation workers .................................................................................................................................................................. Painters, construction and maintenance ............................................................................................................................... Paperhangers ........................................................................................................................................................................ Pipelayers, plumbers, pipefitters, and steamfitters ............................................................................................................... Plasterers and stucco masons .............................................................................................................................................. Reinforcing iron and rebar workers ....................................................................................................................................... Roofers .................................................................................................................................................................................. Sheet metal workers .............................................................................................................................................................. Structural iron and steel workers ........................................................................................................................................... Helpers, construction trades .................................................................................................................................................. Construction and building inspectors .................................................................................................................................... Elevator installers and repairers ............................................................................................................................................ Fence erectors ....................................................................................................................................................................... Hazardous materials removal workers .................................................................................................................................. Highway maintenance workers .............................................................................................................................................. Rail-track laying and maintenance equipment operators ...................................................................................................... Septic tank servicers and sewer pipe cleaners ..................................................................................................................... Miscellaneous construction and related workers .................................................................................................................. Derrick, rotary drill, and service unit operators, oil, gas, and mining ................................................................................... Earth drillers, except oil and gas ........................................................................................................................................... Explosives workers, ordnance handling experts, and blasters ............................................................................................. Mining machine operators ..................................................................................................................................................... Roof bolters, mining ............................................................................................................................................................... Roustabouts, oil and gas ....................................................................................................................................................... Helpers—extraction workers .................................................................................................................................................. Other extraction workers ....................................................................................................................................................... First-line supervisors/managers of mechanics, installers, and repairers .............................................................................. Computer, automated teller, and office machine repairers ................................................................................................... Radio and telecommunications equipment installers and repairers ..................................................................................... Avionics technicians .............................................................................................................................................................. Electric motor, power tool, and related repairers .................................................................................................................. Electrical and electronics installers and repairers, transportation equipment ....................................................................... Electrical and electronics repairers, industrial and utility ...................................................................................................... Electronic equipment installers and repairers, motor vehicles .............................................................................................. Electronic home entertainment equipment installers and repairers ...................................................................................... Security and fire alarm systems installers ............................................................................................................................. Aircraft mechanics and service technicians .......................................................................................................................... Automotive body and related repairers ................................................................................................................................. Automotive glass installers and repairers ............................................................................................................................. Automotive service technicians and mechanics .................................................................................................................... Bus and truck mechanics and diesel engine specialists ....................................................................................................... Heavy vehicle and mobile equipment service technicians and mechanics .......................................................................... Small engine mechanics ........................................................................................................................................................ Miscellaneous vehicle and mobile equipment mechanics, installers, and repairers ............................................................ Control and valve installers and repairers ............................................................................................................................. Heating, air conditioning, and refrigeration mechanics and installers .................................................................................. Home appliance repairers ..................................................................................................................................................... Industrial and refractory machinery mechanics ..................................................................................................................... Maintenance and repair workers, general ............................................................................................................................. Maintenance workers, machinery .......................................................................................................................................... Millwrights .............................................................................................................................................................................. Electrical power-line installers and repairers ......................................................................................................................... Telecommunications line installers and repairers ................................................................................................................. Precision instrument and equipment repairers ...................................................................................................................... Coin, vending, and amusement machine servicers and repairers ........................................................................................ Commercial divers ................................................................................................................................................................. Locksmiths and safe repairers .............................................................................................................................................. Manufactured building and mobile home installers ............................................................................................................... Riggers ................................................................................................................................................................................... Signal and track switch repairers .......................................................................................................................................... Helpers—installation, maintenance, and repair workers ....................................................................................................... Other installation, maintenance, and repair workers ............................................................................................................. First-line supervisors/managers of production and operating workers ................................................................................. Aircraft structure, surfaces, rigging, and systems assemblers ............................................................................................. Electrical, electronics, and electromechanical assemblers ................................................................................................... VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00085 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 3 0 0 38600 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued tkelley on DSK3SPTVN1PROD with PROPOSALS2 2002 Census code 7730 7740 7750 7800 7810 7830 7840 7850 7900 7920 7930 7940 7950 7960 8000 8010 8020 8030 8040 8060 8100 8120 8130 8140 8150 8160 8200 8210 8220 8230 8240 8250 8260 8300 8310 8320 8330 8340 8350 8360 8400 8410 8420 8430 8440 8450 8460 8500 8510 8520 8530 8540 8550 8600 8610 8620 8630 8640 8650 8710 8720 8730 8740 8750 8760 8800 8810 8830 8840 8850 Probability code Occupation Engine and other machine assemblers ................................................................................................................................. Structural metal fabricators and fitters .................................................................................................................................. Miscellaneous assemblers and fabricators ........................................................................................................................... Bakers .................................................................................................................................................................................... Butchers and other meat, poultry, and fish processing workers ........................................................................................... Food and tobacco roasting, baking, and drying machine operators and tenders ................................................................ Food batchmakers ................................................................................................................................................................. Food cooking machine operators and tenders ...................................................................................................................... Computer control programmers and operators ..................................................................................................................... Extruding and drawing machine setters, operators, and tenders, metal and plastic ............................................................ Forging machine setters, operators, and tenders, metal and plastic .................................................................................... Rolling machine setters, operators, and tenders, metal and plastic ..................................................................................... Cutting, punching, and press machine setters, operators, and tenders, metal and plastic ................................................. Drilling and boring machine tool setters, operators, and tenders, metal and plastic ........................................................... Grinding, lapping, polishing, and buffing machine tool setters, operators, and tenders, metal and plastic ......................... Lathe and turning machine tool setters, operators, and tenders, metal and plastic ............................................................ Milling and planing machine setters, operators, and tenders, metal and plastic .................................................................. Machinists .............................................................................................................................................................................. Metal furnace and kiln operators and tenders ...................................................................................................................... Model makers and patternmakers, metal and plastic ........................................................................................................... Molders and molding machine setters, operators, and tenders, metal and plastic .............................................................. Multiple machine tool setters, operators, and tenders, metal and plastic ............................................................................ Tool and die makers .............................................................................................................................................................. Welding, soldering, and brazing workers .............................................................................................................................. Heat treating equipment setters, operators, and tenders, metal and plastic ........................................................................ Lay-out workers, metal and plastic ........................................................................................................................................ Plating and coating machine setters, operators, and tenders, metal and plastic ................................................................. Tool grinders, filers, and sharpeners ..................................................................................................................................... Metalworkers and plastic workers, all other .......................................................................................................................... Bookbinders and bindery workers ......................................................................................................................................... Job printers ............................................................................................................................................................................ Prepress technicians and workers ........................................................................................................................................ Printing machine operators .................................................................................................................................................... Laundry and dry-cleaning workers ........................................................................................................................................ Pressers, textile, garment, and related materials .................................................................................................................. Sewing machine operators .................................................................................................................................................... Shoe and leather workers and repairers ............................................................................................................................... Shoe machine operators and tenders ................................................................................................................................... Tailors, dressmakers, and sewers ......................................................................................................................................... Textile bleaching and dyeing machine operators and tenders ............................................................................................. Textile cutting machine setters, operators, and tenders ....................................................................................................... Textile knitting and weaving machine setters, operators, and tenders ................................................................................. Textile winding, twisting, and drawing out machine setters, operators, and tenders ........................................................... Extruding and forming machine setters, operators, and tenders, synthetic and glass fibers ............................................... Fabric and apparel patternmakers ........................................................................................................................................ Upholsterers ........................................................................................................................................................................... Textile, apparel, and furnishings workers, all other .............................................................................................................. Cabinetmakers and bench carpenters .................................................................................................................................. Furniture finishers .................................................................................................................................................................. Model makers and patternmakers, wood .............................................................................................................................. Sawing machine setters, operators, and tenders, wood ....................................................................................................... Woodworking machine setters, operators, and tenders, except sawing .............................................................................. Woodworkers, all other .......................................................................................................................................................... Power plant operators, distributors, and dispatchers ............................................................................................................ Stationary engineers and boiler operators ............................................................................................................................ Water and liquid waste treatment plant and system operators ............................................................................................ Miscellaneous plant and system operators ........................................................................................................................... Chemical processing machine setters, operators, and tenders ............................................................................................ Crushing, grinding, polishing, mixing, and blending workers ................................................................................................ Cutting workers ...................................................................................................................................................................... Extruding, forming, pressing, and compacting machine setters, operators, and tenders .................................................... Furnace, kiln, oven, drier, and kettle operators and tenders ................................................................................................ Inspectors, testers, sorters, samplers, and weighers ............................................................................................................ Jewelers and precious stone and metal workers .................................................................................................................. Medical, dental, and ophthalmic laboratory technicians ....................................................................................................... Packaging and filling machine operators and tenders .......................................................................................................... Painting workers .................................................................................................................................................................... Photographic process workers and processing machine operators ..................................................................................... Semiconductor processors .................................................................................................................................................... Cementing and gluing machine operators and tenders ........................................................................................................ VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00086 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 38601 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE A2—PROBABILITY CODES BY OCCUPATION—Continued 2002 Census code 8860 8900 8910 8920 8930 8940 8950 8960 9000 9030 9040 9110 9120 9130 9140 9150 9200 9230 9240 9260 9300 9310 9330 9340 9350 9360 9410 9420 9500 9510 9520 9560 9600 9610 9620 9630 9640 9650 9720 9730 9740 9750 Probability code Occupation Cleaning, washing, and metal pickling equipment operators and tenders ........................................................................... Cooling and freezing equipment operators and tenders ....................................................................................................... Etchers and engravers .......................................................................................................................................................... Molders, shapers, and casters, except metal and plastic ..................................................................................................... Paper goods machine setters, operators, and tenders ......................................................................................................... Tire builders ........................................................................................................................................................................... Helpers—production workers ................................................................................................................................................ Production workers, all other ................................................................................................................................................. Supervisors, transportation and material moving workers .................................................................................................... Aircraft pilots and flight engineers ......................................................................................................................................... Air traffic controllers and airfield operations specialists ........................................................................................................ Ambulance drivers and attendants, except emergency medical technicians ....................................................................... Bus drivers ............................................................................................................................................................................. Driver/sales workers and truck drivers .................................................................................................................................. Taxi drivers and chauffeurs ................................................................................................................................................... Motor vehicle operators, all other .......................................................................................................................................... Locomotive engineers and operators .................................................................................................................................... Railroad brake, signal, and switch operators ........................................................................................................................ Railroad conductors and yardmasters ................................................................................................................................... Subway, streetcar, and other rail transportation workers ..................................................................................................... Sailors and marine oilers ....................................................................................................................................................... Ship and boat captains and operators .................................................................................................................................. Ship engineers ....................................................................................................................................................................... Bridge and lock tenders ......................................................................................................................................................... Parking lot attendants ............................................................................................................................................................ Service station attendants ..................................................................................................................................................... Transportation inspectors ...................................................................................................................................................... Other transportation workers ................................................................................................................................................. Conveyor operators and tenders ........................................................................................................................................... Crane and tower operators .................................................................................................................................................... Dredge, excavating, and loading machine operators ............................................................................................................ Hoist and winch operators ..................................................................................................................................................... Industrial truck and tractor operators .................................................................................................................................... Cleaners of vehicles and equipment ..................................................................................................................................... Laborers and freight, stock, and material movers, hand ...................................................................................................... Machine feeders and offbearers ............................................................................................................................................ Packers and packagers, hand ............................................................................................................................................... Pumping station operators ..................................................................................................................................................... Refuse and recyclable material collectors ............................................................................................................................. Shuttle car operators ............................................................................................................................................................. Tank car, truck, and ship loaders .......................................................................................................................................... Material moving workers, all other ........................................................................................................................................ 0 0 0 0 0 0 0 0 3 4 3 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Appendix B. Additional Tables TABLE B1—EAP EXEMPT WORKERS POTENTIALLY AFFECTED BY THIS PROPOSED RULEMAKING, BY INDUSTRY, 2013 Potentially affected EAP workers (millions) tkelley on DSK3SPTVN1PROD with PROPOSALS2 Industry Total a .......................................................................................................................................................... Agriculture .................................................................................................................................................... Forestry, logging, fishing, hunting, and trapping ......................................................................................... Mining .......................................................................................................................................................... Construction ................................................................................................................................................. Nonmetallic mineral product manufacturing ................................................................................................ Primary metals and fabricated metal products ............................................................................................ Machinery manufacturing ............................................................................................................................ Computer and electronic product manufacturing ........................................................................................ Electrical equipment, appliance manufacturing ........................................................................................... Transportation equipment manufacturing .................................................................................................... Wood products ............................................................................................................................................. Furniture and fixtures manufacturing ........................................................................................................... Miscellaneous and not specified manufacturing ......................................................................................... Food manufacturing ..................................................................................................................................... Beverage and tobacco products .................................................................................................................. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00087 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 21.4 0.0 0.0 0.2 0.8 0.1 0.2 0.3 0.6 0.1 0.6 0.0 0.1 0.3 0.2 0.1 06JYP2 As percent of potentially affected EAP workers (percent) 100.0 0.1 0.0 0.8 3.6 0.3 1.0 1.4 2.9 0.5 2.6 0.2 0.2 1.5 0.8 0.3 38602 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE B1—EAP EXEMPT WORKERS POTENTIALLY AFFECTED BY THIS PROPOSED RULEMAKING, BY INDUSTRY, 2013— Continued Potentially affected EAP workers (millions) Industry Textile, apparel, and leather manufacturing ................................................................................................ Paper and printing ....................................................................................................................................... Petroleum and coal products manufacturing ............................................................................................... Chemical manufacturing .............................................................................................................................. Plastics and rubber products ....................................................................................................................... Wholesale trade ........................................................................................................................................... Retail trade .................................................................................................................................................. Transportation and warehousing ................................................................................................................. Utilities ......................................................................................................................................................... Publishing industries (except internet) ........................................................................................................ Motion picture and sound recording industries ........................................................................................... Broadcasting (except internet) ..................................................................................................................... Internet publishing and broadcasting .......................................................................................................... Telecommunications .................................................................................................................................... Internet service providers and data processing services ............................................................................ Other information services ........................................................................................................................... Finance ........................................................................................................................................................ Insurance ..................................................................................................................................................... Real estate ................................................................................................................................................... Rental and leasing services ........................................................................................................................ Professional and technical services ............................................................................................................ Management of companies and enterprises ............................................................................................... Administrative and support services ............................................................................................................ Waste management and remediation services ........................................................................................... Educational services .................................................................................................................................... Hospitals ...................................................................................................................................................... Health care services, except hospitals ........................................................................................................ Social assistance ......................................................................................................................................... Arts, entertainment, and recreation ............................................................................................................. Accommodation ........................................................................................................................................... Food services and drinking places .............................................................................................................. Repair and maintenance ............................................................................................................................. Personal and laundry services .................................................................................................................... Membership associations and organizations .............................................................................................. Private households ...................................................................................................................................... Public administration .................................................................................................................................... As percent of potentially affected EAP workers (percent) 0.1 0.1 0.1 0.4 0.1 0.8 1.6 0.5 0.3 0.2 0.0 0.2 0.0 0.4 0.0 0.1 1.9 1.0 0.3 0.1 3.6 0.1 0.5 0.0 0.8 1.0 1.2 0.4 0.4 0.1 0.3 0.1 0.1 0.4 0.0 0.8 0.3 0.6 0.3 2.0 0.3 3.9 7.5 2.4 1.3 0.9 0.2 0.8 0.2 1.6 0.2 0.3 9.0 4.7 1.4 0.3 16.8 0.3 2.3 0.2 3.9 4.7 5.5 1.8 1.7 0.5 1.2 0.5 0.3 1.8 0.0 3.9 Note: Pooled data for 2011–2013. a Columns may not sum to total due to rounding. tkelley on DSK3SPTVN1PROD with PROPOSALS2 VIII. Initial Regulatory Flexibility Analysis (IRFA) The Regulatory Flexibility Act of 1980 (RFA) as amended by the Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), hereafter jointly referred to as the RFA, requires agencies to prepare regulatory flexibility analyses and make them available for public comment, when proposing regulations that will have a significant economic impact on a substantial number of small entities. See 5 U.S.C. 603. If the rule is not expected to have a significant economic impact on a substantial number of small entities, the RFA allows an agency to certify such, in lieu of preparing an analysis. See 5 U.S.C. 605. The Department specifically invites comment on the impacts of the proposed rule on small businesses, including whether alternatives exist that will reduce burden on small entities VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 while still meeting the objectives of the FLSA. The Chief Counsel for Advocacy of the Small Business Administration (SBA) was notified of a draft of this rule upon submission of the rule to OMB under E.O. 12866. A. Reasons Why Action by the Agency Is Being Considered The EAP exemption salary level test that is the focus of this proposed rulemaking has been updated seven times since the FLSA was originally enacted in 1938. These updates were necessary in order for the required salary level to keep pace with increases in earnings in the economy so that it could continue to serve as an effective bright-line test that separates workers who Congress intended to remain entitled to minimum wage and overtime protection and those who may qualify as bona fide EAP exempt workers. PO 00000 Frm 00088 Fmt 4701 Sfmt 4702 The standard salary level and HCE total compensation levels have not been updated since 2004 and, as described in detail in section VII.A.ii., the standard salary level has declined considerably in real terms relative to both its 2004 and 1975 values. As a result, the exemption removes workers from overtime protection who were not intended to be within the exemption. Similarly, the HCE annual compensation requirement is out of date; more than twice as many workers earn at least $100,000 annually compared to when it was adopted in 2004. Therefore, the Department believes that rulemaking is necessary in order to restore the effectiveness of these levels. B. Statement of Objectives and Legal Basis for the Proposed Rule Section 13(a)(1) creates a minimum wage and overtime pay exemption for E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules bona fide executive, administrative, professional, outside sales employees, and teachers and academic administrative personnel, as those terms are defined and delimited by the Secretary of Labor. The regulations in part 541 contain specific criteria that define each category of exemption. The regulations also define those computer employees who are exempt under section 13(a)(1) and section 13(a)(17). To qualify for exemption, employees must meet certain tests regarding their job duties and generally must be paid on a salary basis at not less than $455 per week. The Department’s primary objective in this rulemaking is to ensure that the revised salary levels will continue to provide a useful and effective test for exemption. The salary levels were designed to operate as a ready guide to assist employers in deciding which employees were more likely to meet the duties tests for the exemptions. If left unchanged, however, the effectiveness of the salary level test as a means of determining exempt status diminishes as nonexempt employee wages increase over time. The Department last updated the salary levels in the 2004 Final Rule, setting the standard test threshold at $455 per week for EAP employees. The 2004 Final Rule also created a new ‘‘highly compensated’’ test for exemption. Under the HCE exemption, employees who are paid total annual compensation of at least $100,000 (which must include at least $455 per week paid on a salary or fee basis) are exempt from the FLSA’s overtime requirements if they customarily and regularly perform at least one of the duties or responsibilities of an exempt EAP employee identified in the standard tests for exemption. § 541.601. Employees who meet the requirements of part 541 are excluded from the Act’s minimum wage and overtime pay protections. As a result, employees may work any number of hours in the workweek and not be subject to the FLSA’s overtime pay requirements. Some State laws have stricter exemption standards than those described above. The FLSA does not preempt any such stricter State standards. If a State law establishes a higher standard than the provisions of the FLSA, the higher standard applies in that specific state. See 29 U.S.C. 218. In order to restore the ability of the standard salary level and the HCE compensation requirement to serve as appropriate bright-line tests between overtime-protected employees and employees who may be EAP exempt, the Department proposes to increase the VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 minimum salary level test from $455 to the 40th percentile of the weekly wages of all full-time salaried employees ($921 per week), and the level for the HCE test from $100,000 to the annual equivalent of the 90th percentile of weekly earnings for full-time salaried employees ($122,148 in annual earnings). The Department reached the proposed salary levels after considering available data on actual salary levels currently being paid in the economy. In order to ensure that these levels continue to function appropriately in the future, the Department also proposes to automatically update them annually either by maintaining the respective earnings percentile or updating the levels based on changes in the CPI–U. C. Description of the Number of Small Entities To Which the Proposed Rule Will Apply i. Definition of Small Entity The RFA defines a ‘‘small entity’’ as a (1) small not-for-profit organization, (2) small governmental jurisdiction, or (3) small business. The Department used the entity size standards defined by SBA to classify entities as small for the purpose of this analysis. SBA establishes separate standards for individual 6-digit NAICS industry codes, and standard cutoffs are typically based on either the average annual number of employees, or the average annual receipts. For example, the SBA has two widely used size standards: 500 employees for manufacturing, and $7 million in annual receipts for nonmanufacturing services. However, some exceptions do exist, the most notable being that depository institutions (including credit unions, commercial banks, and non-commercial banks) are classified by total assets. Small governmental jurisdictions are another noteworthy exception; they are defined as the governments of cities, counties, towns, townships, villages, school districts, or special districts with population of less than 50,000 people.181 ii. Data Sources and Methods The Department obtained data from several different sources to determine employment in small entities for each industry. Categorical tabulations from the Statistics of U.S. Businesses (SUSB, 2007 and 2011) were used for most industries. Industries that used data from alternative sources include Credit Unions (National Credit Union Association, 2010), Commercial and Non-Commercial Banks (Federal 181 See https://www.sba.gov/advocacy/regulatoryflexibility-act for details. PO 00000 Frm 00089 Fmt 4701 Sfmt 4702 38603 Depository Insurance Corporation, 2013), and Public Administration, where employees were classified based on employment estimates from the Census of Governments (2012), and local population estimates from the Census of Population and Housing (2012). The Department used the latest available data in each case, so data years differ between sources.182 For each industry, the total number of employees is organized in categories based on different characteristics of the employing entity. The categories are defined using employment, annual revenue, and assets. The Department combined these categories with the corresponding SBA standards to estimate the proportion of workers in each industry who are employed by a small entity. The general methodological approach was to classify all employees in categories below the SBA cutoff as in ‘‘small entity’’ employment. If a cutoff fell in the middle of a defined category, a uniform distribution of employees across that bracket was assumed in order to determine what proportion should be classified as in small entity employment. The Department assumed that the small entity distribution across revenue categories for Other Depository Institutions, which was not separately represented in FDIC asset data, was similar to that of Credit Unions. iii. Number of Small Entities Impacted by Proposed Rule It is difficult to estimate precisely the number of small entities that will be impacted by the proposed rule. The employee, payroll, and receipts data in SUSB are tabulated by ‘‘enterprise size,’’ where the definition of ‘‘enterprise’’ is equivalent to ‘‘entity’’ for the purposes of the current discussion. However, this data does not directly report the number of enterprises, but instead provides data on ‘‘establishments’’ (individual plants, regardless of ownership), and ‘‘firms’’ (a collection of all plants with a single owner within a given state and industry). Therefore, an enterprise may consist of multiple firms, depending on the number of states and industries it operates in. Using the SUSB number of small firms as a proxy may thus overestimate the number of small entities nationally. However, this effect is unlikely to be large, because most small entities would probably operate on smaller scales (i.e., will either consist 182 Latest available year of data for each source in parentheses. SUSB employment data are for 2011 (although since the time of writing 2012 data have become available) and receipts data are for 2007. E:\FR\FM\06JYP2.SGM 06JYP2 38604 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules of a single establishment, or operate within a single state and industry). The estimated probability that an EAP exempt worker is employed by a small entity is set equal to the calculated proportion of workers employed in the corresponding industry. For example, if an industry has 50 percent of workers employed in small entities, then on average one out of every two EAP exempt workers in this industry is expected to be small-entity employed. The Department applied these probabilities to the population of EAP exempt workers in order to find the number of workers (total and affected by the rule) employed by small entities, their payroll under the current and the proposed salary levels, and the number of small entities employing affected workers. The Department also tabulated the total number of affected entities and employees by industry group. With these limitations, the Department estimates that the proposed rule will affect 4.7 million workers in an estimated 290,800 establishments (Table 33).183 Among affected workers, 1.8 million were estimated to be employed by small entities, working in 211,000 small establishments (Table 34). While nearly 40 percent of affected EAP workers are employed in small entities, this composes a very small percentage of overall small entity employment in the economy; affected workers account for 3.5 percent of small establishment employment on average, with at most 7.0 percent of workers affected in any industry. The industries with the most affected small entity employees are: • Education and health services with 336,800 affected workers (3.5 percent of employees) in 26,800 establishments; • Professional and business services with 319,200 affected workers (5.0 percent of employees) in 56,100 establishments; and • Wholesale and retail trade with 241,700 affected workers (3.7 percent of employees) in 38,000 establishments. The financial activities industry has the largest percent of affected small entity employees; 7 percent are affected. TABLE 33—AFFECTED ENTITIES UNDER PROPOSED STANDARD SALARY AND HCE COMPENSATION LEVEL INCREASES Workers (1,000s) a Establishments (1,000s) Industry Total Total ..................................................................................................... Agriculture, forestry, fishing, & hunting ................................................ Mining .................................................................................................. Construction ......................................................................................... Manufacturing ...................................................................................... Wholesale & retail trade ...................................................................... Transportation & utilities ...................................................................... Information ........................................................................................... Financial activities ................................................................................ Professional & business services ........................................................ Education & health services ................................................................ Leisure & hospitality ............................................................................. Other services ...................................................................................... Public administration c .......................................................................... Affected b 7,427 21 28 658 296 1,475 229 134 809 1,281 910 772 722 90 Total 290.8 0.1 0.6 15.1 8.1 52.1 5.2 8.3 61.4 69.2 28.1 16.3 23.2 3.0 132,084 1,150 931 6,804 14,844 18,733 6,911 2,969 9,009 13,573 32,120 12,166 5,699 7,175 Affected Annual payroll (billions) 4,682.4 7.1 20.4 155.7 406.1 662.1 156.7 183.0 683.3 733.0 992.4 256.7 183.2 242.7 $5,881 35 61 314 759 657 334 164 499 734 1,427 303 193 399 Note: Establishment data from the Survey of U.S. Businesses 2011; worker data from CPS MORG using pooled data for 2011–2013. a Excludes the self-employed and unpaid workers. Affected workers are those who would become overtime eligible under the proposed increased salary levels if weekly earnings did not change. b The number of affected establishments depends on assumptions made by the Department. The numbers presented here assume the share of establishments that are affected is equal to the share of workers who are affected within an industry. c Establishment number represents the total number of governments, including state and local. Data from Government Organization Summary Report: 2012. TABLE 34—AFFECTED SMALL ENTITIES AND WORKERS UNDER PROPOSED STANDARD SALARY AND HCE COMPENSATION LEVEL INCREASES Small entity establishments (1,000s) Industry tkelley on DSK3SPTVN1PROD with PROPOSALS2 Total Total d ................................................................................................... Agriculture, forestry, fishing, & hunting ................................................ Mining .................................................................................................. Construction ......................................................................................... Manufacturing ...................................................................................... Wholesale & retail trade ...................................................................... Transportation & utilities ...................................................................... Information ........................................................................................... Financial activities ................................................................................ Professional & business services ........................................................ Education & health services ................................................................ Leisure & hospitality ............................................................................. Other services ...................................................................................... Public administration e .......................................................................... Affected b 6,045 20 23 640 265 1,038 178 73 550 1,121 763 632 668 73 Small entity workers (1,000s) a Total 210.6 0.1 0.6 14.3 7.2 38.0 4.1 4.6 38.7 56.1 26.8 13.0 23.4 2.5 50,355 624 351 4,373 6,372 6,600 1,711 768 2,812 6,374 9,573 6,380 3,724 692 Affected c Annual small entity payroll (billions) 1,754.0 3.9 9.8 97.8 172.6 241.7 39.7 48.6 198.2 319.2 336.8 131.6 130.2 23.9 $2,110 18 23 201 309 251 76 40 147 339 382 155 134 34 Note: Establishment data from the Survey of U.S. Businesses 2011; worker data from CPS MORG using pooled data for 2011–2013. 183 To estimate the number of establishments the ratio of affected workers to total workers was applied to the total number of establishments. For VerDate Sep<11>2014 20:32 Jul 02, 2015 Jkt 235001 example, 4.7 million of the total 132 million workers are affected, or 3.5 percent; 3.5 percent of PO 00000 Frm 00090 Fmt 4701 Sfmt 4702 the total 7.4 million establishments is 290,000 establishments with affected workers. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules 38605 a Excludes the self-employed and unpaid workers. Affected workers are those who would become overtime eligible under the proposed increased salary levels if weekly earnings did not change. b The number of affected establishments depends on assumptions made by the Department. The numbers presented here assume the share of workers in small entities who are affected is also the share of small entity establishments that are affected. c These numbers are also equal to the number of small entity establishments under the assumption that each affected establishment has one affected worker. d The components do not necessarily equal the totals due to when averages are taken. e Establishment number represents the total number of governments, including state and local. i. Costs to Small Entities As detailed in section VII.D.iv., three direct costs to employers are quantified in this analysis: (1) Regulatory familiarization costs; (2) adjustment costs; and (3) managerial costs. Regulatory familiarization costs are the costs incurred to read and become familiar with the requirements of the rule. Adjustment costs are the costs accrued to determine workers’ new exemption statuses, notify employees of policy changes, and update payroll systems. Managerial costs associated with this proposed rulemaking occur because hours of workers who are newly entitled to overtime may be more closely scheduled and monitored to minimize or avoid paying the overtime premium. Regardless of business size, the Department estimates that each establishment will spend one hour of time for regulatory familiarization; one hour per each affected worker in adjustment costs; and five minutes per week scheduling and monitoring each affected worker expected to be reclassified as overtime eligible as a result of this proposed rule. For small entities, the Department projected annual regulatory familiarization, adjustment, and managerial costs, and payments to employees in terms of extra wages paid. The Department believes that the minimum and maximum perestablishment costs are the most accurate possible estimates for the range of impact of the proposed rule on individual employers. As a direct result of this proposed rule, the Department expects total direct employer costs (regulatory familiarization, adjustment, and managerial costs) of $134.5 to 186.6 million will be incurred by small entities in the first year after the promulgation of the proposed rule (Table 35). The three industries with the most affected small entity employees (educational and health services, professional and business services, and wholesale and retail trade) account for more than 50 percent of direct costs. Average weekly earnings for affected EAP workers in small entities are expected to increase by $6.16 per week per affected worker due to both the standard salary level and HCE total annual compensation level proposed increases. This results in costs to employer of $561.5 million in wage increases to employees, which compose 0.1 to 0.8 percent of aggregate affected entity payroll (Table 36). The Department evaluated the impacts to small entities employing affected workers using a range to represent minimum and maximum costs incurred by an average establishment. To define the average establishment, the Department divided the total number of employees and payroll among small establishments by the total number of small establishments on an industryspecific basis. The minimum level of impacts is defined by assuming only one worker employed by the average establishment is affected by the revised salary level. The maximum level is defined by assuming 100 percent of workers employed by the average establishment are affected by the revised salary level.184 On average, depending on the number of affected workers it employs, an affected establishment is expected to incur $100 to $600 in direct costs and $320 to $2,700 in additional payroll to employees in the first year after the promulgation of the proposed rule. On average, these combined first year costs and transfers account for approximately 0.11 to 0.95 percent of average establishment payroll (depending on how affected small establishments are defined). 184 Larger than average small establishments in each industry might employ a larger number of affected employees, and such establishments might incur larger costs and transfers than the ‘‘average’’ establishment used as a benchmark in this analysis. However, although such establishments’ costs and transfers will increase in proportion to the number of affected workers, these establishments’ payroll will also increase in approximate proportion to the number of workers they employ. Since such establishments can never have more than 100 percent of their employees affected by the proposed rule, the rule’s impact as measured by costs and transfers as a percentage of establishment payroll will be roughly the same magnitude as an average establishment with 100 percent of employees affected. Thus, the scalability of the average establishment impacts adequately captures impacts to establishments both larger and smaller than average. tkelley on DSK3SPTVN1PROD with PROPOSALS2 D. Projected Reporting, Recordkeeping, and Other Compliance Requirements of the Proposed Rule The FLSA sets minimum wage, overtime pay, and recordkeeping requirements for employment subject to its provisions. Unless exempt, covered employees must be paid at least the minimum wage and not less than one and one-half times their regular rates of pay for overtime hours worked. Every covered employer must keep certain records for each nonexempt worker. The regulations at part 516 require employers to maintain records for employees subject to the minimum wage and overtime pay provisions of the FLSA. Thus, the recordkeeping requirements are not new requirements; however, employers would need to keep some additional records for additional affected employees if the NPRM were to be made final without change. As indicated in this analysis, the NPRM would expand minimum wage and overtime pay coverage to 4.6 million affected EAP workers (including HCE workers and excluding Type 4 workers who remain exempt). This would result in an increase in employer burden and was estimated in the PRA portion (section VI.) of this NPRM. Note that the burdens reported for the PRA section of this NPRM include the entire information collection and not merely the additional burden estimated as a result of this NPRM. VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 PO 00000 Frm 00091 Fmt 4701 Sfmt 4702 E:\FR\FM\06JYP2.SGM 06JYP2 38606 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 35—COSTS TO SMALL ENTITIES UNDER PROPOSED STANDARD SALARY AND HCE COMPENSATION LEVEL INCREASES Cost to small entities in year 1 a Industry Per affected establishment (1,000s) Total (millions) Min b Total ................................................................................. Agriculture, forestry, fishing, and hunting ........................ Mining ............................................................................... Construction ..................................................................... Manufacturing .................................................................. Wholesale and retail trade ............................................... Transportation and utilities ............................................... Information ....................................................................... Financial activities ............................................................ Professional and business services ................................ Educational and health services ...................................... Leisure and hospitality ..................................................... Other services .................................................................. Public administration ........................................................ $186.6 0.4 1.0 10.4 18.4 25.7 4.2 5.2 21.1 34.0 35.8 14.0 13.9 2.5 Max b Min b $134.5 0.3 0.7 7.6 12.7 18.8 3.0 3.7 15.6 25.0 25.2 10.0 10.2 1.8 Max b $0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 $0.6 2.2 1.2 0.5 1.8 0.5 0.7 0.8 0.4 0.4 0.9 0.8 0.4 0.7 Percent of annual payroll Min b $0.03 0.01 0.01 0.03 0.01 0.04 0.03 0.02 0.04 0.04 0.02 0.04 0.05 0.02 Max b $0.18% 0.25 0.12 0.17 0.15 0.20 0.17 0.14 0.15 0.15 0.19 0.31 0.22 0.15 Note: Pooled data for 2011–2013. a Direct costs include regulatory familiarization, adjustment, and managerial costs. b The range of costs per establishment depends on the number of affected establishments. The minimum assumes that each affected establishment has one affected worker (therefore, the number of affected establishments is equal to the number of affected workers). The maximum assumes the share of workers in small entities who are affected is also the share of small entity establishments that are affected. TABLE 36—TRANSFERS FOR SMALL ENTITIES UNDER PROPOSED STANDARD SALARY AND HCE COMPENSATION LEVEL INCREASES Transfers for small entities in year 1 a Industry Total (millions) Per affected establishment (1,000s) Min b Total ......................................................................................................... Agriculture, forestry, fishing, and hunting ................................................ Mining ...................................................................................................... Construction ............................................................................................. Manufacturing .......................................................................................... Wholesale and retail trade ....................................................................... Transportation and utilities ....................................................................... Information ............................................................................................... Financial activities .................................................................................... Professional and business services ........................................................ Educational and health services .............................................................. Leisure and hospitality ............................................................................. Other services .......................................................................................... Public administration ................................................................................ $561.5 0.5 3.1 54.4 53.5 101.4 10.2 19.9 53.1 84.2 75.1 70.0 31.4 4.7 Max b $0.32 0.12 0.31 0.56 0.31 0.42 0.26 0.41 0.27 0.26 0.22 0.53 0.24 0.20 $2.7 3.8 4.9 3.8 7.4 2.7 2.5 4.3 1.4 1.5 2.8 5.4 1.3 1.9 Percent of annual payroll Min b $0.09 0.01 0.03 0.18 0.03 0.17 0.06 0.07 0.10 0.09 0.04 0.22 0.12 0.04 Max b $0.76% 0.42 0.49 1.21 0.64 1.10 0.58 0.78 0.51 0.50 0.56 2.19 0.67 0.39 Note: Pooled data for 2011–2013. a Aggregate change in total annual payroll experienced by small entities under the proposed salary levels after labor market adjustments. This amount represents the total amount of (wage) transfers from employers to employees. b The range of transfers per establishment depends on the number of affected establishments (the denominator). The minimum assumes that each affected establishment has one affected worker (therefore, the number of affected establishments is equal to the number of affected workers). The maximum assumes the share of workers in small entities who are affected is also the share of small entity establishments that are affected. tkelley on DSK3SPTVN1PROD with PROPOSALS2 ii. Differing Compliance and Reporting Requirements for Small Entities records under usual or customary business practices. This NPRM provides no differing compliance requirements and reporting requirements for small entities. The Department has strived to minimize respondent recordkeeping burden by requiring no specific form or order of records under the FLSA and its corresponding regulations. Moreover, employers would normally maintain the iii. Least Burdensome Option or Explanation Required VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 The Department believes it has chosen the most effective option that updates and clarifies the rule and which results in the least burden. Among the options considered by the Department, the least restrictive option was taking no PO 00000 Frm 00092 Fmt 4701 Sfmt 4702 regulatory action and the most restrictive was updating the 1975 short test salary level for inflation based upon the CPI–U (which would result in a standard salary level of $1,083 per week). Taking no regulatory action does not address the Department’s concerns discussed above under Need for Regulation. The Department found the most restrictive option to be overly burdensome on business in general and E:\FR\FM\06JYP2.SGM 06JYP2 tkelley on DSK3SPTVN1PROD with PROPOSALS2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules specifically small business, and high in light of the fact that there no longer is a long duties test with an associated lower salary level that employers may use to establish that employees are exempt. Pursuant to section 603(c) of the RFA, the following alternatives are to be addressed: i. Differing compliance or reporting requirements that take into account the resources available to small entities. The FLSA creates a level playing field for businesses by setting a floor below which employers may not pay their employees. To establish differing compliance or reporting requirements for small businesses would undermine this important purpose of the FLSA and appears to not be necessary given the small annualized cost of the rule. The Year 1 cost of the proposed rule for the average employer that qualifies as small was estimated to range from a minimum of $400 to a maximum of $3,300. The Department makes available a variety of resources to employers for understanding their obligations and achieving compliance. Therefore the Department has not proposed differing compliance or reporting requirements for small businesses. ii. The clarification, consolidation, or simplification of compliance and reporting requirements for small entities. The proposed rule imposes no new reporting requirements. The Department makes available a variety of resources to employers for understanding their obligations and achieving compliance. iii. The use of performance rather than design standards. Under the proposed rule, employers may achieve compliance through a variety of means. Employers may elect to continue to claim the EAP exemption for affected employees by adjusting salary levels, hiring additional workers or spreading overtime hours to other employees, or compensating employees for overtime hours worked. The Department makes available a variety of resources to employers for understanding their obligations and achieving compliance. iv. An exemption from coverage of the rule, or any part thereof, for such small entities. Creating an exemption from coverage of this rule for businesses with as many as 500 employees, those defined as small businesses under SBA’s size standards, is inconsistent VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 with Congressional intent in the enactment of the FLSA, which applies to all employers that satisfy the enterprise coverage threshold or employ individually covered employees. See 29 U.S.C. 203(s). Moreover, creating a regulatory exemption for small businesses would be beyond the scope of the Department’s statutory authority to define and delimit the meaning of the term ‘‘employed in a bona fide executive, administrative, or professional capacity.’’ 29 U.S.C. 213(a)(1). E. Identification, to the Extent Practicable, of All Relevant Federal Rules That May Duplicate, Overlap, or Conflict With the Proposed Rule The Department is not aware of any federal rules that duplicate, overlap, or conflict with this NPRM. IX. Unfunded Mandates Reform Act Analysis The Unfunded Mandates Reform Act of 1995 (UMRA), 2 U.S.C. 1501, requires agencies to prepare a written statement for rules for which a general notice of proposed rulemaking was published and that include any federal mandate that may result in increased expenditures by state, local, and tribal governments, in the aggregate, or by the private sector, of $156 million ($100 million in 1995 dollars adjusted for inflation) or more in any one year. This statement must: (1) Identify the authorizing legislation; (2) present the estimated costs and benefits of the rule and, to the extent that such estimates are feasible and relevant, its estimated effects on the national economy; (3) summarize and evaluate state, local, and tribal government input; and (4) identify reasonable alternatives and select, or explain the non-selection, of the least costly, most cost-effective, or least burdensome alternative. A. Authorizing Legislation This proposed rule is issued pursuant to section 13(a)(1) of the Fair Labor Standards Act, 29 U.S.C. 213(a)(1). The section exempts from the FLSA’s minimum wage and overtime pay requirements ‘‘any employee employed in a bona fide executive, administrative, or professional capacity (including any employee employed in the capacity of academic administrative personnel or teacher in elementary or secondary schools), or in the capacity of outside PO 00000 Frm 00093 Fmt 4701 Sfmt 4702 38607 salesman (as such terms are defined and delimited from time to time by regulations of the Secretary, subject to the provisions of [the Administrative Procedure Act]. . .).’’ 29 U.S.C. 213(a)(1). The requirements of the exemption provided by this section of the Act are contained in part 541 of the Department’s regulations. Section 3(e) of the FLSA, 29 U.S.C. 203(e), defines ‘‘employee’’ to include most individuals employed by a state, political subdivision of a state, or interstate governmental agency. Section 3(x) of the FLSA, 29 U.S.C. 203(x), also defines public agencies to include the government of a state or political subdivision thereof, or any interstate governmental agency. B. Assessment of Costs and Benefits For purposes of the UMRA, this rule includes a Federal mandate that is expected to result in increased expenditures by the private sector of more than $156 million in at least one year, but the rule will not result in increased expenditures by state, local and tribal governments, in the aggregate, of $156 million or more in any one year. Costs to state and local governments: Based on the RIA, the Department determined that the proposed rule will result in Year 1 costs for state and local governments totaling $111.5 million; of which $28.3 million are direct employer costs and $83.2 million are transfers (Table 37). Additionally, the proposed rule will lead to $0.3 million in DWL. In subsequent years, the Department estimated that state and local governments may experience payroll increases of as much as $79.1 million in any one year when the salary level is automatically updated (with automatic updating using the fixed percentile method). Costs to the private sector: The Department determined that the proposed rule will result in Year 1 costs to the private sector of approximately $2.0 billion, of which $563.8 million are direct employer costs and $1,396.2 million are transfers. Additionally, the proposed rule will result in $7.0 million in DWL. In subsequent years, the Department estimated that the private sector may experience a payroll increase of as much as $1,219.1 million per year (with automatic updating using the fixed percentile method). E:\FR\FM\06JYP2.SGM 06JYP2 38608 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules TABLE 37—SUMMARY OF YEAR 1 AFFECTED EAP WORKERS, REGULATORY COSTS, AND TRANSFERS BY TYPE OF EMPLOYER Total Government a Private Affected EAP Workers (1,000s) Number ................................................................................................................ 4,682 4,163 507 $254.5 160.1 178.1 592.7 $251.4 142.3 170.0 563.8 $3.1 17.3 7.9 28.3 $1,482.5 $1,396.2 $83.2 $2,075.2 $1,960.0 $111.5 $7.4 $7.0 $0.3 Direct Employer Costs (Millions) Regulatory familiarization .................................................................................... Adjustment ........................................................................................................... Managerial ........................................................................................................... Total direct costs ................................................................................................. Transfers (Millions) From employers to workers ................................................................................. Direct Employer Costs & Transfers (Millions) From employers ................................................................................................... DWL (Millions) DWL b ................................................................................................................... a Includes tkelley on DSK3SPTVN1PROD with PROPOSALS2 b DWL only state, local, and tribal governments. was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. The largest benefit to workers is the transfer of income from employers; but, to the extent that the benefits to workers outweigh the costs to employers, there may be a societal welfare increase due to this transfer. The channels through which societal welfare may increase, and other secondary benefits may occur, include: Decreased litigation costs due to fewer workers subject to the duties test, the multiplier effect of the transfer, increased productivity, reduced dependence on social assistance, and a potential increase in time off and its associated benefits to the social welfare of workers. Additionally, because of the increased salary level, overtime protection will be strengthened for 6.3 million salaried white collar workers and 3.7 million salaried blue collar workers who do not meet the duties requirements for the EAP exemption, but who earn between the current minimum salary level of $455 per week and the proposed salary level because their right to minimum wage and overtime protection will be clear rather than depend upon an analysis of their duties. UMRA requires agencies to estimate the effect of a regulation on the national economy if, at its discretion, such estimates are reasonably feasible and the effect is relevant and material. 5 U.S.C. 1532(a)(4). However, OMB guidance on this requirement notes that such macroeconomic effects tend to be measurable in nationwide econometric models only if the economic impact of the regulation reaches 0.25 percent to 0.5 percent of VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 GDP, or in the range of $41.9 billion to $83.8 billion (using 2013 GDP). A regulation with smaller aggregate effect is not likely to have a measurable impact in macro-economic terms unless it is highly focused on a particular geographic region or economic sector, which is not the case with this proposed rule. The Department’s RIA estimates that the total first-year costs (direct employer costs, transfers from employers to workers, and deadweight loss) of the proposed rule will be approximately $2.0 billion for private employers and $111.8 million for state and local governments. Given OMB’s guidance, the Department has determined that a full macro-economic analysis is not likely to show any measurable impact on the economy. Therefore, these costs are compared to payroll costs and revenue to demonstrate the feasibility of adapting to these new rules. Total first-year private sector costs compose less than 0.04 percent of private sector payrolls nationwide (2013 payroll costs were estimated to be $5.4 trillion).185 Total private sector first-year costs compose less than 0.006 percent of national private sector revenues (2013 revenues were estimated to be $32.9 trillion).186 The Department concludes 185 Private sector payroll costs in 2007 were $4.8 trillion using the 2007 Economic Census of the United States. This was inflated to 2013 dollars using the CPI–U. Table EC0700A1: All sectors: Geographic Area Series: Economy-Wide Key Statistics: 2007. 186 Private sector revenues in 2007 were $29.3 trillion using the 2007 Economic Census of the PO 00000 Frm 00094 Fmt 4701 Sfmt 4702 that impacts of this magnitude are affordable and will not result in significant disruptions to typical firms in any of the major industry categories. Total first-year state and local government costs compose approximately 0.01 percent of state and local government payrolls (2013 payroll costs were estimated to be $864 billion).187 First-year state and local government costs compose 0.003 percent of state and local government revenues (2013 revenues were estimated to be $3.5 trillion).188 Impacts of this magnitude will not result in significant disruptions to typical state and local governments. The $111.5 million in state and local government costs constitutes an average of approximately $1,240 for each of the approximately 90,100 state and local entities. The Department considers impacts of this magnitude to be quite small both in absolute terms and in relation to payrolls and revenue. United States. This was inflated to 2013 dollars using the CPI–U. Table EC0700A1: All sectors: Geographic Area Series: Economy-Wide Key Statistics: 2007. 187 State and local payroll costs in 2012 were reported in the Census of Governments data as $852 billion. This was inflated to 2013 dollars using the CPI–U. 2012 Census of Governments: Employment Summary Report. Available at: https:// www2.census.gov/govs/apes/2012_summary_ report.pdf. 188 State and local revenues in 2011 were reported by the Census as $3.4 trillion. This was inflated to 2013 dollars using the CPI–U. State and Local Government Finances Summary: 2011. Available at: https://www2.census.gov/govs/local/summary_ report.pdf. E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules C. Summary of State, Local, and Tribal Government Input As part of the Department’s outreach program prior to the issuance of this NPRM, the Department conducted stakeholder listening sessions with representatives of state and local governments and tribal governments. In these sessions the Department asked stakeholders to address, among other issues, three questions: (1) What is the appropriate salary level for exemption; (2) what, if any, changes should be made to the duties tests; and (3) how can the regulations be simplified. The input received from state, local, and tribal government representatives was similar to that provided by representatives of private businesses and is summarized in section III. of this preamble. The discussions in the listening sessions have informed the development of this NPRM. The Department specifically seeks comments from state, local, and tribal governments concerning the ability of these entities to absorb the costs related to the proposed revisions. tkelley on DSK3SPTVN1PROD with PROPOSALS2 D. Least Burdensome Option or Explanation Required The Department’s consideration of various options has been described throughout the preamble. The Department believes that it has chosen the least burdensome but still costeffective mechanism to update the salary level and index future levels that is also consistent with the Department’s statutory obligation. Although some alternative options considered would have set the standard salary level at a rate lower than the proposed salary level, which might impose lower direct payroll costs on employers, that outcome may not necessarily be the most cost-effective or least burdensome alternative for employers. A lower salary level—or a degraded stagnant level over time—could result in a less effective bright-line test for separating exempt workers from those nonexempt workers intended to be within the Act’s protection. A low salary level will also increase the role of the duties test in determining whether an employee is exempt, which would increase the likelihood of misclassification and, in turn, increase the risk that employees who should receive overtime and minimum wage protections under the FLSA are denied those protections. Selecting a standard salary level inevitably impacts both the risk and cost of misclassification of overtime-eligible employees earning above the salary level as well as the risk and cost of providing overtime protection to VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 employees performing bona fide EAP duties who are paid below the salary level. An unduly low level risks increasing employer liability from unintentionally misclassifying workers as exempt; but an unduly high standard salary level increases labor costs to employers precluded from claiming the exemption for employees performing bona fide EAP duties. Thus the ultimate cost of the regulation is increased if the standard salary level is set either too low or too high. The Department has determined that setting the standard salary level at the 40th percentile of earnings for full-time salaried workers and automatically updating this level annually either by maintaining that earnings percentile or using the CPI–U best balances the risks and costs of misclassification of exempt status. X. Executive Order 13132, Federalism The Department has (1) reviewed this proposed rule in accordance with Executive Order 13132 regarding federalism and (2) determined that it does not have federalism implications. The proposed rule would not have substantial direct effects on the States, on the relationship between the national government and the States, or on the distribution of power and responsibilities among the various levels of government. XI. Executive Order 13175, Indian Tribal Governments This proposed rule would not have substantial direct effects on one or more Indian tribes, on the relationship between the Federal Government and Indian tribes, or on the distribution of power and responsibilities between the Federal Government and Indian tribes. XII. Effects on Families The undersigned hereby certifies that the proposed rule would not adversely affect the well-being of families, as discussed under section 654 of the Treasury and General Government Appropriations Act, 1999. XIII. Executive Order 13045, Protection of Children This proposed rule would have no environmental health risk or safety risk that may disproportionately affect children. XIV. Environmental Impact Assessment A review of this proposed rule in accordance with the requirements of the National Environmental Policy Act of 1969 (NEPA), 42 U.S.C. 4321 et seq.; the regulations of the Council on Environmental Quality, 40 CFR part 1500 et seq.; and the Departmental PO 00000 Frm 00095 Fmt 4701 Sfmt 4702 38609 NEPA procedures, 29 CFR part 11, indicates that the rule would not have a significant impact on the quality of the human environment. There is, thus, no corresponding environmental assessment or an environmental impact statement. XV. Executive Order 13211, Energy Supply This proposed rule is not subject to Executive Order 13211. It will not have a significant adverse effect on the supply, distribution, or use of energy. XVI. Executive Order 12630, Constitutionally Protected Property Rights This proposed rule is not subject to Executive Order 12630 because it does not involve implementation of a policy that has takings implications or that could impose limitations on private property use. XVII. Executive Order 12988, Civil Justice Reform Analysis This proposed rule was drafted and reviewed in accordance with Executive Order 12988 and will not unduly burden the Federal court system. The proposed rule was: (1) Reviewed to eliminate drafting errors and ambiguities; (2) written to minimize litigation; and (3) written to provide a clear legal standard for affected conduct and to promote burden reduction. List of Subjects in 29 CFR Part 541 Labor, Minimum wages, Overtime pay, Salaries, Teachers, Wages. Signed at Washington, DC this 18th day of June, 2015. David Weil, Administrator, Wage and Hour Division. For the reasons set out in the preamble, the Department of Labor proposes to amend title 29 of the Code of Federal Regulations, part 541 as follows: PART 541—DEFINING AND DELIMITING THE EXEMPTIONS FOR EXECUTIVE, ADMINISTRATIVE, PROFESSIONAL, COMPUTER AND OUTSIDE SALES EMPLOYEES 1. The authority citation for part 541 is revised to read as follows: ■ Authority: 29 U.S.C. 213; Pub. L. 101–583, 104 Stat. 2871; Reorganization Plan No. 6 of 1950 (3 CFR, 1945–53 Comp., p. 1004); Secretary’s Order 01–2014 (Dec. 10, 2014), 79 FR 77527 (Dec. 24, 2014). 2. Revise paragraph (a)(1) of § 541.100 to read as follows: ■ E:\FR\FM\06JYP2.SGM 06JYP2 38610 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules § 541.100 General rule for executive employees. (a) * * * (1) Compensated on a salary basis as of [EFFECTIVE DATE OF FINAL RULE] at a rate per week of not less than $921 (or $774 per week, if employed in American Samoa by employers other than the Federal government), exclusive of board, lodging or other facilities. As of [DATE TBD] on each subsequent year, compensated on a salary basis at a rate per week of not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier (with the rate for American Samoa to be calculated at 84 percent of the updated salary rate, provided that when the highest industry minimum wage for American Samoa equals the minimum wage under 29 U.S.C. 206(a)(1), exempt employees employed in all industries in American Samoa shall be paid the full salary rate), exclusive of board, lodging or other facilities; * * * * * ■ 3. Revise paragraph (a)(1) of § 541.200 to read as follows: § 541.200 General rule for administrative employees. tkelley on DSK3SPTVN1PROD with PROPOSALS2 (a) * * * (1) Compensated on a salary or fee basis as of [EFFECTIVE DATE OF FINAL RULE] at a rate per week of not less than $921 (or $774 per week, if employed in American Samoa by employers other than the Federal government), exclusive of board, lodging or other facilities. As of [DATE TBD] on each subsequent year, compensated on a salary or fee basis at a rate per week of not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier (with the rate for American Samoa to be calculated at 84 percent of the updated salary rate, provided that when the highest industry minimum wage for American Samoa equals the minimum wage under 29 U.S.C. 206(a)(1), exempt employees employed in all industries in American Samoa shall be paid the full salary rate), exclusive of board, lodging or other facilities; * * * * * ■ 4. Revise paragraph (a)(1) of § 541.204 to read as follows: § 541.204 Educational establishments. (a) * * * (1) Compensated on a salary or fee basis as of [EFFECTIVE DATE OF FINAL RULE] at a rate per week of not less than $921 (or $774 per week, if employed in American Samoa by employers other than the Federal VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 government), exclusive of board, lodging or other facilities; or on a salary basis which is at least equal to the entrance salary for teachers in the educational establishment by which employed. As of [DATE TBD] on each subsequent year, compensated on a salary or fee basis at a rate per week of not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier (with the rate for American Samoa to be calculated at 84 percent of the updated salary rate, provided that when the highest industry minimum wage for American Samoa equals the minimum wage under 29 U.S.C. 206(a)(1), exempt employees employed in all industries in American Samoa shall be paid the full salary rate), exclusive of board, lodging or other facilities; or on a salary basis which is at least equal to the entrance salary for teachers in the educational establishment by which employed; and * * * * * ■ 5. Revise paragraph (a)(1) of § 541.300 to read as follows: § 541.300 General rule for professional employees. (a) * * * (1) Compensated on a salary or fee basis as of [EFFECTIVE DATE OF FINAL RULE] at a rate per week of not less than $921 (or $774 per week, if employed in American Samoa by employers other than the Federal government), exclusive of board, lodging or other facilities. As of [DATE TBD] on each subsequent year, compensated on a salary or fee basis at a rate per week of not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier (with the rate for American Samoa to be calculated at 84 percent of the updated salary rate, provided that when the highest industry minimum wage for American Samoa equals the minimum wage under 29 U.S.C. 206(a)(1), exempt employees employed in all industries in American Samoa shall be paid the full salary rate), exclusive of board, lodging or other facilities; and * * * * * ■ 6. Remove the first sentence of § 541.400(b) introductory text and add three sentences in its place to read as follows: § 541.400 General rule for computer employees. * * * * * (b) The section 13(a)(1) exemption applies to any computer employee who, as of [EFFECTIVE DATE OF FINAL RULE] is compensated on a salary or fee PO 00000 Frm 00096 Fmt 4701 Sfmt 4702 basis at a rate per week of not less than $921 (or $774 per week, if employed in American Samoa by employers other than the Federal government), exclusive of board, lodging or other facilities. As of [DATE TBD] on each subsequent year, the section 13(a)(1) exemption applies to any computer employee who is compensated on a salary or fee basis at a rate per week of not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier (with the rate for American Samoa to be calculated at 84 percent of the updated salary rate, provided that when the highest industry minimum wage for American Samoa equals the minimum wage under 29 U.S.C. 206(a)(1), exempt employees employed in all industries in American Samoa shall be paid the full salary rate), exclusive of board, lodging or other facilities. The section 13(a)(17) exemption applies to any computer employee compensated on an hourly basis at a rate of not less than $27.63 an hour. * * * * * * * * ■ 7. Amend § 541.600 by: ■ a. Removing the first sentence of paragraph (a) and adding two sentences in its place; and ■ b. Removing the first sentence of paragraph (b) and adding two sentences in its place. The additions read as follows: § 541.600 Amount of salary required. (a) To qualify as an exempt executive, administrative or professional employee under section 13(a)(1) of the Act, an employee must be compensated on a salary basis as of [EFFECTIVE DATE OF FINAL RULE] at a rate per week of not less than $921 (or $774 per week, if employed in American Samoa by employers other than the Federal government), exclusive of board, lodging or other facilities. As of [DATE TBD] on each subsequent year, such employee must be compensated on a salary basis at a rate per week of not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier (with the rate for American Samoa to be calculated at 84 percent of the updated salary rate, provided that when the highest industry minimum wage for American Samoa equals the minimum wage under 29 U.S.C. 206(a)(1), exempt employees employed in all industries in American Samoa shall be paid the full salary rate), exclusive of board, lodging or other facilities. * * * (b) The required amount of compensation per week may be translated into equivalent amounts for periods longer than one week. The E:\FR\FM\06JYP2.SGM 06JYP2 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules requirement will be met if the employee is compensated biweekly on a salary basis of $[DOUBLE THE 40th PERCENTILE AMOUNT], semimonthly on a salary basis of $[THE 40th PERCENTILE AMOUNT, MULTIPLIED BY 52 AND DIVIDED BY 24], or monthly on a salary basis of $[THE 40th PERCENTILE AMOUNT MULTIPLIED BY 52 AND DIVIDED BY 12]. * * * * * * * * ■ 8. Amend § 541.601 by: ■ a. Revising paragraph (a); ■ b. Removing the first sentence of paragraph (b)(1) and adding two sentences in its place; and ■ c. Revising paragraph (b)(2). The revisions read as follows: tkelley on DSK3SPTVN1PROD with PROPOSALS2 § 541.601 Highly compensated employees. (a) An employee with total annual compensation of at least $122,148 as of [EFFECTIVE DATE OF FINAL RULE] is deemed exempt under section 13(a)(1) of the Act if the employee customarily and regularly performs any one or more of the exempt duties or responsibilities of an executive, administrative or professional employee identified in subparts B, C, or D of this part. As of [DATE TBD] on each subsequent year, an employee with total annual compensation of at least the updated compensation rate published annually by the Secretary in the Federal Register at least 60 days earlier is deemed exempt under section 13(a)(1) of the Act if the employee customarily and regularly performs any one or more of the exempt duties or responsibilities of an executive, administrative or professional employee identified in subparts B, C, or D of this part. (b)(1) ‘‘Total annual compensation’’ must include at least a weekly amount that is, as of [EFFECTIVE DATE OF FINAL RULE] $921 paid on a salary or fee basis. As of [DATE TBD] of each year, ‘‘total annual compensation’’ must include a weekly amount that is not less than the updated salary rate published annually by the Secretary in the Federal Register at least 60 days earlier , paid on a salary or fee basis. * * * (2) If an employee’s total annual compensation does not total at least the minimum amount established in paragraph (a) of this section by the last pay period of the 52-week period, the employer may, during the last pay period or within one month after the end of the 52-week period, make one final payment sufficient to achieve the required level. For example, if the current annual salary level for a highly compensated employee is $122,148, an employee may earn $100,000 in base salary, and the employer may anticipate based upon past sales that the employee VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 also will earn $25,000 in commissions. However, due to poor sales in the final quarter of the year, the employee actually only earns $10,000 in commissions. In this situation, the employer may within one month after the end of the year make a payment of at least $12,148 to the employee. Any such final payment made after the end of the 52-week period may count only toward the prior year’s total annual compensation and not toward the total annual compensation in the year it was paid. If the employer fails to make such a payment, the employee does not qualify as a highly compensated employee, but may still qualify as exempt under subparts B, C, or D of this part. * * * * * ■ 9. Revise § 541.604 to read as follows: § 541.604 Minimum guarantee plus extras. (a) An employer may provide an exempt employee with additional compensation without losing the exemption or violating the salary basis requirement, if the employment arrangement also includes a guarantee of at least the minimum weeklyrequired amount paid on a salary basis. Thus, for example, if the current weekly salary level is $921, an exempt employee guaranteed at least $921 each week paid on a salary basis may also receive additional compensation of a one percent commission on sales. An exempt employee also may receive a percentage of the sales or profits of the employer if the employment arrangement also includes a guarantee of at least $921 each week paid on a salary basis. Similarly, the exemption is not lost if an exempt employee who is guaranteed at least $921 each week paid on a salary basis also receives additional compensation based on hours worked for work beyond the normal workweek. Such additional compensation may be paid on any basis (e.g., flat sum, bonus payment, straight-time hourly amount, time and one-half or any other basis), and may include paid time off. (b) An exempt employee’s earnings may be computed on an hourly, a daily or a shift basis, without losing the exemption or violating the salary basis requirement, if the employment arrangement also includes a guarantee of at least the minimum weekly required amount paid on a salary basis regardless of the number of hours, days or shifts worked, and a reasonable relationship exists between the guaranteed amount and the amount actually earned. The reasonable relationship test will be met if the weekly guarantee is roughly equivalent to the employee’s usual earnings at the assigned hourly, daily or PO 00000 Frm 00097 Fmt 4701 Sfmt 4702 38611 shift rate for the employee’s normal scheduled workweek. Thus, for example, if the weekly salary level is $921, an exempt employee guaranteed compensation of at least $1,000 for any week in which the employee performs any work, and who normally works four or five shifts each week, may be paid $300 per shift without violating the salary basis requirement. The reasonable relationship requirement applies only if the employee’s pay is computed on an hourly, daily or shift basis. It does not apply, for example, to an exempt store manager paid a guaranteed salary per week that exceeds the current salary level who also receives a commission of one-half percent of all sales in the store or five percent of the store’s profits, which in some weeks may total as much as, or even more than, the guaranteed salary. ■ 10. Revise paragraph (b) of § 541.605 to read as follows: § 541.605 Fee basis. * * * * * (b) To determine whether the fee payment meets the minimum amount of salary required for exemption under these regulations, the amount paid to the employee will be tested by determining the time worked on the job and whether the fee payment is at a rate that would amount to at least the minimum required salary per week if the employee worked 40 hours. Thus, if the salary level were $921, an artist paid $500 for a picture that took 20 hours to complete meets the minimum salary requirement for exemption since earnings at this rate would yield the artist $1000 if 40 hours were worked. ■ 11. Revise § 541.709 to read as follows: § 541.709 industry. Motion picture producing The requirement that the employee be paid ‘‘on a salary basis’’ does not apply to an employee in the motion picture producing industry who is compensated, as of [EFFECTIVE DATE OF FINAL RULE], at a base rate of at least $1,404 per week (exclusive of board, lodging, or other facilities); and as of [DATE TBD] on each subsequent year, is compensated at a base rate of at least $[MOST RECENTLY EFFECTIVE MOTION PICTURE INDUSTRY BASE RATE INCREASED AT THE SAME RATIO AS THE STANDARD SALARY LEVEL IS INCREASED] (exclusive of board, lodging, or other facilities). Thus, an employee in this industry who is otherwise exempt under subparts B, C, or D of this part, and who is employed at a base rate of at least the applicable current minimum amount a week is E:\FR\FM\06JYP2.SGM 06JYP2 38612 Federal Register / Vol. 80, No. 128 / Monday, July 6, 2015 / Proposed Rules tkelley on DSK3SPTVN1PROD with PROPOSALS2 exempt if paid a proportionate amount (based on a week of not more than 6 days) for any week in which the employee does not work a full workweek for any reason. Moreover, an otherwise exempt employee in this industry qualifies for exemption if the VerDate Sep<11>2014 19:25 Jul 02, 2015 Jkt 235001 employee is employed at a daily rate under the following circumstances: (a) The employee is in a job category for which a weekly base rate is not provided and the daily base rate would yield at least the minimum weekly amount if 6 days were worked; or PO 00000 Frm 00098 Fmt 4701 Sfmt 9990 (b) The employee is in a job category having the minimum weekly base rate and the daily base rate is at least onesixth of such weekly base rate. [FR Doc. 2015–15464 Filed 7–2–15; 8:45 am] BILLING CODE P E:\FR\FM\06JYP2.SGM 06JYP2

Agencies

[Federal Register Volume 80, Number 128 (Monday, July 6, 2015)]
[Proposed Rules]
[Pages 38515-38612]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2015-15464]



[[Page 38515]]

Vol. 80

Monday,

No. 128

July 6, 2015

Part II





Department of Labor





-----------------------------------------------------------------------





Wage and Hour Division





-----------------------------------------------------------------------





29 CFR Part 541





 Defining and Delimiting the Exemptions for Executive, Administrative, 
Professional, Outside Sales and Computer Employees; Proposed Rule

Federal Register / Vol. 80 , No. 128 / Monday, July 6, 2015 / 
Proposed Rules

[[Page 38516]]


-----------------------------------------------------------------------

DEPARTMENT OF LABOR

Wage and Hour Division

29 CFR Part 541

RIN 1235-AA11


Defining and Delimiting the Exemptions for Executive, 
Administrative, Professional, Outside Sales and Computer Employees

AGENCY: Wage and Hour Division, Department of Labor.

ACTION: Proposed rule and request for comments.

-----------------------------------------------------------------------

SUMMARY: The Fair Labor Standards Act (FLSA or Act) guarantees a 
minimum wage and overtime pay at a rate of not less than one and one-
half times the employee's regular rate for hours worked over 40 in a 
workweek. While these protections extend to most workers, the FLSA does 
provide a number of exemptions. The Department of Labor (Department) 
proposes to update and revise the regulations issued under the FLSA 
implementing the exemption from minimum wage and overtime pay for 
executive, administrative, professional, outside sales, and computer 
employees. This exemption is referred to as the FLSA's ``EAP'' or 
``white collar'' exemption. To be considered exempt, employees must 
meet certain minimum tests related to their primary job duties and be 
paid on a salary basis at not less than a specified minimum amount. The 
standard salary level required for exemption is currently $455 a week 
($23,660 for a full-year worker) and was last updated in 2004.
    By way of this rulemaking, the Department seeks to update the 
salary level to ensure that the FLSA's intended overtime protections 
are fully implemented, and to simplify the identification of nonexempt 
employees, thus making the EAP exemption easier for employers and 
workers to understand. The Department also proposes automatically 
updating the salary level to prevent the level from becoming outdated 
with the often lengthy passage of time between rulemakings. Lastly, the 
Department is considering whether revisions to the duties tests are 
necessary in order to ensure that these tests fully reflect the purpose 
of the exemption.

DATES: Submit written comments on or before September 4, 2015.

ADDRESSES: You may submit comments, identified by Regulatory 
Information Number (RIN) 1235-AA11, by either of the following methods: 
Electronic Comments: Submit comments through the Federal eRulemaking 
Portal https://www.regulations.gov. Follow the instructions for 
submitting comments. Mail: Address written submissions to Mary Ziegler, 
Director of the Division of Regulations, Legislation, and 
Interpretation, Wage and Hour Division, U.S. Department of Labor, Room 
S-3502, 200 Constitution Avenue NW., Washington, DC 20210. 
Instructions: Please submit only one copy of your comments by only one 
method. All submissions must include the agency name and RIN, 
identified above, for this rulemaking. Please be advised that comments 
received will become a matter of public record and will be posted 
without change to https://www.regulations.gov, including any personal 
information provided. All comments must be received by 11:59 p.m. on 
the date indicated for consideration in this rulemaking. Commenters 
should transmit comments early to ensure timely receipt prior to the 
close of the comment period as the Department continues to experience 
delays in the receipt of mail in our area. For additional information 
on submitting comments and the rulemaking process, see the ``Public 
Participation'' section of this document. For questions concerning the 
interpretation and enforcement of labor standards related to the FLSA, 
individuals may contact the Wage and Hour Division (WHD) local district 
offices (see contact information below). Docket: For access to the 
docket to read background documents or comments, go to the Federal 
eRulemaking Portal at https://www.regulations.gov.

FOR FURTHER INFORMATION CONTACT: Mary Ziegler, Director of the Division 
of Regulations, Legislation, and Interpretation, Wage and Hour 
Division, U.S. Department of Labor, Room S-3502, 200 Constitution 
Avenue NW., Washington, DC 20210; telephone: (202) 693-0406 (this is 
not a toll-free number). Copies of this proposed rule may be obtained 
in alternative formats (Large Print, Braille, Audio Tape or Disc), upon 
request, by calling (202) 693-0675 (this is not a toll-free number). 
TTY/TDD callers may dial toll-free 1-877-889-5627 to obtain information 
or request materials in alternative formats.
    Questions of interpretation and/or enforcement of the agency's 
regulations may be directed to the nearest WHD district office. Locate 
the nearest office by calling WHD's toll-free help line at (866) 4US-
WAGE ((866) 487-9243) between 8 a.m. and 5 p.m. in your local time 
zone, or log onto WHD's Web site at https://www.dol.gov/whd/america2.htm 
for a nationwide listing of WHD district and area offices.

Electronic Access and Filing Comments

    Public Participation: This proposed rule is available through the 
Federal Register and the https://www.regulations.gov Web site. You may 
also access this document via WHD's Web site at https://www.dol.gov/whd/
. To comment electronically on Federal rulemakings, go to the Federal 
eRulemaking Portal at https://www.regulations.gov, which will allow you 
to find, review, and submit comments on Federal documents that are open 
for comment and published in the Federal Register. You must identify 
all comments submitted by including ``RIN 1235-AA11'' in your 
submission. Commenters should transmit comments early to ensure timely 
receipt prior to the close of the comment period (11:59 p.m. on the 
date identified above in the DATES section); comments received after 
the comment period closes will not be considered. Submit only one copy 
of your comments by only one method. Please be advised that all 
comments received will be posted without change to https://www.regulations.gov, including any personal information provided.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Executive Summary
II. Background
    A. What the FLSA Provides
    B. Legislative History
    C. Regulatory History
    D. Overview of Existing Regulatory Requirements
III. Presidential Memorandum
IV. Need for Rulemaking
V. Proposed Regulatory Revisions
    A. Setting the Standard Salary Level
    B. Special Salary Tests
    C. Inclusion of Nondiscretionary Bonuses in the Salary Level 
Requirement
    D. Highly Compensated Employees
    E. Automatically Updating the Salary Levels
    F. Duties Requirements for Exemption
VI. Paperwork Reduction Act
VII. Analysis Conducted In Accordance With Executive Order 12866, 
Regulatory Planning and Review, and Executive Order 13563, Improving 
Regulation and Regulatory Review
    A. Introduction
    B. Methodology To Determine the Number of Potentially Affected 
EAP Workers
    C. Determining the Revised Salary Level Test Values
    D. Impacts of Revised Salary and Compensation Level Test Values
    E. Automatic Updates
    F. Duties Test
Appendix A: Methodology for Estimating Exemption Status
Appendix B: Additional Tables
VIII. Initial Regulatory Flexibility Analysis (IRFA)

[[Page 38517]]

    A. Reasons Why Action by the Agency Is Being Considered
    B. Statement of Objectives and Legal Basis for the Proposed Rule
    C. Description of the Number of Small Entities to Which the 
Proposed Rule Will Apply
    D. Projected Reporting, Recordkeeping, and Other Compliance 
Requirements of the Proposed Rule
    E. Identification to the Extent Practicable, of All Relevant 
Federal Rules That May Duplicate, Overlap, or Conflict With the 
Proposed Rule
IX. Unfunded Mandates Reform Act Analysis
    A. Authorizing Legislation
    B. Assessment of Costs and Benefits
    C. Summary of State, Local, and Tribal Government Input
    D. Least Burdensome Option or Explanation Required
X. Executive Order 13132, Federalism
XI. Executive Order 13175, Indian Tribal Governments
XII. Effects on Families
XIII. Executive Order 13045, Protection of Children
XIV. Environmental Impact Assessment
XV. Executive Order 13211, Energy Supply
XVI. Executive Order 12630, Constitutionally Protected Property 
Rights
XVII. Executive Order 12988, Civil Justice Reform Analysis
Proposed Amendments to Regulatory Text

I. Executive Summary

    The FLSA was passed to both guarantee a minimum wage and to limit 
the number of hours an employee could work without additional 
compensation. Section 13(a)(1), which excludes certain white collar 
employees from minimum wage and overtime pay protections, was included 
in the original Act in 1938. The exemption was premised on the belief 
that the exempted workers earned salaries well above the minimum wage 
and enjoyed other privileges, including above-average fringe benefits, 
greater job security, and better opportunities for advancement, setting 
them apart from workers entitled to overtime pay. The statute delegates 
to the Secretary of Labor the authority to define and delimit the terms 
of the exemption.
    On March 13, 2014, President Obama signed a Presidential Memorandum 
directing the Department to update the regulations defining which white 
collar workers are protected by the FLSA's minimum wage and overtime 
standards. 79 FR 18737 (Apr. 3, 2014). Consistent with the President's 
goal of ensuring workers are paid a fair day's pay for a fair day's 
work, the memorandum instructed the Department to look for ways to 
modernize and simplify the regulations while ensuring that the FLSA's 
intended overtime protections are fully implemented.
    Since 1940, the regulations implementing the white collar exemption 
have generally required each of three tests to be met for the exemption 
to apply: (1) The employee must be paid a predetermined and fixed 
salary that is not subject to reduction because of variations in the 
quality or quantity of work performed (the ``salary basis test''); (2) 
the amount of salary paid must meet a minimum specified amount (the 
``salary level test''); and (3) the employee's job duties must 
primarily involve executive, administrative, or professional duties as 
defined by the regulations (the ``duties test'').
    One of the Department's primary goals in this rulemaking is 
updating the section 13(a)(1) exemption's salary requirements. The 
Department has updated the salary level requirements seven times since 
1938, most recently in 2004. Under the current regulations, an 
executive, administrative, or professional employee must be paid at 
least $455 per week ($23,660 per year for a full-year worker) in order 
to come within the standard exemption; in order to come within the 
exemption for highly compensated employees (HCE), such an employee must 
earn at least $100,000 in total annual compensation.
    The Department has long recognized the salary level test as ``the 
best single test'' of exempt status. If left at the same amount over 
time, however, the effectiveness of the salary level test as a means of 
determining exempt status diminishes as the wages of employees entitled 
to overtime increase and the real value of the salary threshold falls. 
In order to maintain the effectiveness of the salary level test, the 
Department proposes to set the standard salary level equal to the 40th 
percentile of earnings for full-time salaried workers ($921 per week, 
or $47,892 annually for a full-year worker, in 2013).\1\ The Department 
is also proposing to set the highly compensated employee annual 
compensation level equal to the 90th percentile of earnings for full-
time salaried workers ($122,148 annually). Furthermore, in order to 
prevent the levels from becoming outdated, the Department is proposing 
to include in the regulations a mechanism to automatically update the 
salary and compensation thresholds on an annual basis using either a 
fixed percentile of wages or the CPI-U.
---------------------------------------------------------------------------

    \1\ The BLS data set used to set the salary level for this 
rulemaking consists of earnings for full-time (defined as at least 
35 hours per week) non-hourly paid employees. For the purpose of 
this rulemaking, the Department considers data representing 
compensation paid to non-hourly workers to be an appropriate proxy 
for compensation paid to salaried workers. The Department relied 
upon 2013 data in the development of the NPRM. The Department will 
update the data used in the Final Rule resulting from this proposal, 
which will change the dollar figures. If, after consideration of 
comments received, the Final Rule were to adopt the proposed salary 
level of the 40th percentile of weekly earnings, the Department 
would likely rely on data from the first quarter of 2016. The latest 
data currently available are for the first quarter of 2015, in which 
the 40th percentile of weekly earnings is $951, which translates 
into $49,452 for a full-year worker. Assuming two percent growth 
between the first quarter of 2015 and the first quarter of 2016, the 
Department projects that the 40th percentile weekly wage in the 
final rule would likely be $970, or $50,440 for a full-year worker.
---------------------------------------------------------------------------

    The Department is proposing to update the salary and compensation 
levels to ensure that the FLSA's intended overtime protections are 
fully implemented and to simplify the identification of overtime-
protected and exempt employees, thus making the exemptions easier for 
employers and workers to understand. The proposed increase to the 
standard salary level is also intended to address the Department's 
conclusion that the salary level set in 2004 was too low to efficiently 
screen out from the exemption overtime-protected white collar employees 
when paired with the standard duties test. The Department believes that 
a standard salary level at the 40th percentile of all full-time 
salaried employees ($921 per week, or $47,892 for a full-year worker, 
in 2013) will accomplish the goal of setting a salary threshold that 
adequately distinguishes between employees who may meet the duties 
requirements of the EAP exemption and those who likely do not, without 
necessitating a return to the more detailed long duties test.\2\ The 
Department believes that the proposed salary compensates for the 
absence of a long test, which would have allowed employers to claim the 
exemption at a lower salary level, but only if they could satisfy a 
more restrictive duties test; moreover, it does so without setting the 
salary at a level that excludes from exemption an unacceptably high 
number of employees who meet the duties test. The Department also 
believes that, by reducing the number of workers for whom employers 
must apply the duties test to determine exempt status, this proposal is 
responsive to the President's directive to simplify the exemption. 
Similarly, the Department believes that the proposal to set the HCE 
total annual compensation level at the annualized value of the 90th 
percentile of weekly wages of all full-time salaried employees 
($122,148 per year) will ensure that the HCE

[[Page 38518]]

exemption continues to cover only employees who almost invariably meet 
all the other requirements for exemption. Finally, the Department 
proposes to automatically update the standard salary and compensation 
levels annually to ensure that they maintain their effectiveness going 
forward, either by maintaining the levels at a fixed percentile of 
earnings or by updating the amounts based on changes in the CPI-U. The 
Department believes that regularly updating the salary and compensation 
levels is the best method to ensure that these tests continue to 
provide an effective means of distinguishing between overtime-eligible 
white collar employees and those who may be bona fide EAP employees. 
The Department is not making specific proposals to modify the standard 
duties tests but is seeking comments on whether the tests are working 
as intended to screen out employees who are not bona fide EAP 
employees; in particular, the Department is concerned that in some 
instances the current tests may allow exemption of employees who are 
performing such a disproportionate amount of nonexempt work that they 
are not EAP employees in any meaningful sense.
---------------------------------------------------------------------------

    \2\ From 1949 until 2004 the regulations contained two different 
tests for exemption--a long duties test for employees paid a lower 
salary, and a short duties test for employees paid at a higher 
salary level.
---------------------------------------------------------------------------

    In 2013, there were an estimated 144.2 million wage and salary 
workers in the United States, of whom the Department estimates that 
43.0 million are white collar salaried employees who may be impacted by 
a change to the Department's part 541 regulations. Of these workers, 
the Department estimates that 21.4 million are currently exempt EAP 
workers who are subject to the salary level requirement and may be 
potentially affected by the proposed rule.\3\
---------------------------------------------------------------------------

    \3\ White collar salaried workers not subject to the EAP salary 
level test include teachers, academic administrative personnel, 
physicians, lawyers, judges, and outside sales workers.
---------------------------------------------------------------------------

    In Year 1 the Department estimates 4.6 million currently exempt 
workers who earn at least the current weekly salary level of $455 but 
less than the 40th earnings percentile ($921) would, without some 
intervening action by their employers, become entitled to minimum wage 
and overtime protection under the FLSA (Table ES1). Similarly, an 
estimated 36,000 currently exempt workers who earn at least $100,000 
but less than the 90th earnings percentile ($122,148) per year and who 
meet the HCE duties test but not the standard duties test may also 
become eligible for minimum wage and overtime protection. In Year 10, 
with automatic updating of the salary levels, the Department projects 
that between 5.1 and 5.6 million workers will be affected by the change 
in the standard salary level test and between 33,000 and 42,000 workers 
will be affected by the change in the HCE total annual compensation 
test, depending on the updating methodology used (CPI-U or fixed 
percentile of wage earnings, respectively). Additionally, the 
Department estimates that an additional 6.3 million white collar 
workers who are currently overtime eligible because they do not satisfy 
the EAP duties tests and who currently earn at least $455 per week but 
less than the proposed salary level would have their overtime 
protection strengthened in Year 1 because their exemption status would 
be clear based on the salary test alone without the need to examine 
their duties.
    Three direct costs to employers are quantified in this analysis: 
(1) Regulatory familiarization costs; (2) adjustment costs; and (3) 
managerial costs. Assuming a 7 percent discount rate, the Department 
estimates that average annualized direct employer costs will total 
between $239.6 and $255.3 million per year, depending on the updating 
methodology used as shown in (Table ES1). In addition to the direct 
costs, this proposed rulemaking will also transfer income from 
employers to employees in the form of higher earnings. Average 
annualized transfers are estimated to be between $1,178.0 and $1,271.4 
million, depending on which of the two updating methodologies analyzed 
in this proposal is used. The Department also projects average 
annualized deadweight loss of between $9.5 and $10.5 million, and notes 
that the projected deadweight loss is small in comparison to the amount 
of estimated costs.
    Impacts of the proposed rule extend beyond those quantitatively 
estimated. For example, a potential impact of the rule's proposed 
increase in the salary threshold is a reduction in litigation costs. 
Other unquantified transfers, costs, and benefits are discussed in 
section VII.D.vii.

                      Table ES1--Summary of Regulatory Costs and Transfers, Standard and HCE Salary Levels With Automatic Updating
                                                                    [Millions 2013$]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                 Future years \c\            Average annualized value
             Cost/Transfer \a\                Automatic updating method       Year 1     ---------------------------------------------------------------
                                                         \b\                                  Year 2          Year 10      3% Real rate    7% Real rate
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                Affected Workers (1,000s)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Standard..................................  Percentile..................           4,646           4,747           5,568              --              --
                                            CPI-U.......................           4,646           4,634           5,062              --              --
HCE.......................................  Percentile..................              36              36              42              --              --
                                            CPI-U.......................              36              35              33              --              --
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                          Costs and Transfers (Millions 2013$)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Direct employer costs.....................  Percentile..................           592.7           188.8           225.3           248.8           255.3
                                            CPI-U.......................           592.7           181.1           198.6           232.3           239.6
Transfers \d\.............................  Percentile..................         1,482.5         1,160.2         1,339.6         1,271.9         1,271.4
                                            CPI-U.......................         1,482.5         1,126.4         1,191.4         1,173.7         1,178.0
DWL.......................................  Percentile..................             7.4            10.8            11.2            10.5            10.5
                                            CPI-U.......................             7.4            10.3             9.7             9.6             9.5
--------------------------------------------------------------------------------------------------------------------------------------------------------
\a\ Costs and transfers for affected workers passing the standard and HCE tests are combined.
\b\ The percentile method sets the standard salary level at the 40th percentile of weekly earnings for full-time salaried workers and the HCE
  compensation level at the 90th percentile. The CPI-U method adjusts both levels based on the annual percent change in the CPI-U.
\c\ These costs/transfers represent a range over the nine-year span.

[[Page 38519]]

 
\d\ This is the net transfer from employers to workers. There may also be transfers of hours and income from some workers to other workers. Unquantified
  transfers, costs and benefits are addressed in Section VII.

    The Department believes that the proposed increase in the standard 
salary level to the 40th percentile of weekly earnings for full-time 
salaried workers and increasing the HCE compensation level to the 90th 
percentile of full-time salaried workers' earnings, combined with 
annual updating, is the simplest method for securing the effectiveness 
of the salary level as a bright-line for ensuring that employees 
entitled to the Act's overtime provisions are not exempted. The 
Department recognizes that the proposed standard salary threshold is 
lower than the historical average salary for the short duties test (the 
basis for the standard duties test) but believes that it will 
appropriately distinguish between overtime-eligible white collar 
salaried employees and those who may meet the EAP duties test without 
necessitating a return to the more rigorous long duties test. A 
standard salary threshold significantly below the 40th percentile, or 
the absence of a mechanism for automatically updating the salary level, 
however, would require a more rigorous duties test than the current 
standard duties test in order to effectively distinguish between white 
collar employees who are overtime protected and those who may be bona 
fide EAP employees. The Department believes that this proposal is the 
least burdensome but still cost-effective mechanism for updating the 
salary and compensation levels, and indexing future levels, and is 
consistent with the Department's statutory obligations.

II. Background

A. What the FLSA Provides

    The FLSA generally requires covered employers to pay their 
employees at least the federal minimum wage (currently $7.25 an hour) 
for all hours worked, and overtime premium pay of one and one-half 
times the employee's regular rate of pay for all hours worked over 40 
in a workweek.\4\ However, there are a number of exemptions from the 
FLSA's minimum wage and overtime requirements. Section 13(a)(1) of the 
FLSA, codified at 29 U.S.C. 213(a)(1), exempts from both minimum wage 
and overtime protection ``any employee employed in a bona fide 
executive, administrative, or professional capacity . . . or in the 
capacity of outside salesman (as such terms are defined and delimited 
from time to time by regulations of the Secretary, subject to the 
provisions of [the Administrative Procedure Act] . . .).'' The FLSA 
does not define the terms ``executive,'' ``administrative,'' 
``professional,'' or ``outside salesman.'' Pursuant to Congress' grant 
of rulemaking authority, the Department in 1938 issued the first 
regulations at 29 CFR part 541, defining the scope of the section 
13(a)(1) exemptions. Because Congress explicitly delegated to the 
Secretary of Labor the power to define and delimit the specific terms 
of the exemptions through notice and comment rulemaking, the 
regulations so issued have the binding effect of law. See Batterton v. 
Francis, 432 U.S. 416, 425 n.9 (1977).
---------------------------------------------------------------------------

    \4\ As discussed infra, the Department estimates that 128.5 
million workers are subject to the FLSA and the Department's 
regulations. Most of these workers are covered by the Act's minimum 
wage and overtime pay protections.
---------------------------------------------------------------------------

    The Department has consistently used its rulemaking authority to 
define and clarify the section 13(a)(1) exemptions. Since 1940, the 
implementing regulations have generally required each of three tests to 
be met for the exemptions to apply: (1) The employee must be paid a 
predetermined and fixed salary that is not subject to reduction because 
of variations in the quality or quantity of work performed (the 
``salary basis test''); (2) the amount of salary paid must meet a 
minimum specified amount (the ``salary level test''); and (3) the 
employee's job duties must primarily involve executive, administrative, 
or professional duties as defined by the regulations (the ``duties 
test'').

B. Legislative History

    Although section 13(a)(1) exempts covered employees from both the 
FLSA's minimum wage and overtime requirements, its most significant 
impact is its removal of these employees from the Act's overtime 
protections. It is widely recognized that the general requirement that 
employers pay a premium rate of pay for all hours worked over 40 in a 
workweek is a cornerstone of the Act, grounded in two policy 
objectives. The first is to spread employment by incentivizing 
employers to hire more employees rather than requiring existing 
employees to work longer hours, thereby reducing involuntary 
unemployment. See, e.g., Davis v. J.P. Morgan Chase, 587 F.3d 529, 535 
(2d Cir. 2009) (``The overtime requirements of the FLSA were meant to 
apply financial pressure to spread employment to avoid the extra wage 
and to assure workers additional pay to compensate them for the burden 
of a workweek beyond the hours fixed in the act.'') (internal quotation 
marks omitted). The second policy objective is to reduce overwork and 
its detrimental effect on the health and well-being of workers. See, 
e.g., Barrentine v. Arkansas-Best Freight System, Inc., 450 U.S. 728, 
739 (1981) (``The FLSA was designed to give specific minimum 
protections to individual workers and to ensure that each employee 
covered by the Act would receive a fair day's pay for a fair day's work 
and would be protected from the evil of overwork as well as 
underpay.'') (internal quotation marks and brackets omitted).
    Section 13(a)(1) was included in the original Act in 1938 and was 
based on provisions contained in the earlier National Industrial 
Recovery Act of 1933 (NIRA) and state law precedents. Specific 
references in the legislative history to the exemptions contained in 
section 13(a)(1) are scant. However, the exemptions were premised on 
the belief that the exempted workers typically earned salaries well 
above the minimum wage and were presumed to enjoy other privileges to 
compensate them for their long hours of work, such as above-average 
fringe benefits, greater job security, and better opportunities for 
advancement, setting them apart from the nonexempt workers entitled to 
overtime pay. See Report of the Minimum Wage Study Commission, Volume 
IV, pp. 236 and 240 (June 1981).\5\ Further, the type of work exempt 
employees performed was difficult to standardize to any time frame and 
could not be easily spread to other workers after 40 hours in a week, 
making enforcement of the overtime provisions difficult and generally 
precluding the potential job expansion intended by the FLSA's time-and-
a-half overtime premium. Id.
---------------------------------------------------------------------------

    \5\ Congress created the Minimum Wage Study Commission as part 
of the Fair Labor Standards Amendments of 1977. See Sec. 2(e)(1), 
Public Law 95-151, 91 Stat. 1246 (Nov. 1, 1977). This independent 
commission was tasked with examining many FLSA issues, including the 
Act's minimum wage and overtime exemptions, and issuing a report to 
the President and to Congress with the results of its study.
---------------------------------------------------------------------------

    The universe of employees eligible for the exemptions has 
fluctuated with amendments to the FLSA. Initially, persons employed in 
a ``local retailing capacity'' were exempt, but Congress eliminated 
that language from section 13(a)(1) in 1961 when the FLSA was expanded 
to cover retail and service enterprises. See Public Law 87-30, 75 Stat. 
65 (May 5, 1961). Teachers and

[[Page 38520]]

academic administrative personnel were added to the exemption when 
elementary and secondary schools were made subject to the FLSA in 1966. 
Sec. 214, Public Law 89-601, 80 Stat. 830 (Sept. 23, 1966). The 
Education Amendments of 1972 made the Equal Pay provisions, section 
6(d) of the FLSA, expressly applicable to employees who were otherwise 
exempt from the FLSA under section 13(a)(1). Sec. 906(b)(1), Public Law 
92-318, 86 Stat. 235 (June 23, 1972).
    A 1990 enactment expanded the exemptions to include in the 
regulations defining exempt executive, administrative, and professional 
employees, computer systems analysts, computer programmers, software 
engineers, and similarly skilled professional workers, including those 
paid on an hourly basis if paid at least 6\1/2\ times the minimum wage. 
Sec. 2, Public Law 101-583, 104 Stat. 2871 (Nov. 15, 1990). The 
compensation test for computer-related occupations was subsequently 
capped at $27.63 an hour (6\1/2\ times the minimum wage in effect at 
the time) as part of the 1996 FLSA Amendments, when Congress enacted 
the new section 13(a)(17) exemption for such computer employees. 
Section 13(a)(17) also incorporated much of the regulatory language 
that resulted from the 1990 enactment. See 29 U.S.C. 213(a)(17), as 
added by the 1996 FLSA Amendments (Sec. 2105(a), Public Law 104-188, 
110 Stat. 1755 (Aug. 20, 1996)).

C. Regulatory History

    The FLSA became law on June 25, 1938, and the first version of part 
541, setting forth the criteria for exempt status under section 
13(a)(1), was issued that October. 3 FR 2518 (Oct. 20, 1938). Following 
a series of public hearings, which were discussed in a report issued by 
WHD,\6\ the Department published revised regulations in 1940, which, 
among other things, updated and expanded the salary level test. 5 FR 
4077 (Oct. 15, 1940). Further hearings were convened in 1947, as 
discussed in a WHD-issued report,\7\ and revised regulations, which 
updated the salary levels required to meet the salary level test for 
the various exemptions, were issued in 1949. 14 FR 7705 (Dec. 24, 
1949). An explanatory bulletin interpreting some of the terms used in 
the regulations was published as subpart B of part 541 in 1949. 14 FR 
7730 (Dec. 28, 1949). In 1954, the Department issued revisions to the 
regulatory interpretations of the salary basis test. 19 FR 4405 (July 
17, 1954). In 1958, based on another WHD-issued report,\8\ the 
regulations were revised to update the required salary levels. 23 FR 
8962 (Nov. 18, 1958). Additional changes, including periodic salary 
level updates, were made to the regulations in 1961 (26 FR 8635, Sept. 
15, 1961), 1963 (28 FR 9505, Aug. 30, 1963), 1967 (32 FR 7823, May 30, 
1967), 1970 (35 FR 883, Jan. 22, 1970), 1973 (38 FR 11390, May 7, 
1973), and 1975 (40 FR 7091, Feb. 19, 1975). Revisions to increase the 
salary levels in 1981 were stayed indefinitely by the Department. 46 FR 
11972 (Feb. 12, 1981). In 1985, the Department published an Advance 
Notice of Proposed Rulemaking that reopened the comment period on the 
1981 proposal and broadened the review to all aspects of the 
regulations, including whether to increase the salary levels, but this 
rulemaking was never finalized. 50 FR 47696 (Nov. 19, 1985).
---------------------------------------------------------------------------

    \6\ Executive, Administrative, Professional . . . Outside 
Salesman Redefined, Wage and Hour Division, U.S. Department of 
Labor, Report and Recommendations of the Presiding Officer (Harold 
Stein) at Hearings Preliminary to Redefinition (Oct. 10, 1940) 
(``Stein Report'').
    \7\ Report and Recommendations on Proposed Revisions of 
Regulations, Part 541, by Harry Weiss, Presiding Officer, Wage and 
Hour and Public Contracts Divisions, U.S. Department of Labor (June 
30, 1949) (``Weiss Report'').
    \8\ Report and Recommendations on Proposed Revision of 
Regulations, Part 541, Under the Fair Labor Standards Act, by Harry 
S. Kantor, Presiding Officer, Wage and Hour and Public Contracts 
Divisions, U.S. Department of Labor (Mar. 3, 1958) (``Kantor 
Report'').
---------------------------------------------------------------------------

    The Department revised the part 541 regulations twice in 1992. 
First, the Department created a limited exception from the salary basis 
test for public employees, permitting public employers to follow public 
sector pay and leave systems requiring partial-day deductions from pay 
for absences for personal reasons or due to illness or injury not 
covered by accrued paid leave, or due to budget-driven furloughs, 
without defeating the salary basis test required for exemption. 57 FR 
37677 (Aug. 19, 1992). The Department also implemented the 1990 law 
requiring it to promulgate regulations permitting employees in certain 
computer-related occupations to qualify as exempt under section 
13(a)(1) of the FLSA. 57 FR 46744 (Oct. 9, 1992); see Sec. 2, Public 
Law 101-583, 104 Stat. 2871 (Nov. 15, 1990).
    On March 31, 2003, the Department published a Notice of Proposed 
Rulemaking proposing significant changes to the part 541 regulations. 
68 FR 15560 (Mar. 31, 2003). On April 23, 2004, the Department issued a 
Final Rule (2004 Final Rule), which raised the salary level for the 
first time since 1975, and made other changes, some of which are 
discussed below. 69 FR 22122 (Apr. 23, 2004). Current regulations 
retain the three tests for exempt status that have been in effect since 
1940: A salary basis test, a salary level test, and a job duties test.

D. Overview of Existing Regulatory Requirements

    The regulations in part 541 contain specific criteria that define 
each category of exemption provided by section 13(a)(1) for bona fide 
executive, administrative, professional, outside sales employees, and 
teachers and academic administrative personnel. The regulations also 
define those computer employees who are exempt under section 13(a)(1) 
and section 13(a)(17). See Sec. Sec.  541.400-.402. The employer bears 
the burden of establishing the applicability of any exemption from the 
FLSA's pay requirements. Job titles and job descriptions do not 
determine exempt status, nor does paying a salary rather than an hourly 
rate. To qualify for the EAP exemption, employees must meet certain 
tests regarding their job duties and generally must be paid on a salary 
basis of not less than $455 per week.\9\ In order for the exemption to 
apply, an employee's specific job duties and salary must meet all the 
requirements of the Department's regulations. The duties tests differ 
for each category of exemption.
---------------------------------------------------------------------------

    \9\ Alternatively, administrative and professional employees may 
be paid on a ``fee basis.'' This occurs where an employee is paid an 
agreed sum for a single job regardless of the time required for its 
completion. Sec.  541.605(a). Salary level test compliance for fee 
basis employees is assessed by determining whether the hourly rate 
for work performed (i.e., the fee payment divided by the number of 
hours worked) would total at least $455 per week if the employee 
worked 40 hours. See Sec.  541.605(b). Some employees, such as 
doctors and lawyers (Sec.  541.600(e)), teachers (Sec.  541.303(d); 
Sec.  541.600(e)), and outside sales employees (Sec.  541.500(c)), 
are not subject to a salary or fee basis test. Some, such as 
academic administrative personnel, are subject to a special, 
contingent salary level. See Sec.  541.600(c). There is also a 
separate salary level in effect for workers in American Samoa (Sec.  
541.600(a)), and a special salary test for motion picture industry 
employees (Sec.  541.709).
---------------------------------------------------------------------------

    The Department last updated the salary levels in the 2004 Final 
Rule, setting the standard test threshold at $455 per week for 
executive, administrative, and professional employees. Since its prior 
revision in 1975, the salary level tests had grown outdated and were 
thus no longer effective at distinguishing between exempt and nonexempt 
employees. Mindful that nearly 30 years had elapsed between salary 
level increases, and in response to commenter concerns that similar 
lapses would occur in the future, in the 2004 Final Rule the Department 
expressed the intent to

[[Page 38521]]

``update the salary levels on a more regular basis.'' 69 FR 22171.
    Under the current part 541 regulations, an exempt executive 
employee must be compensated on a salary basis at a rate of not less 
than $455 per week and have a primary duty of managing the enterprise 
or a department or subdivision of the enterprise. Sec.  541.100(a)(1)-
(2). An exempt executive must also customarily and regularly direct the 
work of at least two employees and have the authority to hire or fire, 
or the employee's suggestions and recommendations as to the hiring, 
firing, or other change of status of employees must be given particular 
weight. Sec.  541.100(a)(3)-(4).
    An exempt administrative employee must be compensated on a salary 
or fee basis at a rate of not less than $455 per week and have a 
primary duty of the performance of office or non-manual work directly 
related to the management or general business operations of the 
employer or the employer's customers. Sec.  541.200. An exempt 
administrative employee's primary duty must include the exercise of 
discretion and independent judgment with respect to matters of 
significance. Id.
    An exempt professional employee must be compensated on a salary or 
fee basis at a rate of not less than $455 per week and have a primary 
duty of (1) work requiring knowledge of an advanced type in a field of 
science or learning customarily acquired by prolonged, specialized, 
intellectual instruction and study, or (2) work that is original and 
creative in a recognized field of artistic endeavor, or (3) teaching in 
a school system or educational institution, or (4) work as a computer 
systems analyst, computer programmer, software engineer, or other 
similarly-skilled worker in the computer field. Sec. Sec.  541.300; 
541.303; 541.400. An exempt professional employee must perform work 
requiring the consistent exercise of discretion and judgment, or 
requiring invention, imagination, or talent in a recognized field of 
artistic endeavor. Sec.  541.300(a)(2). The salary requirements do not 
apply to certain licensed or certified doctors, lawyers, and teachers. 
Sec. Sec.  541.303(d); 541.304(d).
    An exempt outside salesperson must be customarily and regularly 
engaged away from the employer's place of business and have a primary 
duty of making sales, or obtaining orders or contracts for services or 
for the use of facilities. Sec.  541.500. There are no salary or fee 
requirements for exempt outside sales employees. Id.
    The 2004 Final Rule created a new ``highly compensated'' test for 
exemption. Under the HCE exemption, employees who are paid total annual 
compensation of at least $100,000 (which must include at least $455 per 
week paid on a salary or fee basis) are exempt from the FLSA's overtime 
requirements if they customarily and regularly perform at least one of 
the exempt duties or responsibilities of an executive, administrative, 
or professional employee identified in the standard tests for 
exemption. Sec.  541.601. The HCE exemption applies only to employees 
whose primary duty includes performing office or non-manual work; non-
management production line workers and employees who perform work 
involving repetitive operations with their hands, physical skill, and 
energy are not exempt under this section no matter how highly paid. Id.
    Employees who meet the requirements of part 541 are excluded from 
both the Act's minimum wage and overtime pay protections. As a result, 
employees may work any number of hours in the workweek and not be 
subject to the FLSA's minimum wage and overtime pay requirements. Some 
state laws have stricter exemption standards than those described 
above. The FLSA does not preempt any such stricter state standards. If 
a State establishes a higher standard than the provisions of the FLSA, 
the higher standard applies in that State. See 29 U.S.C. 218.

III. Presidential Memorandum

    On March 13, 2014, President Obama signed a Presidential Memorandum 
directing the Department to update the regulations defining which 
``white collar'' workers are protected by the FLSA's minimum wage and 
overtime standards. 79 FR 18737 (Apr. 3, 2014). The memorandum 
instructed the Department to look for ways to modernize and simplify 
the regulations while ensuring that the FLSA's intended overtime 
protections are fully implemented. As the President noted at the time, 
the FLSA's overtime protections are a linchpin of the middle class and 
the failure to keep the salary level requirement for the white collar 
exemption up-to-date has left millions of low-paid salaried workers 
without this basic protection.\10\ The current salary level threshold 
for exemption of $455 per week, or $23,660 annually, is below the 
poverty threshold for a family of four.\11\
---------------------------------------------------------------------------

    \10\ https://www.whitehouse.gov/the-press-office/2014/03/13/fact-sheet-opportunity-all-rewarding-hard-work-strengthening-overtime-pr.
    \11\ See https://www.census.gov/hhes/www/poverty/data/threshld/. The current salary level is less than the 10th 
percentile of full-time salaried workers.
---------------------------------------------------------------------------

    Following issuance of the memorandum, the Department embarked on an 
extensive outreach program, conducting listening sessions in 
Washington, DC, and several other locations, as well as by conference 
call. The listening sessions were attended by a wide range of 
stakeholders: Employees, employers, business associations, non-profit 
organizations, employee advocates, unions, state and local government 
representatives, tribal representatives, and small businesses. In these 
sessions the Department asked stakeholders to address, among other 
issues: (1) What is the appropriate salary level for exemption; (2) 
what, if any, changes should be made to the duties tests; and (3) how 
can the regulations be simplified.
    Stakeholders representing employers expressed a wide variety of 
views on the appropriate salary level, ranging from a few who said the 
salary should not be raised, to several who noted their entry level 
managers already earned salaries far above the current annual salary 
level of $23,660. A number of representatives of national employers 
also noted regional variations in the salary levels they pay to EAP 
employees. Several employers encouraged the Department to consider 
nondiscretionary bonuses in determining whether the salary level is 
met, noting that such bonuses are a key part of exempt employees' 
compensation in their industries and contribute to an ``ownership 
mindset.'' Many employer stakeholders stated that they consider first-
line managerial positions to be the gateway to developing their future 
senior managers and organizational leadership. A number of these 
employer stakeholders also raised concerns about changing currently 
exempt employees to nonexempt employees as a result of an increase in 
the salary requirement, stating that employees are attached to the 
perceived higher status of being in exempt salaried positions, and 
value the time flexibility and steady income that comes with such 
positions. These stakeholders also stressed the need for flexibility 
under the regulations, in particular emphasizing the value they place 
on a work culture that encourages managers to lead by example and 
``pitch in'' to assist nonexempt employees. They stressed that changing 
the duties tests to limit exempt employees' ability to perform 
nonexempt work--such as California's 50 percent primary duty rule--
would negatively impact the culture of the workplace, be difficult and 
costly to implement, and lead to increased litigation. They also noted 
the significant investment they made in

[[Page 38522]]

reviewing employee classifications as a result of the 2004 Final Rule 
to determine whether employees met the revised duties tests. Finally, 
several employer representatives suggested that adding to the 
regulations additional examples of how the exemptions may apply to 
specific occupations would simplify employers' determinations of EAP 
exemption status.
    Stakeholders representing employees universally endorsed the need 
to increase the salary level, noting that it has not been updated since 
2004. Several employee advocates also stressed the need to index the 
salary level to ensure that it maintains its effectiveness as a 
demarcation line between exempt and overtime-eligible employees without 
having to rely on time consuming future rulemaking. Both individual 
employees and their representatives shared their concerns that some 
employers are taking advantage of exempt employees, requiring them to 
perform large amounts of routine work in order to keep down labor 
costs, and a few suggested that there needs to be a maximum hours cap 
for EAP exempt employees. They stressed that employees in 
``management'' positions who are required to spend disproportionate 
amounts of time performing routine nonexempt tasks (ringing up 
customers, stocking shelves, bussing tables, cleaning stores and 
restaurants, etc., alongside or in place of front line workers) are not 
bona fide executives and do not, in fact, enjoy the flexibility and 
status traditionally associated with such positions and therefore are 
entitled to the overtime protections the FLSA was designed to provide. 
Employee advocates pointed to the California overtime rule as more 
protective of such workers.
    While the HCE exemption was not a primary focus of any of the 
listening sessions, a number of business stakeholders stated that the 
$100,000 total annual compensation requirement was too high, and a few 
suggested that the duties test for the HCE exemption should be dropped 
and the exemption should be based on compensation level alone. In 
contrast, the employee stakeholders who addressed the issue argued that 
the HCE duties test was too lax and that the $100,000 total annual 
compensation requirement was too low, particularly in light of the wage 
gains at the top end of the earnings spectrum since 2004. Some employee 
advocates suggested eliminating the HCE exemption. While the outside 
sales exemption was also not a central focus of the sessions, several 
stakeholders representing employer interests argued that the 
distinction between inside and outside sales positions in the 
application of the EAP exemption does not reflect the realities of the 
modern workplace.\12\
---------------------------------------------------------------------------

    \12\ Section 13(a)(1) expressly includes within the EAP 
exemption ``any employee employed . . . in the capacity of outside 
salesman.'' 29 U.S.C. 213(a)(1). As discussed in the 2004 Final 
Rule, ``the Administrator does not have statutory authority to 
exempt inside sales employees from the FLSA minimum wage and 
overtime requirements under the outside sales exemption.'' 69 FR 
22162.
---------------------------------------------------------------------------

    The Department's outreach has made clear that there are also some 
widespread misconceptions about overtime eligibility under the FLSA. 
For example, many employers and employees mistakenly believe that 
payment of a salary automatically disqualifies an employee from 
entitlement to overtime compensation irrespective of the duties 
performed. Many employees are also unaware of the duties required to be 
performed in order for the exemption to apply. Additionally, many 
employers seem to mistakenly believe that nonexempt white collar 
employees must be converted to hourly compensation. Similarly, other 
employers erroneously believe that they are prohibited from paying 
nondiscretionary bonuses to EAP employees, given that they cannot be 
used to satisfy the salary requirement. Some employers also mistakenly 
believe that the EAP regulations limit their ability to permit white 
collar employees to work part-time or job share.\13\ The Department 
believes that many of these misconceptions can be addressed through its 
education and outreach efforts.\14\
---------------------------------------------------------------------------

    \13\ As the Department has previously explained, there is no 
special salary level for EAP employees working less than full-time. 
69 FR 22171. Employers, however, can pay white collar employees 
working part-time or job sharing a salary of less than the required 
EAP salary threshold and will not violate the Act so long as the 
salary equals at least the minimum wage for all hours worked and the 
employee does not work more than 40 hours a week. FLSA2008-1NA (Feb. 
14, 2008).
    \14\ Such misconceptions are not new. In 1940 the Department 
responded to the related argument that employers would convert 
overtime-eligible white collar employees to hourly pay instead of 
more secure salaries, stating: ``Without underestimating the general 
desirability of weekly or monthly salaries which enable employees to 
adjust their expenditures on the basis of an assured income (so long 
as they remain employed), there is little advantage in salaried 
employment if it serves merely as a cloak for long hours of work. 
Further, such salaried employment may well conceal excessively low 
hourly rates of pay.'' Stein Report at 7.
---------------------------------------------------------------------------

    Lastly, the Department notes that multiple stakeholders on both 
sides of the issue expressed frustration with the exempt/nonexempt 
terminology and asked the Department to consider more descriptive 
terms. The Department recognizes that the terms ``exempt'' and 
``nonexempt'' are not intuitive and can be confusing to both employers 
and employees. In an attempt to address this concern, the Department 
uses the terms ``overtime protected'' and ``overtime eligible'' at 
times in this NPRM as synonyms for nonexempt, and ``not overtime 
protected'' and ``overtime ineligible'' as synonyms for exempt. While 
the Department will continue to use the terms exempt and nonexempt as 
technical terms to ensure accuracy and continuity, we will, where 
appropriate, endeavor to use these more descriptive terms to aid the 
regulated community. The Department also uses the term ``EAP 
exemption'' throughout this NPRM to reflect the section 13(a)(1) 
exemption for executive, administrative, and professional employees.
    The discussions in the listening sessions have informed not just 
the development of this NPRM, but also the Department's understanding 
of the role of overtime in the modern workplace. Some of the issues 
raised in the listening sessions are specifically referenced below in 
the Department's proposals; some issues that were raised are either 
beyond the scope of this rulemaking or beyond the Department's 
authority under the FLSA. For example, several employers expressed 
concern that employees who would become newly entitled to overtime 
under a higher salary level requirement would lose the flexibility they 
currently enjoy to work remotely on electronic devices because of 
employer concerns about overtime liability. Because this concern 
involves compensation for hours worked by overtime-protected employees, 
it is beyond the scope of this rulemaking. The Department, however, 
understands the importance of this concern and will publish a Request 
for Information in the near future seeking information from 
stakeholders on the use of electronic devices by overtime-protected 
employees outside of scheduled work hours.
    The Department appreciates the views of all the participants in the 
listening sessions and welcomes further input from the public in 
response to this NPRM. Finally, consistent with the President's 
commitment to a 21st-century regulatory system, the Department would 
consider conducting a retrospective review of the Final Rule resulting 
from this proposal at an appropriate time in the future.

IV. Need for Rulemaking

    One of the Department's primary goals in this rulemaking is 
updating the section 13(a)(1) exemption's salary level requirement. A 
salary level test has been part of the regulations since 1938 and

[[Page 38523]]

has been long recognized as ``the best single test'' of exempt status. 
Stein Report at 19, 42; see Weiss Report at 8-9; Kantor Report at 2-3. 
The salary an employer pays an employee provides ``a valuable and 
easily applied index to the `bona fide' character of the employment for 
which exemption is claimed'' and ensures that section 13(a)(1) of the 
FLSA ``will not invite evasion of section 6 and section 7 for large 
numbers of workers to whom the wage-and-hour provisions should apply.'' 
Stein Report at 19. The 1949 Weiss Report's statement remains true 
today: ``The experience of [the Department] since 1940 supports the 
soundness of the inclusion of the salary criteria in the regulations.'' 
Weiss Report at 8. In setting the salary level for the long test (which 
paired a lower salary with a limitation on the amount of non-exempt 
work an exempt worker could perform) the Department sought to provide a 
ready guide to assist employers in identifying employees who were 
unlikely to meet the duties tests for the exemptions.
    The salary level's function in differentiating exempt from 
nonexempt employees takes on greater importance when there is only one 
duties test that has no limitation on the amount of nonexempt work that 
an exempt employee may perform, as has been the case since 2004. The 
Department set the standard salary level in 2004 equivalent to the 
former long test salary level, thus not adjusting the salary threshold 
to account for the absence of the more rigorous long duties test. The 
long test salary level was designed to operate as a ready guide to 
assist employers in identifying employees who were unlikely to meet the 
duties tests for the EAP exemption. The salary level required for 
exemption under section 13(a)(1) is currently $455 a week and has not 
been updated in more than 10 years. The annual value of the salary 
level ($23,660) is now lower than the poverty threshold for a family of 
four. If left at the same amount, the effectiveness of the salary level 
test as a means of helping determine exempt status diminishes as the 
wages of employees entitled to overtime pay increase and the real value 
of the salary threshold falls.
    By way of this rulemaking, the Department seeks to update the 
salary level to ensure that the FLSA's intended overtime protections 
are fully implemented, and to simplify the identification of overtime-
eligible employees, thus making the exemptions easier for employers and 
workers to understand. For similar reasons, the Department also 
proposes to update the total annual compensation required for the HCE 
exemption, since it too has been unchanged since 2004, and the current 
level could lead to inappropriate classification given the minimal 
duties test for that exemption.
    In a further effort to respond to changing conditions in the 
workplace, the Department is also considering whether to allow 
nondiscretionary bonuses to satisfy some portion of the standard test 
salary requirement. Currently, such bonuses are only included in 
calculating total annual compensation under the HCE test, but some 
stakeholders have urged broader inclusion, pointing out that in some 
industries, particularly the retail and restaurant industries, 
significant portions of salaried EAP employees' earnings may be in the 
form of such bonuses.
    The Department also proposes automatically updating the salary 
levels based on changes in the economy to prevent the levels from 
becoming outdated with the often lengthy passage of time between 
rulemakings. The Department proposes to automatically update the 
standard salary test, the annual compensation requirement for highly 
compensated employees, and the special salary levels for American Samoa 
and for motion picture industry employees, in order to ensure the 
continued utility of these tests over time. As explained in the Weiss 
Report, the salary test is only a strong measure of exempt status if it 
is up to date, and a weakness of the salary test is that increases in 
wage rates and salary levels over time gradually diminish its 
effectiveness. See Weiss Report at 8. In the 1970 rulemaking, in 
response to a comment requesting that the regulations provide for 
annual review and updating of the salary level, the Department noted 
that the idea ``appears to have some merit particularly since past 
practice has indicated that approximately 7 years elapse between 
amendment of these salary requirements,'' but concluded that such a 
proposal would require further study. 35 FR 884. In the 2004 Final 
Rule, the Department declined to adopt a process for automatically 
updating the salary level and instead stated our intent ``in the future 
to update the salary levels on a more regular basis'' as we did prior 
to 1975. Yet competing regulatory priorities, overall agency workload, 
and the time-intensive nature of the notice and comment process have 
hindered the Department's ability to achieve this goal, which would 
require nearly continuous future rulemaking. A rule providing for 
automatic updates to the salary level using a methodology that has been 
subject to notice and comment rulemaking would maintain the utility of 
the dividing line set by the salary level without the need for frequent 
rulemaking. This modernization of the regulations would provide 
predictability for employers and employees by replacing infrequent, and 
thus more drastic, salary level increases with gradual changes 
occurring at set intervals. Regular annual increases in the salary and 
compensation levels, instead of large changes that result from sporadic 
rulemaking, will provide more certainty and stability for employers.
    The Department is also considering revisions to the duties tests in 
order to ensure that they fully reflect the purpose of the exemption. 
Possible revisions include requiring overtime-ineligible employees to 
spend a specified amount of time performing their primary duty (e.g., a 
50 percent primary duty requirement as required under California state 
law) or otherwise limiting the amount of nonexempt work an overtime-
ineligible employee may perform, and adding to the regulations 
additional examples illustrating how the exemption may apply to 
particular occupations. As previously discussed, during listening 
sessions held in advance of this proposed rule, the Department asked 
stakeholders what, if any, changes should be made to the existing 
duties tests for exemption. Stakeholders from the business community, 
while noting the uncertainty caused by litigation surrounding their 
application of the current duties tests, generally advocated for no 
changes to the current duties tests and raised specific concerns about 
the difficulty of imposing any limit on the amount of nonexempt work 
that exempt employees may perform. These stakeholders indicated that 
the uncertainty which would result from any changes in the duties tests 
would be much more problematic than the challenges encountered with the 
current tests. Employees and stakeholders representing employee 
interests, however, generally advocated for stricter requirements to 
ensure that overtime-ineligible employees spend a sufficient amount of 
time performing exempt duties, and do not spend excessive amounts of 
time on nonexempt work. These stakeholders argued that such 
requirements would clarify the application of the exemption and restore 
overtime protection to employees whose duties are not, in fact, those 
of a bona fide executive, administrative, or professional employee. 
Several business stakeholders also suggested that adding additional 
examples of how the exemptions apply

[[Page 38524]]

to particular occupations would simplify application of the exemption 
for employers and increase the clarity of the current duties tests.

V. Proposed Regulatory Revisions

    The Department's current proposal focuses primarily on updating the 
salary and compensation levels by proposing that the standard salary 
level be set at the 40th percentile of weekly earnings for full-time 
salaried workers, proposing to increase the HCE annual compensation 
requirement to the annualized value of the 90th percentile of weekly 
earnings of full-time salaried workers, and proposing a mechanism for 
automatically updating the salary and compensation levels going forward 
to ensure that they will continue to provide a useful and effective 
test for exemption. While the primary regulatory changes proposed are 
in Sec. Sec.  541.600 and 541.601, additional conforming changes are 
proposed to update references to the salary level throughout part 541 
as well as to update the special salary provisions for American Samoa 
and the motion picture industry. The proposal also discusses the 
inclusion of nondiscretionary bonuses to satisfy a portion of the 
standard salary requirement but does not propose specific regulatory 
changes. Additionally, the proposal discusses the duties tests, 
requests comments on the current requirements, and solicits suggestions 
for additional occupation examples, but does not make any specific 
proposals for revisions to these sections.

A. Setting the Standard Salary Level

i. History
    The FLSA became law on June 25, 1938, and the first version of part 
541, issued later that year, set a minimum salary level of $30 per week 
for executive and administrative employees. 3 FR 2518. Since 1938, the 
Department has increased the salary levels seven times--in 1940, 1949, 
1958, 1963, 1970, 1975, and 2004. See Table A. While the Department's 
method for calculating the salary level has evolved to fulfill its 
mandate, the purpose of the salary level requirement has remained 
consistent--to define and delimit the scope of the executive, 
administrative, and professional exemptions. 29 U.S.C. 213(a)(1). The 
Department has long recognized that the salary paid to an employee is 
the ``best single test'' of exempt status (Stein Report at 19) and that 
setting a minimum salary threshold provides a ``ready method of 
screening out the obviously nonexempt employees'' while furnishing a 
``completely objective and precise measure which is not subject to 
differences of opinion or variations in judgment.'' Weiss Report at 8-
9. The Department reaffirmed this position in the 2004 Final Rule, 
explaining that the ``salary level test is intended to help distinguish 
bona fide executive, administrative, and professional employees from 
those who were not intended by Congress to come within these exempt 
categories[,]'' and reiterating that any increase in the salary level 
must ``have as its primary objective the drawing of a line separating 
exempt from nonexempt employees.'' 69 FR 22165.

                                   Table A--Weekly Salary Levels for Exemption
----------------------------------------------------------------------------------------------------------------
                                                                    Long test
                 Date enacted                  --------------------------------------------------   Short test
                                                   Executive     Administrative    Professional        (all)
----------------------------------------------------------------------------------------------------------------
1938..........................................             $30               $30  ..............  ..............
1940..........................................              30                50             $50  ..............
1949..........................................              55                75              75            $100
1958..........................................              80                95              95             125
1963..........................................             100               100             115             150
1970..........................................             125               125             140             200
1975..........................................             155               155             170             250
----------------------------------------------------------------------------------------------------------------
                                                  Standard Test
----------------------------------------------------------------------------------------------------------------
2004..........................................                                $455
----------------------------------------------------------------------------------------------------------------

    In 1940, the Department maintained the $30 per week salary level 
set in 1938 for executive employees, increased the salary level for 
administrative employees, and established a salary level for 
professional employees. The Department used salary surveys from federal 
and state government agencies, experience gained under the National 
Industrial Recovery Act, and federal government salaries to determine 
the salary level that was the ``dividing line'' between employees 
performing exempt and nonexempt work. Stein Report at 9, 20-21, 31-32. 
The Department recognized that the salary level falls within a 
continuum of salaries that overlaps the outer boundaries of exempt and 
nonexempt employees. Specifically, the Department stated:

    To make enforcement possible and to provide for equity in 
competition, a rate should be selected in each of the three 
definitions which will be reasonable in the light of average 
conditions for industry as a whole. In some instances the rate 
selected will inevitably deny exemption to a few employees who might 
not unreasonably be exempted, but, conversely, in other instances it 
will undoubtedly permit the exemption of some persons who should 
properly be entitled to the benefits of the act.

Id. at 6. Taking into account the average salary levels for employees 
in numerous industries, and the percentage of employees earning below 
these amounts, the Department set the salary level for each exemption 
slightly below the ``dividing line'' suggested by these averages.
    In 1949, the Department again looked at salary data from state and 
federal agencies, including the Bureau of Labor Statistics (BLS). The 
data reviewed included wages in small towns and low-wage industries, 
earnings of federal employees, average weekly earnings for exempt 
employees, starting salaries for college graduates, and salary ranges 
for different occupations such as bookkeepers, accountants, chemists, 
and mining engineers. Weiss Report at 10, 14-17, 19-20. The Department 
noted that the ``salary level adopted must exclude the great bulk of 
nonexempt persons if it is to be effective''. Id. at 18. Recognizing 
that the ``increase in wage rates and salary levels'' since 1940 had 
``gradually weakened the effectiveness of the present salary tests as a 
dividing line between exempt and nonexempt employees,'' the Department 
calculated the percentage increase in weekly

[[Page 38525]]

earnings from 1940 to 1949, and then adopted new salary levels ``at a 
figure slightly lower than might be indicated by the data'' in order to 
protect small businesses. Id. at 8, 14. The Department also cautioned 
that ``a dividing line cannot be drawn with great precision but can at 
best be only approximate.'' Id. at 11
    In 1949, the Department also established a second, less-stringent 
duties test for each exemption, but only for those employees who were 
paid at or above a higher ``short test'' salary level. Those paid above 
the higher salary level were exempt if they also met a ``short'' duties 
test, which lessened the duties requirements for exemption.\15\ The 
rationale for this short test was that employees who met the higher 
salary level were more likely to meet all the requirements for 
exemption, and thus a ``short-cut test for exemption . . . would 
facilitate the administration of the regulations without defeating the 
purposes of section 13(a)(1).'' Id. at 22-23. Employees who met only 
the lower ``long test'' salary level, and not the higher short test 
salary level, were still required to satisfy the default ``long'' 
duties test, which included a 20 percent limitation on the amount of 
nonexempt work that could be performed by an exempt employee. While the 
long test salary level was set based on an analysis of the defined 
sample, the short test salary level was set in relation to the long 
test salary. The existence of separate short and long tests--with short 
test salary levels ranging from approximately 130 to 180 percent of the 
long test salary levels--remained part of the Department's regulations 
until 2004.\16\ See Table A.
---------------------------------------------------------------------------

    \15\ These higher salary levels are presented under the ``Short 
Test'' heading in Table A.
    \16\ The smallest ratio was in 1963 between the long test salary 
requirement for professionals ($115) and the short test salary level 
($150). The largest ratio was in 1949 between the long test salary 
requirement for executives ($55) and the short test salary level 
($100).
---------------------------------------------------------------------------

    In setting the long test salary level in 1958, the Department 
considered data collected during 1955 WHD investigations on the 
``actual salaries paid'' to employees who ``qualified for exemption'' 
(i.e., met the applicable salary and duties tests), grouped by 
geographic region, broad industry groups, number of employees, and city 
size, and supplemented with BLS and Census data to reflect income 
increases of white collar and manufacturing employees during the period 
not covered by the Department's investigations. Kantor Report at 6. The 
Department then set the salary level tests for exempt employees ``at 
about the levels at which no more than about 10 percent of those in the 
lowest-wage region, or in the smallest size establishment group, or in 
the smallest-sized city group, or in the lowest-wage industry of each 
of the categories would fail to meet the tests.'' Id. at 6-7. In other 
words, the Department set the salary level so that only a limited 
number of workers performing EAP duties (about 10 percent) in the 
lowest-wage regions and industries would fail to meet the salary level 
test and therefore be overtime protected. In laying out this 
methodology, the Department echoed comments from the Weiss Report that 
the salary tests ``simplify enforcement by providing a ready method of 
screening out the obviously nonexempt employees[,]'' and that 
``[e]mployees that do not meet the salary test are generally also found 
not to meet the other requirements of the regulations.'' Id. at 2-3. 
The Department also noted that in our experience misclassification of 
overtime-protected employees occurs more frequently when the salary 
levels have ``become outdated by a marked upward movement of wages and 
salaries.'' Id. at 5.
    The Department followed a similar methodology when determining the 
appropriate long test salary level increase in 1963, using data 
regarding salaries paid to exempt workers collected in a 1961 WHD 
survey. 28 FR 7002. The salary level for executive and administrative 
employees was increased to $100 per week, for example, when the 1961 
survey data showed that 13 percent of establishments paid one or more 
exempt executives less than $100 per week, and 4 percent of 
establishments paid one or more exempt administrative employees less 
than $100 a week. 28 FR 7004. The professional exemption salary level 
was increased to $115 per week, when the 1961 survey data showed that 
12 percent of establishments surveyed paid one or more professional 
employees less than $115 per week. Id. The Department noted that these 
salary levels approximated the same percentages used in 1958:

    Salary tests set at this level would bear approximately the same 
relationship to the minimum salaries reflected in the 1961 survey 
data as the tests adopted in 1958, on the occasion of the last 
previous adjustment, bore to the minimum salaries reflected in a 
comparable survey, adjusted by trend data to early 1958. At that 
time, 10 percent of the establishments employing executive employees 
paid one or more executive employees less than the minimum salary 
adopted for executive employees and 15 percent of the establishments 
employing administrative or professional employees paid one or more 
employees employed in such capacities less than the minimum salary 
adopted for administrative and professional employees.

Id.
    The Department continued to use a similar methodology when updating 
the long test salary level in 1970. After examining data from 1968 WHD 
investigations, 1969 BLS wage data, and information provided in a 
report issued by the Department in 1969 that included salary data for 
executive, administrative, and professional employees,\17\ the 
Department increased the long test salary level for executive employees 
to $125 per week when the salary data showed that 20 percent of 
executive employees from all regions and 12 percent of executive 
employees in the West earned less than $130 a week. 35 FR 884-85. The 
Department also increased the long test salary levels for 
administrative and professional employees to $125 and $140, 
respectively.
---------------------------------------------------------------------------

    \17\ Earnings Data Pertinent to a Review of the Salary Tests for 
Executive, Administrative and Professional Employees As Defined in 
Regulations Part 541, (1969), cited in 34 FR 9935.
---------------------------------------------------------------------------

    In 1975, instead of following these prior approaches, the 
Department set the long test salary levels based on increases in the 
Consumer Price Index (CPI), although the Department adjusted the salary 
level downward ``in order to eliminate any inflationary impact.'' 40 FR 
7091. As a result of this recalibration of the 1970 levels, the long 
test salary level for the executive and administrative exemptions was 
set at $155, while the professional level was set at $170. The salary 
levels adopted were intended as interim levels ``pending the completion 
and analysis of a study by [BLS] covering a six month period in 
1975[,]'' and were not meant to set a precedent for future salary level 
increases. Id. at 7091-92. Although the Department intended to increase 
the salary levels after completion of the BLS study of actual salaries 
paid to employees, the envisioned process was never completed, and the 
``interim'' salary levels remained unchanged for the next 29 years.
    As reflected in Table A, the short test salary level increased in 
tandem with the long test level throughout the various rulemakings 
since 1949. Because the short test was designed to capture only those 
white collar employees whose salary was high enough to indicate a 
stronger likelihood of exempt status and thus warrant a less stringent 
duties requirement, the short test salary level was always set 
significantly higher than the long test

[[Page 38526]]

salary level. Thus, in 1975 while the long test salary levels ranged 
from $155 to $170, the short test level was $250.
    The salary level test was most recently updated in 2004, when the 
Department abandoned the concept of separate long and short tests, 
opting instead for one ``standard'' test, and set the salary level 
under a new standard duties test at $455 for executive, administrative, 
and professional employees. Due to the lapse in time between the 1975 
and 2004 rulemakings, the salary threshold for the long duties tests 
(i.e., the lower salary level) did not reflect salaries being paid in 
the economy and had become ineffective at distinguishing between 
overtime-eligible and overtime-ineligible white collar employees. For 
example, at the time of the 2004 Final Rule, the salary levels for the 
long duties tests were $155 for executive and administrative employees 
and $170 for professional employees, while a full-time employee working 
40 hours per week at the federal minimum wage ($5.15 per hour) at that 
time earned $206 per week. 69 FR 22164. Even the short test salary 
level at $250 per week was not far above the minimum wage.
    The Department in the 2004 Final Rule based the new ``standard'' 
duties tests on the short duties tests (which did not limit the amount 
of nonexempt work that could be performed), and tied them to a single 
salary test level that was updated from the long test salary (which 
historically had been paired with a cap on nonexempt work). 69 FR 
22164, 22168-69; see also 68 FR 15570 (``Under the proposal, the 
minimum salary level to qualify for exemption from the FLSA minimum 
wage and overtime requirements as an executive, administrative, or 
professional employee would be increased from $155 per week to $425 per 
week. This salary level would be referred to as the `standard test,' 
thus eliminating the `short test' and `long test' terminology. The 
separate, higher salary level test for professional employees also 
would be eliminated.''). The Department concluded that it would be 
burdensome to require employers to comply with a more complicated long 
duties test given that the passage of time had rendered the long test 
salary level largely obsolete. 69 FR 22164; 68 FR 15564-65. The 
Department believed at the time that the new standard test salary level 
accounted for the elimination of the long duties test. 69 FR 22167.
    In determining the new salary level in 2004, the Department 
reaffirmed our oft-repeated position that the salary level is the 
``best single test'' of exempt status. 69 FR 22165. Consistent with 
prior rulemakings, the Department relied on actual earnings data and 
set the salary level near the lower end of the current range of 
salaries. Specifically, the Department used Current Population Survey 
(CPS) data that encompassed most salaried employees, and set the salary 
level to exclude roughly the bottom 20 percent of these salaried 
employees in each of the subpopulations: (1) The South and (2) the 
retail industry. Although several prior salary levels were based on 
salaries of approximately the lowest 10 percent of exempt salaried 
employees (the Kantor method), the Department stated that the change in 
methodology was warranted in part to account for the elimination of the 
short and long duties tests, and because the utilized data sample 
included nonexempt salaried employees, as opposed to only exempt 
salaried employees. However, as the Department acknowledged, the salary 
arrived at by this method was, in fact, equivalent to the salary 
derived from the Kantor method. 69 FR 22168. Based on the adopted 
methodology, the Department ultimately set the salary level for the new 
standard test at $455 per week.
    In the 2004 Final Rule the Department also created a test for 
highly compensated employees, which provided a minimal duties test for 
workers within the highest compensation range. Reasoning that an 
especially high salary level negated the need for a probing duties 
analysis, the Department provided that employees who earned at least 
$100,000 in total annual compensation (of which at least $455 was paid 
weekly on a salary or fee basis) were covered by the exemption if they 
customarily and regularly spent time on one or more exempt duties, and 
were not engaged in manual work. 69 FR 22172.
    In summary, the regulatory history reveals a common methodology 
used, with some variations, to determine appropriate salary levels. In 
almost every case, the Department examined a broad set of data on 
actual wages paid to salaried employees and then set the salary level 
at an amount slightly lower than might be indicated by the data. In 
1940 and 1949, the Department looked to the average salary paid to the 
lowest level of exempt employees. Beginning in 1958, the Department set 
salary levels to exclude approximately the lowest-paid 10 percent of 
exempt salaried employees in low-wage regions, employment size groups, 
city size, and industry sectors, and we followed a similar methodology 
in 1963 and 1970. The levels were based on salaries in low-wage 
categories in order to protect the ability of employers in those areas 
and industries to utilize the exemptions and in order to mitigate the 
impact of higher-paid regions and sectors. In 1975, the Department 
increased the salary levels based on changes in the CPI, adjusting 
downward to eliminate any potential inflationary impact. 40 FR 7091 
(``However, in order to eliminate any inflationary impact, the interim 
rates hereinafter specified are set at a level slightly below the rates 
based on the CPI.''). In 2004, the Department raised the salary level 
to $455 per week using earnings data of full-time salaried employees 
(both exempt and nonexempt) in the South and in the retail sector. As 
in the past, the use of lower-salary data sets was intended to 
accommodate those businesses for which salaries were generally lower 
due to geographic or industry-specific reasons. This most recent 
revision eliminated the short and long duties requirements in favor of 
a standard duties test for each exemption and a single salary level for 
executive, administrative, and professional employees.
    Between 1938 and 1975, the Department increased the salary level 
every five to nine years. Following the 1975 rulemaking, however, 29 
years passed before the salary level was again raised. In the 2004 
Final Rule, the Department expressed a commitment to updating the 
salary levels ``on a more regular basis,'' particularly when ``wage 
survey data and other policy concerns support such a change.'' 69 FR 
22171. Regular updates to the salary level test are imperative to 
ensuring that the salary level does not become obsolete over time, and 
providing predictability for employers and employees. Not only does the 
annualized current salary level of $23,660 a year not reflect increases 
in nationwide salary levels since 2004, but this figure, as noted 
above, is below the 2014 poverty threshold of $24,008 per year for a 
family of four.\18\ Moreover, since the salary level test was last 
increased in 2004, the federal minimum wage has increased three times, 
from $5.15 to the current rate of $7.25 an hour,\19\ raising the wages 
of overtime-protected employees. The absence of an

[[Page 38527]]

increase in the salary level when combined with past (and future) 
increases to the minimum wage further undermines the effectiveness of 
the salary level to serve as a line of demarcation between overtime-
protected and exempt workers. Mindful of such developments, the 
Department proposes to increase the salary level annually to ensure the 
test's ability to serve as an effective dividing line between exempt 
and nonexempt employees.
---------------------------------------------------------------------------

    \18\ The 2014 poverty threshold for a family of four with two 
related people under 18 in the household. Available at: https://www.census.gov/hhes/www/poverty/data/threshld/.
    \19\ The U.S. Troop Readiness, Veterans' Care, Katrina Recovery, 
and Iraq Accountability Appropriations Act, 2007, Public Law 110-28, 
121 Stat. 112 (Mary 25, 2007), included an amendment to the FLSA 
that increased the applicable Federal minimum wage under section 
6(a) of the FLSA in three steps: To $5.85 per hour effective July 
24, 2007; to $6.55 per hour effective July 24, 2008; and to $7.25 
per hour effective July 24, 2009.
---------------------------------------------------------------------------

ii. Purpose of the Salary Level Requirement
    The Department has long recognized that the line of demarcation 
between the salaries of white collar employees who are overtime-
protected and those who are exempt EAP employees cannot be reduced to a 
standard formula. There will always be white collar overtime-eligible 
employees who are paid above the salary threshold, and employees 
performing EAP duties who are paid below the salary threshold. The 
salary level selected will inevitably affect the number of workers 
falling into each of these categories. As the Department has noted:

    Inevitably, if the salary tests are to serve their purpose in a 
situation where salaries and wages have risen, some employees who 
have been classified as exempt under the present salary tests will 
no longer be within the exemption under any new tests adopted. Such 
employees include some whose status in management or the professions 
is questionable in view of their low salaries. Also included in the 
group who would not be exempt are employees whose exempt status, on 
the basis of their duties and responsibilities, is questionable.

Kantor Report at 5. Historically, when setting the lower, long test 
salary level, the Department strived to ensure that the salary 
threshold reasonably served to reduce instances where obviously 
overtime-protected white collar employees were classified as exempt, 
while avoiding undue exclusions from exemption of employees performing 
bona fide executive, administrative, and professional duties. In 1949, 
the Department noted:

    Regulations of general applicability such as these must be drawn 
in general terms to apply to many thousands of different situations 
throughout the country. In view of the wide variation in their 
applicability the regulations cannot have the precision of a 
mathematical formula. The addition to the regulations of a salary 
requirement furnishes a completely objective and precise measure 
which is not subject to differences of opinion or variations in 
judgment. The usefulness of such a precise measure as an aid in 
drawing the line between exempt and nonexempt employees, 
particularly in borderline cases, seems . . . to be established 
beyond doubt.

Weiss Report at 9. Since 1958, the Department's approach has emphasized 
minimizing the number of white collar employees performing bona fide 
EAP duties who are excluded from the exemption by the salary level. 
This approach was appropriate when there was a long duties test with a 
specific cap on the amount of time that overtime-ineligible employees 
could spend performing nonexempt work. However, this approach is not 
effective in the absence of that limitation, as it does not take into 
sufficient account the inefficiencies (in terms of the administrative 
costs of classifying positions) of applying the duties test to large 
numbers of overtime-eligible white collar employees and the possibility 
of misclassification of those employees as exempt (and possible 
litigation costs associated with misclassification).
    A thorough review of the regulatory history of the seven previous 
increases to the salary levels reveals an essentially common 
methodology to determine the appropriate level, which has been refined 
periodically in order to better meet the salary level test's goals. In 
almost every case, the Department considered a broad set of salary data 
and then set the salary level at an amount slightly lower than the 
dividing line between exempt and nonexempt that might be indicated by 
the data, or otherwise set it ``at points near the lower end of the 
current range of salaries for each of the [EAP] categories.'' Kantor 
Report at 5. The exact line of demarcation set by the Department, 
however, has varied, and is guided by practical considerations that 
allow it to best serve the underlying principles of the exemption, that 
is, to differentiate exempt and nonexempt white collar employees.
    With that objective in mind, the Department proposes to increase 
the minimum salary level required to qualify for the EAP exemptions 
from $455 per week to the 40th percentile of weekly earnings for full-
time salaried workers ($921 per week).\20\ This proposed methodology is 
conceptually similar to the methodology utilized by the Department in 
the 2004 Final Rule, which in turn was largely modeled on the salary 
level methodology first set forth in the Kantor Report in 1958 and used 
by the Department in nearly every salary level rulemaking thereafter. 
See 69 FR 22167-68; Kantor Report at 6-7. Both the proposed methodology 
and its predecessors set the salary level based on a percentile of the 
salaries actually paid to a specified pool of salaried employees.
---------------------------------------------------------------------------

    \20\ The BLS sample used for this rulemaking consists of usual 
weekly earnings for full-time (defined as at least 35 hours per 
week) non-hourly paid employees. For the purpose of this rulemaking, 
the Department considers data representing compensation paid to non-
hourly workers to be an appropriate proxy for compensation paid to 
salaried workers.
---------------------------------------------------------------------------

iii. Sources for the Salary Level Requirement
    After a careful review of the guidance articulated in the 
Department's previous part 541 rulemakings, and observing more than a 
decade of experience since the 2004 salary level test update, the 
Department has chosen to rely on the general methodology used in every 
previous update except 1975, with a few changes designed to simplify 
and improve the methodology as a tool for differentiating exempt and 
nonexempt workers. Specifically, in the interest of making the salary 
methodology simpler and more transparent, the Department is using 
nationwide CPS data on full-time salaried employees (both exempt and 
nonexempt) to set the proposed salary level. As discussed infra, the 
Department is not further modifying the sample as we did in 2004. See 
69 FR 22168.\21\
---------------------------------------------------------------------------

    \21\ As discussed infra, the CPS data on full-time salaried 
workers which the Department is now proposing to use excludes 
certain groups, such as the self-employed, unpaid volunteers, 
workers under age 16, and members of the military on active duty. 
However, BLS automatically excludes these groups when it generates 
the sample. In 2004, the Department took additional steps to exclude 
other categories of workers from the sample.
---------------------------------------------------------------------------

    This is not the first time the Department has modified the 
methodology, in part because the specific sources of the Department's 
data have changed over the years. In 1940, the Department considered 
salary surveys by government agencies, experience under the NIRA, state 
laws, and federal government salaries. Stein Report at 9, 20-21, 31-32. 
In 1949, the Department looked at salary data collected by state and 
federal agencies, including the BLS, and considered wages in small 
towns and low-wage industries, earnings of federal employees, average 
weekly earnings for exempt employees, wages of clerical employees, and 
starting salaries for college graduates. Weiss Report at 10, 13-20. In 
1958, the Department used a data set that consisted of data collected 
during WHD investigations on actual salaries paid to employees who 
qualified for the exemption, grouped by geographic region, broad 
industry groups, number of employees, and size of city, and the 
Department supplemented the investigation data with BLS and Census data 
on the income increases of white collar and

[[Page 38528]]

manufacturing employees for the period not covered by the Department's 
investigations. Kantor Report at 6-9. Subsequent salary level updates 
in 1963 and 1970 followed a similar approach, looking to WHD data on 
actual salaries paid to exempt employees and augmenting the 1970 
analysis with BLS data. 28 FR 7002; 35 FR 884. The Department diverged 
from our practice of looking to actual salary data in the 1975 rule, 
when the Department increased the salary levels set in 1970 based on 
the CPI and adjusted slightly ``in order to eliminate any inflationary 
impact''; those salary levels, however, were intended to be ``interim'' 
levels, pending receipt and review of data on actual salary levels. 40 
FR 7091.
    The Department made some adjustments in 2004 to broaden the data 
set used, rather than continuing to rely upon WHD's limited enforcement 
data. The Department continued to carefully review actual salary 
levels, but did so by using the CPS as the data source. The CPS is a 
large, statistically robust survey jointly administered by the Census 
Bureau and BLS, and it is widely used and cited by industry analysts. 
It surveys 60,000 households a month, covering a nationally 
representative sample of workers, industries, and geographic areas.\22\ 
Households are surveyed for four months, excluded from the survey for 
eight months, surveyed for an additional four months, then permanently 
dropped from the sample. During months 4 and 16 in the sample (the 
outgoing rotation months), employed respondents complete a 
supplementary questionnaire (the merged outgoing rotation group or 
MORG) in addition to the regular survey, which contains the detailed 
information on earnings necessary to estimate a worker's exemption 
status. However, because the Department was unable to precisely 
identify which workers would qualify for the exemption, the Department 
based the salary level in the 2004 Final Rule on a pool of employees 
that generally included those full-time salaried employees covered by 
the FLSA and by the part 541 regulations. Where possible, the 
Department excluded from our analysis workers who were excluded 
entirely from the FLSA's overtime requirements or from the salary 
tests.\23\ 69 FR 22167-68. The Department concluded that it was 
preferable to move away from using a sample limited to exempt salaried 
employees, as was done in the Kantor method, because in order to create 
such a pool of likely-exempt salaried employees one would have to rely 
upon ``uncertain assumptions regarding which employees are actually 
exempt.'' Id. at 22167. In addition, the Department used CPS data 
rather than salary data from the limited pool of our own investigations 
because there would have been too few observations from these 
investigations to yield statistically meaningful results.
---------------------------------------------------------------------------

    \22\ https://www.census.gov/cps; https://www.census.gov/cps/methodology.
    \23\ The 2004 pool of salaried employees excluded: (1) The self-
employed, unpaid volunteers and religious workers who are not 
covered by the FLSA; (2) agricultural workers, certain 
transportation workers, and certain automobile dealership employees 
who are exempt from overtime under other provisions of the Act; (3) 
teachers, academic administrative personnel, certain medical 
professionals, outside sales employees, lawyers and judges who are 
not subject to the part 541 salary tests; and (4) federal employees 
who are not subject to the part 541 regulations. 69 FR 22168.
---------------------------------------------------------------------------

    In this proposed rule, the Department continues to adhere to the 
basic methodological principle of looking to actual salaries paid to 
employees, but as in the 2004 rulemaking, the Department has reexamined 
the precise contours of the sample to ensure that it is as transparent, 
accessible, and easily replicated as possible. By moving to an even 
more standardized sample than the one used in 2004--the proposed rule 
includes all full-time salaried employees nationwide, without 
exclusions--the Department seeks to further improve upon the 
methodology.
    The proposed rule uses CPS data comprising all full-time salaried 
employees to determine the proposed salary levels, and the Department 
is not further restricting the sample. Inclusion of those employees 
previously excluded by the Department in 2004 achieves a more robust 
sample that is more representative of salary levels throughout the 
economy. For example, while teachers, physicians, lawyers, outside 
sales employees, and federal employees were excluded from the 2004 
sample because they are not subject to the part 541 salary level test, 
they nonetheless are part of the universe of salaried employees and, as 
such, their salaries shed light on the salaries paid to employees 
performing exempt EAP duties. Furthermore, replicating this sample from 
the CPS public-use files would require no adjustments, making it easier 
for members of the public to access it and use it.\24\ In contrast, the 
sample from the 2004 rulemaking required filtering out various 
employees based on interpretations of a number of statutory and 
regulatory exclusions from coverage or the salary requirement--a 
process that is inconsistent with the simplification, streamlining, and 
transparency objectives of the current rulemaking.
---------------------------------------------------------------------------

    \24\ The Department notes that the public will not be able to 
exactly replicate the weekly earnings and percentiles used in this 
NPRM from the public-use data files made available by BLS. As with 
all BLS data, to ensure the confidentiality of survey respondents, 
data in the public-use files use adjusted weights and therefore 
minor discrepancies between internal BLS files and public-use files 
exist. BLS publishes quarterly the earnings deciles of full-time 
salaried workers on which the Department relies to set the proposed 
salary level at https://www.bls.gov/cps/research_series_earnings_nonhourly_workers.htm.
---------------------------------------------------------------------------

    Using a broader sample does not diminish the soundness of the 
ultimate salary level derived. As the Department noted with respect to 
our change in the sample for the 2004 rulemaking, different 
``approaches are capable of reaching exactly the same endpoint [i.e., a 
percentile that accomplishes the purpose of the salary level test].'' 
69 FR 22167.
iv. Setting the Required Salary Level
    In addition to looking to a less-restricted sample, this proposed 
rule also differs from the 2004 Final Rule in that the Department 
proposes to set the standard salary level at a higher percentile of the 
salary distribution and relies upon salaries nationwide rather than 
salaries in a limited geographic area or industry. The Department is 
also proposing to set the salary level as a percentile of weekly 
earnings of full-time salaried workers rather than a specific dollar 
amount because we believe a percentile serves as a better proxy for 
distinguishing between overtime-eligible and exempt white collar 
workers as it is rooted in the relative distribution of earnings which 
are linked to the type of work undertaken by salaried workers. The 
proposed standard salary level of the 40th percentile of weekly 
earnings for all full-time salaried employees is higher than the 
percentile used by the Department in either the 2004 Final Rule or the 
Kantor method. In the 2004 Final Rule, the Department set the required 
standard salary level at approximately the 20th percentile of salaried 
employees in the South region and in the retail industry, and in 1958, 
using the Kantor method which had both the long and short tests, the 
Department set the required salary level at approximately the 10th 
percentile of exempt EAP workers' salaries in low-wage regions, 
employment size groups, city size, and industries. As explained in the 
2004 Final Rule, those two methods produced roughly equivalent salary 
levels when taking into account their differing samples. See 69 FR 
22167-68; Kantor Report at 6. Applying

[[Page 38529]]

these methods today would result in salary levels of $577 per week 
(2004 method) or $657 per week (Kantor method), which would equate to 
approximately the 15th and 20th percentiles of weekly earnings for all 
full-time salaried workers.
    However, the higher percentile proposed here is necessary to 
correct for the current pairing of a salary based on the lower salary 
long test with a duties test based on the less rigorous short duties 
test, and ensure that the proposed salary is consistent with the 
Department's longstanding goal of finding an appropriate line of 
demarcation between exempt and nonexempt employees. See, e.g., Weiss 
Report at 11 (``The salary tests in the regulations are essentially 
guides to help in distinguishing bona fide executive, administrative, 
and professional employees from those who were not intended by the 
Congress to come within these categories.''). Currently, approximately 
85 percent of white collar salaried workers who fail the EAP duties 
test earn at least $455 per week. Because the current salary level is 
only screening from exemption approximately 15 percent of overtime-
eligible white collar salaried employees, it is not an effective test 
for exemption and does not serve the intended purpose of simplifying 
application of the exemption by reducing the number of employees for 
whom employers must perform a duties analysis. Increasing the standard 
salary level to the 40th percentile of weekly earnings for full-time 
salaried workers would reduce by 6.3 million the number of white collar 
employees whose exemption status currently can only be determined by 
applying the duties test.\25\ Conversely, only approximately 4 percent 
of all white collar salaried employees who meet the duties test earn 
less than the current salary level. The proposed increase in the 
standard salary level would increase the number of overtime-eligible 
white collar salaried employees who meet the duties test and earn less 
than the proposed salary level to approximately 25 percent.
---------------------------------------------------------------------------

    \25\ These workers are salaried, white collar workers who do not 
satisfy the EAP duties tests and who earn at least $455 per week but 
less than the proposed salary level. Some workers in this group may 
be overtime ineligible due to another non-EAP exemption.
---------------------------------------------------------------------------

    The proposed percentile diverges from the percentiles adopted in 
both the 2004 Final Rule and the Kantor method because it more fully 
accounts for the Department's elimination of the long duties test. As 
discussed in detail below, the Department acknowledged in the 2004 
Final Rule that it was necessary in setting the salary level to account 
for the shift to a single standard duties test that was equivalent to 
the less rigorous short duties test. The Department intended the change 
from the 10th to the 20th percentile to address, in part, the 
elimination of the long duties test. 69 FR 22167. The Department also 
intended this change, however, to account for the use of a different 
data set. 69 FR 22168. Based on further consideration of our analysis 
of the 2004 salary, the Department has now concluded that the $455 
salary level did not adequately account for both the shift to a sample 
including all salaried workers covered by the part 541 regulations, 
rather than just EAP exempt workers, and the elimination of the long 
duties test that had historically been paired with the lower salary 
level. Accordingly, this proposal is intended to correct for that error 
by setting a salary level that fully accounts for the fact that the 
standard duties test is significantly less rigorous than the long 
duties test and, therefore, the salary threshold must play a greater 
role in protecting overtime-eligible employees. This proposal is also 
responsive to the President's desire to simplify the exemption, and it 
addresses the Department's concern that overtime-eligible workers may 
be misclassified as exempt based solely on the salaries they receive.
    This is the first time that the Department has needed to correct 
for such a mismatch between the existing salary level and the 
applicable duties test. Under the old short test/long test structure, 
the Department routinely focused on setting a long test salary level 
that would minimize the number of employees performing bona fide EAP 
duties deemed overtime-eligible based on their salaries (keeping the 
number of such excluded employees to about 10 percent of those who 
qualified for exemption based upon their duties). This approach was 
possible because the long duties test included a limit on the amount of 
nonexempt work that could be performed and thus provided an adequate 
safeguard against the exemption of white collar workers who should be 
overtime-protected but who exceeded the salary level. The creation of a 
single standard test based on the less rigorous short duties test 
caused new uncertainty as to what salary level is sufficient to ensure 
that employees intended to be overtime-protected are not subject to 
inappropriate classification as exempt, while minimizing the number of 
employees disqualified from the exemption even though their primary 
duty is EAP exempt work.
    A brief history of the long duties test illustrates the importance 
of offsetting its elimination with a corresponding increase in the 
salary level. The so-called long test was the sole test for all 
employees until 1949. The Department devised a separate short test in 
1949 to supplement the long test with a short-cut, more permissive, 
method for determining exempt status for only those employees meeting a 
higher salary requirement. For example, the long duties test in effect 
from 1949 to 2004 for administrative employees required that an exempt 
employee: (1) Have a primary duty consisting of the performance of 
office or non-manual work directly related to management policies or 
general business operations of the employer or the employer's 
customers; (2) customarily and regularly exercise discretion and 
independent judgment; (3) regularly and directly assist a proprietor or 
a bona fide executive or administrative employee, or perform under only 
general supervision work along specialized or technical lines requiring 
special training, experience, or knowledge, or execute under only 
general supervision special assignments and tasks; and (4) not devote 
more than 20 percent (or 40 percent in a retail or service 
establishment) of hours worked in the workweek to activities that are 
not directly and closely related to the performance of the work 
described above. 29 CFR 541.2 (2003). By contrast, the short duties 
test in effect during the 1949 to 2004 period provided that an 
administrative employee paid at or above the short test salary level 
qualified for exemption if the employee's primary duty consisted of the 
performance of office or non-manual work directly related to management 
policies or general business operations of the employer or the 
employer's customers which includes work requiring the exercise of 
discretion and independent judgment. Id.
    Between 1949 and 2004, employers were only able to claim the 
exemption based on the less-stringent short duties test for employees 
who were paid a specified higher salary level. The Department reasoned 
that, ``in the categories of employees under consideration the higher 
the salaries paid the more likely the employees are to meet all the 
requirements for exemption, and the less productive are the hours of 
inspection time spent in analysis of the duties performed.'' Weiss 
Report at 22. The original, more thorough duties test became known as 
the long test, and remained for decades

[[Page 38530]]

the test employers were required to satisfy for those employees whose 
salary was insufficient to meet the higher short test salary level.
    Apart from the differing salary requirements, the most significant 
difference between the short test and the long test was the long test's 
limit on the amount of time an exempt employee could spend on nonexempt 
duties while allowing the employer to claim the exemption. For all 
three EAP exemptions, the long duties test imposed a limit on nonexempt 
duties. A bright-line, 20 percent cap on nonexempt work was instituted 
in 1940 for executive and professional employees, and in 1949 for 
administrative employees.\26\ The short duties tests did not include a 
limitation on nonexempt work because employees paid the higher short 
test salary level were likely to ``meet all of the requirements of the 
Administrator's basic definitions of exempt employees, including the 
requirements with respect to nonexempt work.'' Weiss Report at 23. The 
Department reasoned that if the test were to exempt those for whom 
``the nonexempt work is substantial,'' this would be ``contrary to the 
objectives of the Fair Labor Standards Act.'' Id. at 33.
---------------------------------------------------------------------------

    \26\ By statute, beginning in 1961, retail employees could spend 
up to 40 percent of their hours worked performing nonexempt work and 
still be found to meet the duties tests for EAP exemption. 29 U.S.C. 
213(a)(1).
---------------------------------------------------------------------------

    In 2004 the Department discontinued the use of the long duties test 
because it had effectively become dormant due to the passage of time 
since the required salary level had last been raised in 1975, and 
because the Department believed that reinstituting it would be 
administratively burdensome. Instead the Department essentially adopted 
the short duties tests as the standard duties tests, stating that the 
new standard duties tests ``are substantially similar to the current 
short duties tests,'' 69 FR 22214, and that ``it is impossible to 
quantitatively estimate the number of exempt workers resulting from the 
de minimis differences in the standard duties tests compared to the 
current short duties tests.'' Id. at 22192-93. The Department 
recognized the need to adjust the salary percentile previously used to 
set the long test salary level upward to account for the transition to 
a single more lenient duties test. Indeed, the Department stated that 
the increase to the 20th percentile instead of the 10th percentile was 
intended to account for two changes made in 2004: ``because of the 
proposed change from the `short' and `long' test structure and because 
the data included nonexempt salaried employees.'' 69 FR 22167; see 68 
FR 15571. However, although the Department recognized the need to make 
an adjustment because of the elimination of the long duties test, the 
amount of the increase in the required salary actually only accounted 
for the fact that the data set used to set the salary level included 
nonexempt workers while the Kantor method considered only the salaries 
paid to exempt employees. As the data tables in the 2004 Final Rule 
show, a salary of $455 excluded from the exemption 20.2 percent of all 
salaried employees in the South and 20.0 percent of all salaried 
employees in retail. 69 FR 22169, Table 3. However, that same $455 
salary level excluded only 8.2 percent of likely exempt employees in 
the South and 10.2 percent of likely exempt employees in retail. 69 FR 
22169, Table 4. In other words, ``by setting a salary level excluding 
from the exemptions approximately the lowest 20 percent of all salaried 
employees, rather than the Kantor report's 10 percent of exempt 
employees,'' the Department in 2004 actually adopted a percentile that 
produced a salary amount roughly equivalent to the long test salary 
yielded at the 10th percentile using the Kantor method's data set. Id. 
at 22168 (emphases in original). The Department had not, in fact, made 
any additional adjustment to account for the elimination of the long 
duties test.
    Thus, although the Department had identified the need to adjust the 
required salary percentile to account for the elimination of the long 
duties test, the Department effectively paired the short test's less 
stringent duties requirements with the lower salary level historically 
associated with the long duties test.\27\ The long duties tests had 
limited the amount of nonexempt work that could be performed by 
employees for whom the employer claimed the EAP exemption; only 
employees who were paid the higher short test salary level were not 
required to meet the nonexempt duties caps. Because the standard duties 
tests do not contain a cap on the amount of nonexempt work that may be 
performed, after the 2004 rulemaking the salary level test must play a 
larger role in screening out overtime-protected white collar employees.
---------------------------------------------------------------------------

    \27\ Throughout both the 2003 NPRM and 2004 Final Rule, the 
Department emphasized that it was increasing the standard salary 
level from the $155 long test salary level last previously updated 
in 1975. See, e.g., 68 FR 15570; 69 FR 22123 (``The final rule 
nearly triples the current $155 per week minimum salary level 
required for exemption to $455 per week.''); id. at 22171. Neither 
the 2003 NPRM nor the 2004 Final Rule compared the magnitude of the 
new standard salary level against the former $250 per week short 
test salary level.
---------------------------------------------------------------------------

    While the role of the salary level test as an initial test for 
exemption increased in 2004, the Department has always recognized the 
impact of the threshold on overtime-eligible white collar employees. In 
the Stein Report, the Department looked at the impact of various salary 
thresholds on overtime-eligible bookkeepers, noting that approximately 
50 percent of surveyed bookkeepers earned more than the then applicable 
$30 weekly salary threshold, while that number decreased to 
approximately 8 percent at the $50 dollar level at which the applicable 
salary level was ultimately set. Stein Report at 32. The Department 
went on to note that evidence that a salary of $50 ``would not also 
exclude persons who properly deserve the exemption is illustrated by 
the fact that almost 50 percent of the accountants and auditors [many 
of whom are properly considered administrative or professional] earn at 
least $50 a week.'' Id. Similarly, the Weiss Report noted that 
``[a]nother guide of value in determining the appropriate levels of a 
salary test for administrative and professional employees is the 
probable percentage of persons in clerical, subprofessional, or other 
nonexempt occupations who would meet the various salary requirements. 
The salary level adopted must exclude the great bulk of nonexempt 
persons if it is to be effective.'' Weiss Report at 18. The Weiss 
Report went on to look at salaries paid to bookkeepers in New York and 
nine other surveyed cites and noted that, at a salary of $80 per week, 
some hand-bookkeepers in 9 of the 10 cities surveyed would exceed the 
salary level; at $75 per week, the salary test would be met by some 
hand-bookkeepers in all 10 cities. The report noted that the data ``all 
tend to indicate that a salary requirement of about $75 or $80 a week 
for administrative employees is necessary in order to provide adequate 
protection against misclassification since many obviously nonexempt 
employees earn salaries at or near these figures.'' Id. The Department 
set the salary level for administrative employees at $75 per week.
    The Department's 2004 pairing of the lower long test salary level 
with the short test duties requirements also runs contrary to the 
Department's rationale for the short duties test that ``the higher the 
salaries paid the more likely the employees are to meet all the 
requirements for exemption,'' and at ``the higher salary levels in such 
classes of employment, the employees have almost invariably been found 
to meet all

[[Page 38531]]

the other requirements of the regulations for exemption.'' Weiss Report 
at 22. Further, in establishing the short test the Department cautioned 
that ``the salary level must be high enough to include only those 
persons about whose exemption there is normally no question.'' Id. at 
23. Setting the standard salary level at the 40th percentile of 
earnings for full-time salaried workers would effectively correct for 
the Department's establishment in the 2004 Final Rule of a single 
standard duties test that was equivalent to the former short duties 
test without a correspondingly higher salary level. In the absence of 
the protection provided by the long duties test, the lower salary level 
increased the risk that employees who should be entitled to overtime 
protection might be inappropriately classified as exempt and denied 
that protection. The lower salary level associated with the former long 
duties test was never intended to ensure that the employees earning 
that amount meet ``all the requirements for exemption . . . including 
the requirement with respect to nonexempt work.'' Id. at 22-23. 
Therefore, without a more rigorous duties test, the salary level set in 
the 2004 Final Rule is inadequate to serve the salary's intended 
purpose of the ``drawing of a line separating exempt from nonexempt 
employees[.]'' 69 FR 22165.
    The importance of adjusting the salary level threshold upward to 
account for the lack of a long duties test is illustrated by the 
Department's Burger King litigation in the early 1980's, when the long 
test was still actively in use. The Department brought two actions 
arguing that Burger King restaurants in the northeast had misclassified 
their assistant managers as exempt executive employees and that these 
employees were, in fact, entitled to overtime protection. Sec'y of 
Labor v. Burger King Corp., 675 F.2d 516 (2d Cir. 1982); Sec'y of Labor 
v. Burger King Corp., 672 F.2d 221 (1st Cir. 1982). The assistant 
managers at issue all performed the same duties, which included 
spending significant amounts of time performing the same routine, 
nonexempt work as their subordinates. One group of assistant managers 
was paid between $155 and $249 per week--and therefore subject to the 
long duties test; the other group was paid $250 or more--and therefore 
subject to the short duties test. The Department argued that neither 
group of assistant managers had management as their primary duty. Both 
appellate courts found that the employees did have management as their 
primary duty; however, for the lower paid group, both courts found the 
employees to be overtime protected because they spent more than 40 
percent of their time performing nonexempt work and therefore did not 
satisfy the requirements of the long duties test. Accordingly, the 
lower paid employees were protected by application of the more rigorous 
long duties test, while the higher paid employees were found to be 
exempt under the easier short duties test. If the less rigorous short 
duties test had been paired with the long test's lower salary 
threshold--as the Department did in 2004--the lower paid assistant 
managers would have lost their overtime protection.
    The continued extensive litigation regarding employees for whom 
employers assert the EAP exemption also demonstrates that using the 
20th percentile of salaried employees in the South and in retail as the 
threshold has not met the Department's goals as stated in the 2004 
Final Rule of simplifying enforcement and reducing litigation. Id. 
According to a recent Government Accountability Office (GAO) report, 
statistics from the Federal Judicial Center show that the number of 
wage and hour lawsuits filed in federal courts ``has increased 
substantially, with most of this increase occurring in the last 
decade.'' GAO-14-69, ``Fair Labor Standards Act,'' December 2013, at 2, 
6.\28\ A ``total of 8,148 FLSA lawsuits [were] filed in fiscal year 
2012. Since 2001, when 1,947 FLSA lawsuits were filed, the number of 
FLSA lawsuits has increased sharply.'' Id. at 6. Stakeholders advised 
GAO that one of the reasons for the increased litigation was employer 
confusion about which workers should be classified as EAP exempt. Id. 
at 11. Adjusting the salary level upward to account for the absence of 
a more rigorous duties test will ensure that the salary threshold 
serves as a more clear line of demarcation between employees who are 
entitled to overtime and those who are not, and will reduce the number 
of white collar employees who may be misclassified and therefore 
decrease litigation related to application of the EAP duties test. At 
the 40th percentile of full-time salaried workers, there will be 10.9 
million fewer white collar employees for whom employers could be 
subject to potential litigation regarding whether they meet the duties 
test for exemption (4.6 million who would be newly entitled to overtime 
due to the increase in the salary threshold and 6.3 million who 
previously failed the duties test and would now also fail the salary 
level test).
---------------------------------------------------------------------------

    \28\ https://gao.gov/products/GAO-14-69.
---------------------------------------------------------------------------

    As discussed previously, the salary component of the EAP test for 
exemption has always worked hand-in-hand with the duties test in order 
to simplify the application of the exemption. At a lower salary level, 
more overtime-eligible employees will exceed the salary threshold, and 
a more rigorous duties test would be required to ensure that they are 
not classified as falling within an EAP exemption and therefore denied 
overtime pay. At a higher salary level, more employees performing bona 
fide EAP duties will become entitled to overtime because they are paid 
a salary below the salary threshold. Setting the salary threshold too 
low reduces the risk that workers who pass the duties test become 
entitled to overtime protection, but does so at the cost of increasing 
the number of overtime-eligible employees exceeding the salary level 
who are subject to the duties test and possible misclassification. In 
contrast, setting the salary level too high reduces the number of 
overtime-protected employees subject to the duties test and eliminates 
their risk of misclassification, but at the cost of requiring overtime 
protection for workers who pass the duties test. With those concerns in 
mind, the Department has reviewed a variety of data sources to 
ascertain the appropriate amount to increase the required salary level 
in order to ensure that it works effectively with the standard duties 
tests to distinguish between overtime-eligible white collar employees 
and employees performing bona fide EAP duties.
    In the 1949, 1958, 1963, 1970 and 1975 updates to the salary level, 
all of which featured a long test/short test structure, the short test 
salary level was set at approximately 130 to 180 percent of the long 
duties test salary level to adequately establish a salary level that 
obviated the need to engage in a more probing duties analysis. To 
remedy the Department's error from 2004 of pairing the lower long test 
salary with the less stringent short test duties, the Department is 
setting the salary level within the range of the historical short test 
salary ratio so that it will work appropriately with the current 
standard duties test. The Department recognizes that the proposed 
salary amount is only about 140 percent of the long duties test salary 
level under the Kantor method, and thus may be viewed as slightly out 
of line with the historic average of approximately 150 percent of the 
long test at which the short-test salary has

[[Page 38532]]

been set.\29\ This suggests that a salary significantly lower than the 
40th percentile of full-time salaried workers would pose an 
unacceptable risk of inappropriate classification of overtime-protected 
employees without a change in the standard duties test. The Department 
believes that setting the salary level at the 40th percentile of weekly 
wages for all full-time salaried employees will result in a salary 
threshold that properly distinguishes between employees who may meet 
the duties requirements of the EAP exemption and those who likely do 
not, without necessitating a return to the more detailed long duties 
test. The Department notes that currently approximately 75 percent of 
white collar employees who do not meet the duties test earn less than 
the proposed salary threshold. The Department believes that the 40th 
percentile is appropriate because there is no longer a lower salary/
long duties test for EAP exemption to which employers can turn if 
employees do not satisfy the standard salary level. By proposing a 
lower salary level than traditionally used for the short duties test, 
the Department intends to minimize the potential that additional bona 
fide exempt employees might become entitled to overtime because they 
fall below the proposed salary level. The Department notes that 
currently approximately 78 percent of all exempt EAP workers--those who 
are paid on a salary basis of at least $455 per week and meet the 
duties test--earn at least $921 per week.
---------------------------------------------------------------------------

    \29\ The Department estimated the average historic ratio of 149 
percent as the simple average of the fifteen historical ratios of 
the short duties salary level to the long duties salary level 
(salary levels were set in 5 years and in each year the salary level 
varied between the three exemptions: executive, administrative, and 
professional). If the Department had weighted the average ratio 
based on the length of time the historic salary levels were in 
effect, this would have yielded an average historic ratio of 152 
percent.
---------------------------------------------------------------------------

    This salary level also accounts for the fact that the salary 
threshold will apply to all employees nationwide, including employees 
who work in low-wage regions and low-wage industries. In this 
rulemaking, we are proposing a salary level of the 40th percentile of 
the weekly wages of all full-time salaried workers nationwide. The 
Department believes that setting the salary level based on nationwide 
salary data is consistent with the goals of modernizing and simplifying 
the regulations. Using nationwide salary data will also produce a 
salary level appropriate to both low- and high-wage areas and 
industries. While the proposed salary level is lower than the average 
historical short test salary ratio under the Kantor method, a higher 
percentile more in line with the historical short duties test could 
have a negative impact on the ability of employers in low-wage regions 
and industries to claim the EAP exemptions for employees who have bona 
fide executive, administrative, or professional duties as their primary 
duty, particularly in the absence of a long duties test as an 
alternative. As will be discussed in section VII.D., the Department 
believes this proposal is appropriate in low-wage areas and low-wage 
industries.
    The proposal also is consistent with the Department's practice in 
prior rulemakings, including the 2004 Final Rule, of establishing a 
national salary level, rather than multiple levels for different 
regions or industries. As stated in the 2004 Final Rule, the Department 
does not believe that having different salary levels for different 
areas of the country or for different kinds or sizes of businesses ``is 
administratively feasible because of the large number of different 
salary levels this would require.'' 69 FR 22171. The Department came to 
the same conclusion in 1940 when the Department rejected suggestions 
for varying salary levels, stating that it would present serious 
difficulties in enforcement, and that the FLSA is a national law that 
cannot take

into account every small variation occurring over the length and 
breadth of the country. To make enforcement possible and to provide 
for equity in competition, a rate should be selected . . . which 
will be reasonable in light of average conditions for industry as a 
whole. In some instances the rate selected will inevitably deny 
exemption to a few employees who might not unreasonably be exempted, 
but, conversely, in other instances it will undoubtedly permit the 
exemption of some persons who should properly be entitled to the 
benefits of the act.

Stein Report at 6; see Weiss Report at 9 (``Regulations of general 
applicability such as these must be drawn in general terms to apply to 
many thousands of different situations throughout the country.'').
    Setting the salary level at the 40th percentile of full-time 
salaried workers places it far enough above the minimum wage to provide 
an effective means of screening out workers who should be overtime 
protected. As the Stein Report noted, ``[i]t must be assumed that 
[executive employees] enjoy compensatory privileges and this assumption 
will clearly fail if they are not paid a salary substantially higher 
than the wages guaranteed as a mere minimum under section 6 of the 
act.'' Stein Report at 19. Furthermore, the failure to require a salary 
level of substantially more than the minimum wage would ``invite 
evasion of section 6 and 7 for large numbers of workers to whom the 
wage-and-hour provisions should apply.'' Id. Accordingly, following 
each update from 1949 to 1975 (those which included a short duties test 
similar to the current standard test), the ratio of the short test 
salary level to the earnings of a full-time, nonexempt, minimum wage 
worker equaled between approximately 3.0 and 6.25.\30\ See Table B. For 
instance, the ratio was its highest in 1949 at 6.25 ($100 salary level 
divided by the product of $0.40 and 40 hours) and its lowest in 1975 at 
2.98 ($250/($2.10 x 40)). Because the 2004 standard salary level was 
based on the 1975 long test salary and not the short test salary, it 
deviated from the pattern observed over the previous decades, resulting 
in a salary threshold of just 2.21 times full-time minimum wage 
earnings ($455/($5.15 x 40)). The proposed salary level is 3.18 times 
full-time minimum wage earnings ($921/($7.25 x 40)), which is 
consistent with the historical average. Therefore, the Department 
believes that the proposed salary level is appropriate in comparison 
with prior minimum wage ratios.
---------------------------------------------------------------------------

    \30\ The 6.25 ratio is an outlier that was set in December 1949 
(when the short test was created) and the minimum wage increased 
from $.40 to $.75 per hour one month later (which reduced the ratio 
to 3.33). To return to the 6.25 ratio, the weekly salary level would 
have to be set at $1,812.50, which is around the 80th percentile of 
all full-time salaried employees.

                    Table B--Ratios of Salary Test Levels to Full-Time Minimum Wage Earnings
----------------------------------------------------------------------------------------------------------------
                                                             MW earnings for    Exempt short     Ratio of short
                  Year                      Minimum wage        a 40-hour        test salary     salary test to
                                                (MW)            workweek            level          MW earnings
----------------------------------------------------------------------------------------------------------------
1949....................................             $0.40               $16              $100              6.25
1958....................................              1.00                40               125              3.13

[[Page 38533]]

 
1963....................................              1.25                50               150              3.00
1970....................................              1.60                64               200              3.13
1975....................................              2.10                84               250              2.98
----------------------------------------------------------------------------------------------------------------
                  Year                      Minimum wage     MW earnings for    Exempt short     Ratio of short
                                                (MW)            a 40-hour        test salary     salary test to
                                                                workweek            level          MW earnings
----------------------------------------------------------------------------------------------------------------
2004....................................             $5.15              $206              $455              2.21
2015....................................              7.25               290               921              3.18
                                                                                    (proposed)
----------------------------------------------------------------------------------------------------------------

    Moreover, the median earnings for all salaried workers provides 
further support for the proposed salary level. The Weiss Report 
observed approvingly that in the Stein Report, the ``dividing line 
[between subprofessional and professional employees was] based on the 
midpoint salaries'' of federal government service classifications of 
administrative and professional employees, and thus suggested that a 
midpoint value of the aggregated earnings of such workers is an 
appropriate benchmark for the salary level. Weiss Report at 16-17 
(referencing Stein Report at 43). In 1947, 1962, 1969, and 2003, data 
showing median increases in earnings for all employees in various 
industries were generated and considered instructive to a determination 
of an appropriate salary level.\31\ The 2013 national median earnings 
for all full-time salaried workers was $1,065 per week, giving support 
to the Department's proposed salary level of $921. Thus, using median 
earnings as a point of comparison supports that the 40th percentile of 
full-time salaried workers would provide an appropriate line of 
demarcation between overtime-eligible white collar employees and 
potentially exempt EAP employees.
---------------------------------------------------------------------------

    \31\ Statistical Materials Bearing on the Salary Requirement in 
Regulations Part 541 (1947), at 2, 6, 27-30, 56-57; Salary Tests for 
EAP Employees DOL Report--Wage and Hour Public Contracts Division 
(1962), at 3, 7-15, 18, 20; Salary Tests WHD Report (1969), at 19, 
48.
---------------------------------------------------------------------------

    The Department's proposed salary level is further supported by its 
increased ability to distinguish overtime-eligible employees. The 
primary objective of the salary level test has always been the drawing 
of a line separating overtime-eligible white collar salaried employees 
from employees who may be bona fide EAP employees. At the current 
salary threshold, there are 11.6 million salaried white collar workers 
who are overtime protected but are paid at or above the $455 salary 
level and therefore must be subjected to a duties analysis to determine 
their overtime eligibility. At the proposed salary level, the number of 
overtime-eligible salaried white collar employees paid at or above the 
salary level would be reduced by more than 50 percent. Thus a salary 
level at the 40th percentile of weekly earnings for salaried workers 
would be more efficient at distinguishing overtime-eligible employees.
v. Alternatives Considered
    While the Department has largely followed historical precedent in 
determining the proposed salary threshold by basing it on the level of 
salaries that employers currently pay and making only modest changes to 
our time-tested model, the Department did consider other approaches to 
determine the appropriate salary test level.\32\ First, the Department 
considered adjusting either the 2004 standard salary test level or the 
1975 short test salary level for inflation using the CPI, similar to 
the methodology used to set the salary levels in the 1975 interim 
update. The Department noted in 1975 that ``[t]he rapid increase in the 
cost of living since the salary tests were last adjusted justifies an 
interim increase in those tests . . . [and] the widely accepted [CPI] 
may be utilized as a guide for establishing these interim rates.'' 40 
FR 7091. However, the Department noted at that time that the adoption 
of interim rates, while necessary to expeditiously provide protection 
for workers affected by a salary level rendered obsolete by rapid cost-
of-living changes, was not considered a precedent for future rulemaking 
(and those same inflationary conditions do not exist today). Id. at 
7092. In other years, however, the Department has looked at inflation 
when increasing the salary level, but has never established the actual 
numerical salary level based on inflation.
---------------------------------------------------------------------------

    \32\ The alternatives the Department considered are discussed in 
more detail in section VII.C.
---------------------------------------------------------------------------

    The Department has thus recognized that measures of inflation and 
losses in purchasing power provide helpful background for setting the 
salary level because they indicate how far the levels erode between 
updates and underscore the need for an update. They can also point very 
generally to ranges in which new salary levels might be considered. 
Indeed, with respect to the current rulemaking, looking at inflation 
provides added support for the proposed salary level. Updating the 2004 
standard salary level for inflation based on the Consumer Price Index 
for all urban consumers (CPI-U) would result in a salary level of $561 
per week (approximately the 15th percentile of weekly earnings for all 
full-time salaried workers). Updating the 1975 short test salary level 
with the CPI-U would result in a salary level of $1,083 per week 
(approximately the 50th percentile of weekly earnings for all full-time 
salaried workers). Considering that the standard test most closely 
approximates the historic short duties test, looking at an inflation 
adjustment would support a higher salary level than that being 
proposed. However, inflation has been used as a method for setting the 
precise salary level only in the breach, as in 1975 when practical 
considerations prevented a more complete analysis of actual salaries. 
The Department continues to believe that looking to the actual earnings 
of workers provides the best evidence of the rise in prevailing salary 
levels and, thus, constitutes the best source for setting the proposed 
salary requirement. This viewpoint reflects guidance from previous 
updates, including the Weiss Report, where the Department rejected 
suggestions to base the salary level on the change in the cost of 
living. Weiss Report at 12 (``The change in the cost of living which 
was urged by several witnesses as a basis for determining the 
appropriate levels is, in

[[Page 38534]]

my opinion, not a measure for the rise in prevailing minimum 
salaries.'').
    The Department also considered setting the salary level using the 
2004 method (20th percentile of full-time salaried employees in the 
South and retail) or Kantor method (10th percentile of likely exempt 
employees in low-wage regions, employment size groups, city size, and 
industries). While these methods produced similar salaries in 2004 when 
the Department last revised the salary levels, over time they have 
diverged significantly and today would result in salaries of $577 and 
$657 per week, respectively (approximately the 15th and 20th 
percentiles of weekly earnings for all full-time salaried workers). 
Because the Kantor method was based on the long test duties 
requirements (which limited the amount of nonexempt work that EAP 
employees could perform), the Department concluded that the resulting 
salary level was inappropriately low when paired with the standard 
duties test (which was based on the short test). For similar reasons 
the Department concluded that the 2004 method (which paired the lower 
long test salary level with a standard duties test based on the short 
duties test) also resulted in an inappropriately low salary level.
    The Department further considered setting the standard salary level 
equal to the median earnings for all full-time wage and salaried 
workers combined (i.e., not just salaried, also workers paid by the 
hour). This median provides a rough dividing line between the generally 
lower-paid hourly workers who are overtime protected and the generally 
higher-paid salaried workers who may be exempt. The national median 
earnings for all full-time workers, both wage and salary, in all 
occupations and industries, and across metropolitan and rural areas, 
was $776 per week (approximately the 30th percentile of weekly earnings 
for all full-time salaried workers). The Department concluded, however, 
that it would not be appropriate to include the wages of hourly workers 
in setting the EAP salary threshold and that the resulting salary level 
was too low to work effectively with the standard duties test.
    The Department also considered updating the Kantor long test salary 
level of $657 to a short test level, reflecting the historical 
relationship of the short test to the long test which has ranged from 
approximately 130 percent to 180 percent of the long test level and 
averaged approximately 150 percent. This would result in a salary level 
between $854 and $1,183 per week, with the historical average yielding 
a salary level of $979 per week. The end points of the historical range 
are approximately the 35th and 55th percentiles of weekly earnings for 
all full-time salaried workers, respectively. While the Department 
thought that salaries throughout this historical salary range would 
work appropriately with the standard duties test, we were concerned 
that the top end of the resulting range would be too high for low-wage 
regions and industries, particularly because employers no longer have a 
long duties test to fall back on for purposes of exempting lower-
salaried workers performing bona fide EAP duties.
    Finally, the Department considered setting the standard salary 
equal to the 50th percentile, or median, of weekly earnings for all 
full-time salaried workers. This method would be similar to the 
proposed method but would use a higher percentile. Using the 50th 
percentile would result in a standard salary level of $1,065 per week. 
The Department believes that the salary level generated with this 
method would be too high for low-wage regions and industries, 
particularly in light of the absence of a lower salary long duties 
test.
    When measured against inflation or previous methods of setting the 
salary levels (standard, short, and long), the proposed salary level is 
within the range that was the historical norm until the 2004 update. 
For instance, this level falls well below the 1975 inflation-adjusted 
short test level ($1,083 per week) and is lower than the salary level 
comparable to the average historical ratio between the short and long 
test salary ($979 per week). But the proposed salary exceeds the 
inflation-adjusted 2004 salary level and the levels suggested by the 
Kantor and 2004 methods (all of which were based on the long test 
salary). While, for the reasons stated herein, none of these 
alternative measures was used as a methodology to establish the 
proposed salary test level, they confirm that the 40th percentile of 
weekly earnings of all full-time salaried employees ($921) proposed by 
the Department is in line with previous updates.
vi. Summary of Proposed Change to Standard Salary Level
    Therefore, for the reasons stated above, the Department proposes to 
increase the standard salary level to qualify for exemption from the 
FLSA minimum wage and overtime requirements as an executive, 
administrative, or professional employee from $455 a week to the weekly 
earnings of the 40th percentile of full-time salaried employees ($921 a 
week). The Department reached the proposed salary level after 
considering available data on actual salary levels currently being paid 
in the economy. The Department believes that, in view of the regulatory 
history and all other relevant considerations, using the earnings of 
all full-time salaried workers (exempt and nonexempt) as the basis for 
setting the proposed salary level is appropriate here, and setting the 
salary level at the 40th percentile establishes an appropriate dividing 
line helping differentiate between white collar workers who are 
overtime-eligible and those who are not.
    The Department invites comments on this proposed salary level and 
on any alternative salary level amounts, or methodologies for 
determining the salary level, that appropriately distinguish between 
overtime-eligible white collar workers and bona fide EAP workers. In 
addition, the Department invites comments on the effectiveness of the 
proposed salary level to both limit the number of employees who pass 
the EAP duties tests but become overtime eligible because of the 
increased salary level, and reduce the number of employees who fail the 
EAP duties test but are subject to a duties analysis and possible 
misclassification by their employers.

B. Special Salary Tests

 i. American Samoa
    The Department has historically applied a special salary level test 
to employees in American Samoa because minimum wage rates in that 
jurisdiction have remained lower than the federal minimum wage. See 69 
FR 22172. Prior to July 24, 2007, industry-specific minimum wage rates 
for American Samoa were set by a special industry committee appointed 
by the Department. See Sec. 5, Pub. L. 87-30, 75 Stat. 67 (May 5, 
1961). The Fair Minimum Wage Act of 2007 replaced this methodology with 
a system of incremental increases. See Sec. 8103, Pub. L. 110-28, 121 
Stat. 188 (May 25, 2007). As amended, this law provides that the 
American Samoa minimum wage for each industry will increase by $0.50 on 
September 30, 2015, and continue to increase every three years 
thereafter until each equals the federal minimum wage. See Sec. 4, Pub. 
L. 112-149, 126 Stat. 1145 (July 26, 2012). The minimum wage in 
American Samoa currently ranges from $4.18 to $5.59 an hour depending 
on the industry,\33\ and

[[Page 38535]]

so the disparity with the federal minimum wage is expected to remain 
for the foreseeable future. Accordingly, the Department proposes to 
maintain a special salary level test for employees in American Samoa.
---------------------------------------------------------------------------

    \33\ See WHD Minimum Wage Poster for American Samoa, available 
at https://www.dol.gov/whd/minwage/americanSamoa/ASminwagePoster.pdf.
---------------------------------------------------------------------------

    Consistent with our practice since 1975, in the 2004 Final Rule the 
Department set the special salary level test for employees in American 
Samoa at approximately 84 percent of the standard salary test level--
which computed to $380 per week. See 69 FR 22172. The Department 
believes that our approach in the 2004 Final Rule remains appropriate 
given the continued gap between American Samoa and federal minimum wage 
rates. Accordingly, the Department proposes to set the American Samoa 
special salary level test at $774, which equals approximately 84 
percent of the proposed standard salary level of the 40th percentile of 
weekly earnings for full-time salaried workers ($921). The Department 
also proposes that when the minimum wage rate for any industry in 
American Samoa equals the federal minimum wage, the standard salary 
level will then apply in full for all EAP employees in all industries 
in American Samoa.
    The Department invites comments on this special salary level 
proposal.
 ii. Motion Picture Producing Industry
    The Department currently permits employers to classify as exempt 
employees in the motion picture producing industry who are paid at a 
base rate of at least $695 per week (or a proportionate amount based on 
the number of days worked), so long as they meet the duties tests for 
the EAP exemptions. Sec.  541.709. This exception from the ``salary 
basis'' requirement was created to address the ``peculiar employment 
conditions existing in the [motion picture] industry'' (18 FR 2881 (May 
19, 1953)), and applies, for example, when a motion picture industry 
employee works less than a full workweek and is paid a daily base rate 
that would yield at least $695 if six days were worked. Id. The 
Department has provided this industry-specific exception to the salary 
basis requirement since 1953. 18 FR 3930 (July 7, 1953).
    In the 2004 Final Rule the Department increased the base rate for 
motion picture industry employees by the same percentage that the 
salary level tests, on average, increased.\34\ See 69 FR 22190. 
Consistent with the 2004 Final Rule methodology, the Department 
proposes to increase the required base rate proportionally to the 
proposed increase in the standard salary level test. The Department is 
proposing to increase the standard salary level by approximately 102 
percent--from $455 to $921. Accordingly, in Sec.  541.709, the 
Department proposes to increase the current base rate for employees in 
the motion picture industry by approximately 102 percent--from $695 to 
$1,404 per week (or a proportionate amount based on the number of days 
worked).
---------------------------------------------------------------------------

    \34\ Specifically, in the 2004 Final Rule the Department 
increased the standard salary level test by approximately 170 
percent for professional employees (from a long test salary level of 
$170 to a standard test salary level of $455), and by roughly 190 
percent for executive and administrative employees (from a long test 
salary level of $155 to a standard test salary level of $455). The 
Department averaged these two percentiles and increased the base 
rate for motion picture industry employees by 180 percent--from $250 
to $695. See 69 FR 22190.
---------------------------------------------------------------------------

    The Department invites comments on this base rate proposal.

C. Inclusion of Nondiscretionary Bonuses in the Salary Level 
Requirement

    The Department has consistently assessed compliance with the salary 
level test by looking only at actual salary or fee payments made to 
employees and, with the exception of the highly compensated test, has 
not included bonus payments of any kind in this calculation. During 
stakeholder listening sessions several business representatives asked 
the Department to include nondiscretionary bonuses and incentive 
payments as a component of any revised salary level requirement. These 
stakeholders conveyed that nondiscretionary bonuses and incentive 
payments are an important component of employee compensation in many 
industries and stated that such compensation might be curtailed if the 
standard salary level was increased and employers had to shift 
compensation from bonuses to salary to satisfy the new standard salary 
level. They asserted that such a change would have a negative impact on 
the workplace and would undermine managers' sense of ``ownership'' in 
their organizations. A few employer stakeholders also raised the 
possibility of counting fringe benefits and/or commissions toward the 
salary level requirement.
    The Department's longstanding position has been to allow employers 
to pay additional compensation in the form of bonuses in addition to 
the required salary. Sec.  541.604(a). However, in recognition of the 
increased role bonuses play in many compensation systems, and as part 
of the Department's efforts in this rulemaking to modernize these 
regulations, the Department is now considering whether to also permit 
nondiscretionary bonuses and incentive payments to count toward a 
portion of the standard salary level test for the executive, 
administrative, and professional exemptions.\35\ Such payments may 
include, for example, nondiscretionary incentive bonuses tied to 
productivity and profitability. Thus, the Department is considering 
whether compensation such as a nondiscretionary bonus for meeting 
specified performance metrics, in combination with a minimum weekly 
salary amount, may be counted in satisfying the standard salary level 
test.
---------------------------------------------------------------------------

    \35\ The Department notes that overtime-eligible (i.e., 
nonexempt) employees may also receive such bonuses. Where 
nondiscretionary bonuses or incentive payments are made to overtime-
eligible employees, the payments must be included in the regular 
rate when calculating overtime pay. The Department's regulations at 
Sec. Sec.  778.208-.210 explain how to include nondiscretionary 
bonuses in the regular rate calculation. One way to calculate and 
pay such bonuses is as a percentage of the employee's total 
earnings. Under this method, the payment of the bonus includes the 
simultaneous payment of overtime due on the bonus payment. See Sec.  
778.210.
---------------------------------------------------------------------------

    The Department is also considering how to include nondiscretionary 
bonuses and incentive payments as part of the salary level test, if 
such a change is implemented. Compliance with the HCE exemption's 
$100,000 total compensation requirement is assessed annually, and 
employers are permitted to make a ``catch-up'' payment at or shortly 
after the end of the year that counts toward this amount. Employees for 
whom the HCE exemption is claimed must receive the full standard salary 
amount, currently $455, weekly on a salary or fee basis. See Sec.  
541.601(b). The Department believes that a different approach would be 
needed for the standard salary test. Because the only compensation 
guaranteed to employees for whom the employer claims the standard EAP 
exemption is the standard salary threshold amount, the Department 
believes it is important to strictly limit the amount of the salary 
requirement that could be satisfied through the payment of 
nondiscretionary bonuses and incentive pay. The Department is 
considering whether to permit such payments to satisfy 10 percent of 
the standard weekly salary level. The Department recognizes that some 
businesses pay significantly larger bonuses and where larger bonuses 
are paid, the amount attributable toward the EAP standard salary 
requirement would be capped at 10 percent of the salary level if such a 
provision were adopted. The

[[Page 38536]]

Department also believes that the time period over which such 
compensation should be considered must be limited. Permitting bonuses 
to be paid as much as a year out would significantly undermine the 
crucial protection provided by the salary basis requirement, which 
ensures that exempt workers receive a minimum level of compensation on 
a consistent basis. Accordingly, the Department envisions that in order 
for employers to be permitted to credit such compensation toward the 
weekly salary requirement employees would need to receive the bonus 
payments monthly or more frequently. For similar reasons, the 
Department is not considering permitting employers to make a yearly 
catch-up payment like under the HCE exemption.
    With these parameters in mind, the Department seeks comments on 
whether it should modify the standard exemption for executive, 
administrative, and professional employees to permit nondiscretionary 
bonuses and incentive payments to count toward partial satisfaction of 
the salary level test. The Department seeks information on what 
industries commonly have pay arrangements that include nondiscretionary 
bonuses and incentive payments, what types of employees typically earn 
nondiscretionary bonuses and incentive payments, the types of 
nondiscretionary compensation employees receive, and to what extent 
including nondiscretionary bonuses and incentive payments as part of 
the salary level would advance or hinder that test's ability to serve 
as a dividing line between exempt and nonexempt employees. The 
Department also seeks comments on whether payment on a monthly basis is 
the appropriate interval for such nondiscretionary compensation that 
will be credited toward the weekly salary requirement, and whether 10 
percent is the appropriate limit on the amount of the salary 
requirement that can be satisfied by nondiscretionary bonuses and 
incentive payments (with the remaining 90 percent paid on a salary or 
fee basis in accordance with the regulations).
    Consistent with the rule for highly compensated employees (which 
counts nondiscretionary bonuses toward the total annual compensation 
requirement), the Department is not considering expanding the salary 
level test calculation to include discretionary bonuses. The Department 
is also not considering changing the exclusion of board, lodging, or 
other facilities from the salary calculation, a position that it has 
held consistently since the salary requirement was first adopted. 
Similarly, the Department also declines to consider including in the 
salary requirement payments for medical, disability, or life insurance, 
or contributions to retirement plans or other fringe benefits. See 
Sec.  541.601(b)(1). The Department is also concerned it would be 
inappropriate to count commissions toward the salary level requirement, 
as employees who earn commissions are usually sales employees who--with 
the exception of outside sales employees--are generally unable to 
satisfy the standard duties test (which is more stringent than the HCE 
duties test) for the EAP exemptions. However, the Department seeks 
comments on the appropriateness of including commissions as part of 
nondiscretionary bonuses and other incentive payments that could 
partially satisfy the standard salary level test.

D. Highly Compensated Employees

    In the 2004 Final Rule, the Department created a new highly 
compensated exemption for EAP employees. Section 541.601(a) provides 
that such employees are exempt if they earn at least $100,000 in total 
annual compensation and customarily and regularly perform any one or 
more of the exempt duties or responsibilities of an executive, 
administrative, or professional employee. Section 541.601(b)(1) states 
that employees must receive at least $455 per week on a salary or fee 
basis, while the remainder of the total annual compensation may include 
commissions, nondiscretionary bonuses, and other nondiscretionary 
compensation. It also clarifies that total annual compensation does not 
include board, lodging, and other facilities, and does not include 
payments for medical insurance, life insurance, retirement plans, or 
other fringe benefits. Pursuant to Sec.  541.601(b)(2), an employer is 
permitted to make a final payment (catch-up pay) during the final pay 
period or within one month after the end of the 52-week period to bring 
an employee's compensation up to the required level. If an employee 
does not work for a full year, Sec.  541.601(b)(3) permits an employer 
to pay a pro rata portion of the required annual compensation, based 
upon the number of weeks of employment (and one final payment may be 
made, as under paragraph (b)(2), within one month for employees who 
leave employment during the year).
    In the 2003 NPRM, where the HCE test was first introduced, the 
Department had proposed to require total annual compensation of at 
least $65,000. The Department stated that, ``[t]o determine an 
appropriate salary level for highly compensated employees, the 
Department looked to points near the higher end of the current range of 
salaries and found that the top 20 percent of all salaried employees 
earned above $65,000 annually. This level is consistent with setting 
the proposed standard test salary level at the bottom 20 percent of 
salaried employees.'' 68 FR 15571. However, in the 2004 Final Rule, the 
Department recognized that the required compensation level had to ``be 
set high enough to avoid the unintended exemption of large numbers of 
employees--such as secretaries in New York City or Los Angeles--who 
clearly are outside the scope of the exemptions and are entitled to the 
FLSA's minimum wage and overtime pay provisions.'' 69 FR 22174. 
Therefore, the Department increased the required annual compensation to 
$100,000, to ``address commenters' concerns regarding the associated 
duties test, the possibility that workers in high-wage regions and 
industries could inappropriately lose overtime protection, and the 
effect of future inflation.'' Id. at 22175.
    The Department set the level at $100,000 because our experience 
demonstrated that

virtually every salaried ``white collar'' employee with a total 
annual compensation of $100,000 per year would satisfy any duties 
test. Employees earning $100,000 or more per year are at the very 
top of today's economic ladder, and setting the highly compensated 
test at this salary level provides the Department with the 
confidence that, in the words of the Weiss report: ``in the rare 
instances when these employees do not meet all other requirements of 
the regulations, a determination that such employees are exempt 
would not defeat the objectives of section 13(a)(1) of the Act.''

Id. at 22174 (quoting Weiss Report at 22-23). The Department further 
noted that ``[o]nly roughly 10 percent of likely exempt employees who 
are subject to the salary tests earn $100,000 or more per year,'' which 
the Department noted was ``broadly symmetrical with the Kantor approach 
of setting the minimum salary level for exemption at the lowest 10 
percent of likely exempt employees. In contrast, approximately 35 
percent of likely exempt employees subject to the salary tests exceed 
the proposed $65,000 salary threshold.'' Id.
    The Department continues to believe that an HCE test for exemption 
is an appropriate means of testing whether highly compensated employees 
qualify as bona fide executive, administrative, or professional 
employees. In the 2004 Final Rule, the Department concluded that the 
requirement for $100,000 in total annual compensation struck the

[[Page 38537]]

right balance by matching a much higher compensation level than was 
required for the standard salary level test with a duties test that was 
more flexible than the standard duties test, thereby creating a bright-
line test that allowed only appropriate workers to qualify for 
exemption. See 69 FR 22174. This total annual compensation requirement 
was set more than four times higher than the standard salary 
requirement of $455 per week, which totals $23,660 per year. Id. at 
22175. Such a balancing of a substantially higher compensation 
requirement with a minimal duties test still is appropriate, so long as 
the required annual compensation threshold is sufficiently high to 
ensure that it covers only employees who ``have almost invariably been 
found to meet all the other requirements of the regulations for 
exemption.'' Id. at 22174.
    Therefore, the Department proposes to increase the total annual 
compensation required by Sec.  541.601 in order to ensure that it 
remains a meaningful and appropriate standard when matched with the 
minimal duties test. Just as with the standard salary level test, it is 
imperative to increase the compensation level that was established more 
than ten years ago to ensure that it continues to allow for the 
exemption of only bona fide exempt employees. Over the past decade, the 
percentage of salaried employees who earn more than $100,000 annually 
has increased substantially to approximately 17 percent of full-time 
salaried workers. Accordingly, the Department proposes to increase the 
total annual compensation requirement to the annualized weekly earnings 
of the 90th percentile of all full-time salaried workers ($122,148). As 
discussed earlier with respect to the standard salary level, the 
Department is proposing to set the annual compensation requirement as 
the annualized value of a percentile of weekly earnings of full-time 
salaried workers rather than a specific dollar amount because we 
believe it serves as a better proxy for distinguishing those white 
collar workers who meet the requirements of the HCE exemption. 
Consistent with the current regulations, the Department also proposes 
that at least the standard salary requirement must be paid on a salary 
or fee basis.\36\ The Department is not proposing any changes to the 
HCE duties test created in 2004.
---------------------------------------------------------------------------

    \36\ Should the Department implement in the final rule resulting 
from this proposed rule a provision allowing employers to take a 
credit against the standard salary level for nondiscretionary 
bonuses paid to the employee, that credit would not be applicable in 
determining compliance with the standard salary requirement for HCE 
workers.
---------------------------------------------------------------------------

    The Department believes that the 90th percentile of full-time 
salaried workers is appropriate because it brings the required 
compensation level more in line with the level established in 2004; 
therefore, it will ensure that, as in 2004, the HCE exemption covers 
only those employees who are at the very top of today's economic ladder 
and minimizes ``the possibility that workers in high-wage regions and 
industries could inappropriately lose overtime protection.'' 69 FR 
22175. The proposed $122,148 requirement also generally corresponds to 
the increase that would result from updating the $100,000 level by the 
amount of the increase in the CPI-U between 2004 and 2013 (the CPI-U 
increase would result in a compensation level of approximately 
$123,000). The Department invites comments on whether the 90th 
percentile is the correct HCE total annual compensation level and 
whether the Department should make any other changes to the 
requirements for the use of the HCE exemption.

E. Automatically Updating the Salary Levels

    As previously discussed, the salary level test plays a crucial role 
in ensuring that the EAP exemptions effectively differentiate between 
exempt and overtime-protected workers. But even a well-calibrated 
salary level that is fixed becomes obsolete as wages for nonexempt 
workers increase over time. Since the EAP regulations were first issued 
in 1938, the Department has increased the salary level only seven 
times--in 1940, 1949, 1958, 1963, 1970, 1975, and 2004. The lapses 
between rulemakings have resulted in salary levels that are based on 
outdated salary data and thus ill-equipped to help employers assess 
which employees are unlikely to meet the duties tests for the 
exemptions. During stakeholder listening sessions several employee 
advocates called on the Department to index the EAP salary level 
requirement to ensure that the revised salary test set in this 
rulemaking does not suffer the same fate as the salary tests in the 
Department's prior rulemakings.
    After careful consideration of the history of EAP salary increases 
and the impact on the regulated community of routine updating of the 
salary test, the Department is proposing to modernize the EAP 
exemptions by establishing a mechanism for automatically updating the 
standard salary test, as well as the total annual compensation 
requirement for highly compensated employees. The addition of automatic 
updating will ensure that the salary test level is based on the best 
available data (and thus remains a meaningful, bright-line test), 
produce more predictable and incremental changes in the salary required 
for the EAP exemptions, and therefore provide certainty to employers, 
and promote government efficiency by removing the need to continually 
revisit this issue through resource-intensive notice and comment 
rulemaking. The Department also proposes to update annually the special 
salary level test for employees in American Samoa and the base rate 
test for motion picture industry employees, as described infra.
    The Department is considering two alternative methodologies for 
annually updating the salary and compensation thresholds. One method 
would update the thresholds based on a fixed percentile of earnings for 
full-time salaried workers. The other method would update the 
thresholds based on changes in the CPI-U. Both methods are described in 
detail below and the Department seeks comments on which methodology 
would be the most appropriate basis for annual updates to the salary 
and compensation thresholds.
i. History of Automatically Updating the Salary Levels
    The Department has only directly commented twice on the subject of 
automatically updating the salary level test for the EAP exemptions. In 
the 1970 rulemaking, the Department stated that a comment ``propos[ing] 
to institute a provision calling for an annual review and adjustment of 
the salary tests . . . appears to have some merit, particularly since 
past practice has indicated that approximately 7 years elapse between 
amendment of the salary level requirements.'' 35 FR 884. Despite 
recognizing the potential value of this approach, the Department 
ultimately determined that ``such a proposal will require further 
study.'' Id. In the 2004 Final Rule the Department declined to adopt 
commenter requests for automatic increases to the salary level, 
reasoning in part that ``the salary levels should be adjusted when wage 
survey data and other policy concerns support such a change'' and that 
``the Department finds nothing in the legislative or regulatory history 
that would support indexing or automatic increases.'' 69 FR 22171. 
Although the Department acknowledged the lack of historical guidance 
related to the automatic updating of salary levels, in the 2004 Final 
Rule we did not discuss the Department's authority to promulgate such 
an approach through notice and comment rulemaking. Rather than explore 
in greater depth whether

[[Page 38538]]

automatic updates to the salary levels posed a viable solution to 
problems created by lapses between rulemakings, the Department 
expressed our intent ``in the future to update the salary levels on a 
more regular basis, as it did prior to 1975.'' Id. As discussed below, 
difficulties in achieving this goal have led the Department to examine 
the possibility of automatically updating salary levels in greater 
detail.
    The lack of Congressional guidance either supporting or prohibiting 
automatic updating is unsurprising given the origin and evolution of 
the salary level test, and does not foreclose the Department's 
proposal. Congress did not specifically set forth precise criteria, 
such as a salary level test, for defining the EAP exemptions, but 
instead delegated that task to the Secretary. The Department 
established the first salary level tests by regulation in 1938, using 
our delegated authority to define and delimit the EAP exemptions. See 
29 U.S.C. 213(a)(1). The fact that the salary level tests were created 
by regulation after the FLSA was enacted helps explain why the FLSA's 
early legislative history does not address the salary level tests or 
methods for updating the salary level. Despite numerous amendments to 
the FLSA over the past 75 years, Congress has continued to entrust the 
Department with promulgating, updating, and enforcing the salary test 
regulations. Significant regulatory changes since 1938 include adding a 
separate salary level for professional employees in 1940, adopting 
separate short and long test salary levels in 1949, and creating a 
single standard salary level test and a new HCE exemption in 2004. 
These changes were all made without express Congressional guidance, and 
none have been superseded by statute. Other than directing the 
Department in 1990 to include in the section 13(a)(1) exemption 
regulations certain computer employees paid at least six-and-a-half 
times the minimum wage on an hourly basis, see Sec. 2, Pub. L. 101-583, 
104 Stat. 2871 (Nov. 15, 1990), Congress has never amended the FLSA in 
a manner that affects the salary level tests. It has also never enacted 
limits on the Department's ability to update the salary levels. Just as 
the Department has authority under 29 U.S.C. 213(a)(1) to establish and 
update the salary level tests, it likewise has authority to adopt a 
methodology through notice and comment rulemaking for automatically 
updating the salary levels to ensure that the tests remain effective. 
This interpretation is consistent with the well-settled principle that 
agencies have authority to `` `fill any gap left, implicitly or 
explicitly, by Congress.' '' Long Island Care at Home, Ltd. v. Coke, 
551 U.S. 158, 165 (2007) (quoting Chevron, U.S.A., Inc. v. Natural Res. 
Def. Council, Inc., 467 U.S. 837, 843 (1984)).
ii. Rationale for Automatically Updating Salary Levels
    The addition of an automatic updating mechanism will ensure that 
the standard salary level and the HCE total annual compensation 
requirement remain meaningful tests for distinguishing between bona 
fide EAP workers who are not entitled to overtime and overtime-
protected white collar workers. Experience has shown that the salary 
level test is only a strong measure of exempt status if it is up to 
date. Left unchanged, the test becomes substantially less effective as 
wages for overtime-protected workers increase over time. See Weiss 
Report at 8 (``The increase in wage rates and salary levels gradually 
weakened the effectiveness of the present salary tests as a dividing 
line between exempt and nonexempt employees.''); see also 69 FR 22164 
(explaining that 1975 salary levels had grown outdated and were ``no 
longer useful in distinguishing between exempt and nonexempt 
employees''). For example, in 2005 18.6 million workers subject to the 
FLSA were potentially covered by the EAP exemptions and in 2013 that 
number had grown to 21.4 million--an increase of 15 percent--while the 
number of workers subject to the FLSA grew only 5.8 percent during that 
period. See Figure A. Automatically updating the salary level using the 
most recent data ensures that the salary level test continues to 
accurately reflect current salary conditions. This specific proposal 
also helps fulfill the President's instruction to modernize the part 
541 regulations. 79 FR 18737.

[[Page 38539]]

Figure A: Employees Subject to EAP Salary Level Requirement
[GRAPHIC] [TIFF OMITTED] TP06JY15.005

    Automatically updating the salary level will ensure that it 
continues to be a reliable proxy for identifying overtime-eligible 
white collar employees, thus reducing one source of uncertainty for 
employers and employees. Regular updates to the salary level will also 
prevent the more drastic and unpredictable salary level increases that 
have resulted from the differing time periods between rulemakings. For 
example, between 1940 and 2004 the time between salary level updates 
ranged from five to 29 years. In part as a result of these breaks, long 
test salary level increases between 1940 and 1975 ranged from roughly 
five to 50 percent, and the 2004 standard salary level test represented 
an average 180 percent increase from the 1975 long test salary levels. 
Automatically updating the standard salary level test will ensure that 
future salary level increases occur at regular intervals and at more 
even increments.
    The Department recognizes that instituting a mechanism for 
automatically updating the salary level is a change to the part 541 
regulations. As explained in the 2004 Final Rule, the Department's 
reluctance to institute automatic updating was tied in part to our 
preference for issuing new salary level regulations when new wage 
survey data necessitated such action. 69 FR 22171. However, a review of 
salary test history shows that the Department has updated the salary 
level only once since 1975, and has gone nine or more years between 
updates on several occasions. This history underscores the difficulty 
in maintaining an up-to-date and effective salary level test, despite 
the Department's best intentions. Competing regulatory priorities, 
overall agency workload, and the time-intensive nature of notice and 
comment rulemaking have all contributed to the Department's difficulty 
in updating the salary level test as frequently as necessary to reflect 
changes in workers' salaries. These impediments are exacerbated because 
unlike most regulations, which can remain both unchanged and forceful 
for many years if not decades, in order for the salary level test to be 
effective, frequent updates are imperative to keep pace with changing 
employee salary levels. Confronted with this regulatory landscape, the 
Department believes automatic updating is the most viable and efficient 
way to ensure that the standard salary level test and the HCE total 
annual compensation requirement remain current and can serve their 
intended function of helping differentiate between white collar workers 
who are overtime-eligible and those who are not.
iii. Proposal for Automatic Updating of the Standard Salary Level Test
    The Department proposes to insert a new provision in the 
regulations in the Final Rule that will establish a set methodology for 
recalculating the required salary level annually. The Department is not 
proposing specific regulatory text because it has not chosen the 
updating methodology and is instead seeking comments on two 
alternatives--using a fixed percentile of wage earnings or using the 
CPI-U. In the 1970 rulemaking, the Department recognized the potential 
merit of automatically updating the salary level test, but determined 
that such action would ``require further study.'' 35 FR 884. The 
Department has now examined a range of possible updating methodologies 
and concluded, for the reasons stated herein, that either maintaining 
the standard salary level at the 40th percentile of weekly wages of all 
full-time salaried workers or

[[Page 38540]]

updating the standard salary threshold based on changes in the CPI-U 
would maintain the effectiveness of the salary level in distinguishing 
overtime-eligible white collar salaried employees from those who may be 
exempt. Regardless of the updating method used, the Department proposes 
to publish the revised salary and compensation levels annually using 
the most recent data as determined and published by BLS. The Department 
will publish a notice with the new salary level in the Federal 
Register, as well as on the WHD Web site, at least sixty days before 
the updated rates would become effective. Should the Department choose 
to make any changes to the updating methodology in the future, such 
changes would require notice and comment rulemaking.
    1. Fixed Percentile Approach to Automatically Updating the Standard 
Salary Level
    The ``fixed percentile'' approach would permit the Department to 
reset the salary level test by applying the same methodology proposed 
in this rulemaking to update the standard salary level. As explained at 
length in section V.A. of this preamble, the proposed salary level test 
methodology closely tracks prior rulemakings, with a few adjustments 
drawn from the Department's long history of administering the part 541 
regulations. The chosen population--all full-time salaried workers--
represents the broadest pool of workers who could potentially be denied 
overtime pay as bona fide EAP workers. The BLS data for this pool is 
readily available and transparent (all full-time salaried workers in 
the CPS data set are included), and at the 40th percentile level is 
representative of those employees who may be bona fide executive, 
administrative or professional workers. The Department has proposed 
raising the salary percentile to the 40th percentile in part to reflect 
our conclusion that the 20th percentile figure used in the 2004 Final 
Rule did not fully account for the elimination of the more stringent 
long duties test; by updating the long--rather than the short--test 
salary level, and effectively pairing it with the less rigorous short 
duties test, we inadvertently made the exemptions over-inclusive and 
increased the risk of misclassification. The proposed salary level 
percentile reflects the Department's best estimate of the appropriate 
line of demarcation between exempt and nonexempt workers, and 
maintaining the salary level at the 40th percentile by updating it 
annually would ensure that the salary level test continues to fulfill 
its intended purpose. Further, because annual salary level updates 
would be based on actual salaries that employers are currently paying, 
it is consistent with the methodology the Department has used in prior 
rulemakings when setting the required salary level.
    Other factors make the fixed percentile approach well-suited for 
automatic updating. For example, on a quarterly basis, BLS publishes a 
table of deciles of the weekly wages of full-time salaried workers, 
calculated using CPS data,\37\ which would provide employers with 
information on changes in salary levels prior to the annual updates. 
While employers may be more familiar with the CPI-U, the quarterly 
publishing of weekly earnings deciles would provide employers with 
information on changes in wages and allow them to plan for changes in 
the salary threshold. The Department would be able to update the salary 
level test annually using this published BLS table, without modifying 
the data in any way or otherwise engaging in complex data analysis. 
This transparent process would further the President's instruction to 
simplify and modernize the part 541 regulations. It would also ensure 
that salary level updates occur in a manner established in the 
regulations and, thus, do not require additional, time-consuming notice 
and comment rulemaking. Additionally, maintaining the standard salary 
level test at the 40th percentile would ensure that increases in 
overtime-protected employee salaries do not render the salary level 
threshold obsolete; such increases have lessened the effectiveness of 
the salary level test in the past when they were not promptly 
recognized. For all of these reasons, the Department believes that 
automatically updating the standard salary level test annually by 
maintaining the salary level at the 40th percentile of weekly earnings 
for all full-time salaried workers would ensure the standard salary 
level remains a meaningful test for distinguishing between overtime-
protected and potentially exempt white collar employees.
---------------------------------------------------------------------------

    \37\ https://www.bls.gov/cps/research_series_earnings_nonhourly_workers.htm.
---------------------------------------------------------------------------

2. Automatically Updating the Standard Salary Level Using the CPI-U
    The Department could also automatically update the salary level 
test based on changes to the CPI-U--a commonly used economic indicator 
for measuring inflation. The CPI-U calculates inflation by measuring 
the average change over time in the prices paid by urban consumers for 
a set basket of consumer goods and services.\38\ The CPI-U is the 
``broadest and most comprehensive'' of the many CPI statistics 
calculated by BLS, and is published monthly.\39\
---------------------------------------------------------------------------

    \38\ https://stats.bls.gov/cpi/cpifaq.htm.
    \39\ Id.
---------------------------------------------------------------------------

    The Department has generally discussed inflation adjustments in the 
context of determining how to raise the salary level from a prior 
rulemaking, not as a method for ensuring the salary level's ongoing 
effectiveness. The Department has expressed concern in prior part 541 
rulemakings with setting a new salary level test by using inflationary 
indicators to update the prior salary level. These sentiments were 
first raised in 1949 in the Weiss Report, which considered and rejected 
proposals to use cost-of-living increases to update the 1940 salary 
levels. Weiss Report at 12. More recently, in the 2003 NPRM the 
Department considered whether to calculate the new salary level by 
adjusting the 1975 salary levels for inflation, and expressed concern 
that the 1975 figure was a potentially inaccurate benchmark and that an 
inflation-based adjustment would not account for changes in working 
conditions over the preceding 28 years. See 68 FR 15570. We also noted 
in the 2003 NPRM that setting the salary level based on inflation was 
inconsistent with the Department's past practice of looking at actual 
salaries and incomes, not inflation-adjusted amounts, id., and we 
expressed concern in the 2004 Final Rule that this approach ``could 
have an inflationary impact or cause job losses.'' 69 FR 22168.
    Although the Department acknowledges these prior concerns regarding 
whether the CPI-U will accurately track the actual salaries and 
incomes, we believe that using the CPI-U to update the proposed salary 
level, which will be set using current data on wages being paid to 
full-time salaried workers, would ensure that the salary level remains 
a useful tool for distinguishing between overtime-eligible white collar 
employees and those who may be exempt. Many of the concerns raised in 
prior rulemakings are less troublesome here because the Department is 
only proposing to use the CPI-U to automatically update the proposed 
salary level going forward; it is not being used to update the salary 
from its 2004 level. The related concerns about using an outdated 
salary level as a baseline for inflation-based adjustments, and the 
inability of inflation-based indicators to account for changes in 
working conditions, are not cause for concern in the context of

[[Page 38541]]

automatically updating a newly set salary level going forward. The 
proposed salary level provides the most appropriate baseline to 
subsequently update using the CPI-U, and year-to-year changes in 
working conditions should be negligible (especially compared to the 
changes between 1975 and 2004). While the Department considers it 
unlikely that cumulative changes in job duties, compensation practices, 
and other relevant working conditions would undermine application of 
the CPI-U over an extended period of time, should such changes occur 
the Department could adjust the salary level test through notice and 
comment rulemaking.
    The Department expressed concern in the 2003 NPRM about the effect 
that adjusting the 1975 salary levels for inflation ``would have on 
certain segments of industry and geographic areas of the county, 
particularly in the retail industry and in the South, which tend to pay 
lower salaries.'' 68 FR 15570. In the 2004 Final Rule the Department 
explained that these concerns applied ``equally when considering 
automatic increases to the salary levels'' and declined to adopt 
commenter requests to institute a mechanism for automatically updating 
the salary level. 69 FR 22171-72.
    The Department continues to believe that any automatic updating 
mechanism must adequately protect low-wage industries and geographic 
areas. However, two related factors have led the Department to conclude 
that updating the salary level using the CPI-U would not harm 
vulnerable business sectors or have other negative economic effects. 
First, the Department's proposal to set the salary level test at the 
40th percentile of the salaries of all full-time salaried workers 
already accounts for and protects low-wage industries and geographic 
areas. In choosing to set the salary level as a percentile of full-time 
salaried workers, the Department set the salary level at the 40th 
percentile rather than a higher percentile to account for low-wage 
regions and industries. Second, the Department has analyzed the 
historical relationship between the 40th percentile benchmark and the 
CPI-U, and determined that the data does not substantiate the 
Department's past concerns about the likely effects on low-wage regions 
and industries of updating the salary level test using an inflation-
based updating mechanism.
    As discussed in section VII.E., the CPI-U has largely tracked the 
earnings rates of the 40th percentile of weekly wages of full-time 
salaried workers. The two updating methodologies are thus expected to 
produce roughly equivalent salary growth in the future; or, put another 
way, past evidence suggests that updating the salary level using the 
CPI-U would result in a comparable salary level to updating using the 
fixed percentile approach. Since the 40th percentile figure adequately 
protects low-wage industries and areas, it follows that CPI-U based 
updating would do likewise, while also maintaining the appropriate line 
of demarcation between white collar workers who are overtime-eligible 
and those who are not. This congruence also supports the conclusion 
that updating the salary level using the CPI-U, as opposed to actual 
salary and income data, would not produce an appreciably different 
result.
    Automatically updating the salary level test using the CPI-U would 
provide a familiar and well understood method for updating the salary 
level and ensure that the real value of the salary level does not 
degrade over time. The CPI-U is commonly applied as an automatic 
updating mechanism. For example, the Internal Revenue Service uses the 
CPI-U to adjust personal tax brackets, 26 U.S.C. 1(f)(3)-(5), and 
multiple federal agencies use the CPI-U to determine eligibility for a 
wide range of government programs.\40\ And although it was not intended 
to serve as a precedent for future rulemakings, in 1975 the Department 
set salary levels using the consumer price index. 40 FR 7092. Most 
importantly, given the comparable growth rates of the 40th percentile 
benchmark and the CPI-U between 1998 and 2013, the Department believes 
that updating the salary levels using the CPI-U would maintain the 
effectiveness of the standard salary level test.
---------------------------------------------------------------------------

    \40\ See https://fas.org/sgp/crs/misc/R42000.pdf.
---------------------------------------------------------------------------

    The Department seeks comments on both methods to update the 
standard salary level test--the fixed percentile approach and the CPI-
U--including comments on whether one approach is better suited to 
maintaining the effectiveness of the salary level test. Additionally, 
the Department seeks comments on whether to schedule updates based on 
the effective date of the Final Rule, on January 1, or some other 
specified date. The Department also seeks comments on how often 
automatic updates to the salary level test should occur. In order to 
ensure that the salary level tests are based on the best available 
data, the Department proposes to update the salary level annually, 
which will produce predictable and incremental changes. However, we 
seek comments identifying whether a different updating period would be 
more appropriate.
v. Automatic Updates to the Special Salary Test for American Samoa
    As discussed in subpart V.B., the Department has historically set a 
special salary test for employees in American Samoa because minimum 
wage rates there are lower than the federal minimum wage. This gap is 
likely to remain for the foreseeable future since American Samoa's 
industry-specific minimum wage rates are scheduled to increase only 
every three years (Sec. 4, Pub. L. 112-149), and as a result the 
industry with the highest minimum wage will not equal the current 
federal minimum wage ($7.25 an hour) until September 30, 2027.
    Consistent with the 2004 Final Rule, the Department is proposing to 
set the special salary level for employees in American Samoa at 84 
percent of the proposed standard salary level ($774 per week). In 
future years, the Department proposes to automatically update the 
special salary level test in American Samoa with the same frequency as 
the standard salary level and to maintain the 84 percent ratio. The 
Department will publish the updated American Samoa special salary level 
and standard salary level simultaneously. Once any industry-specific 
minimum wage rate in American Samoa equals the federal minimum wage, 
the special salary level will no longer be operative and the standard 
salary level test will apply in full to all EAP employees in all 
industries in American Samoa. The Department seeks comments on this 
proposal.
vi. Automatic Updates to the Base Rate for Motion Picture Industry 
Employees
    As discussed in subpart V.B., the Department is proposing to 
increase the base rate for the motion picture industry exception from 
the salary basis requirement with the same frequency and by the same 
percentage as the proposed increase to the standard salary level test. 
This updating method will ensure that the base rate remains a 
meaningful test for helping determine exempt status for motion picture 
industry employees who work partial workweeks and are paid a daily 
rate, rather than a weekly salary. The Department will publish the 
updated base rate and the standard salary test level simultaneously. 
The Department seeks comments on this proposal.

[[Page 38542]]

vii. Proposal for Automatically Updating the Total Annual Compensation 
Requirement for Highly Compensated Employees
    The Department is also proposing to automatically update the total 
annual compensation requirement for highly compensated employees. This 
change is needed to ensure that only those who are ``at the very top of 
[the] economic ladder'' satisfy the total annual compensation 
requirement and are thus subject to a minimal duties test analysis. 69 
FR 22174. Leaving the total annual compensation requirement at a fixed 
dollar amount would risk exempting increasingly large numbers of 
employees, thus diluting the effectiveness of the HCE total annual 
compensation test and allowing exemption of increasing numbers of 
employees who do not meet the standard duties test. Id. Only by 
automatically updating the requirement so that it does not become 
obsolete can the Department ensure that the workers who satisfy the HCE 
compensation test continue to ``almost invariably . . . meet all the 
other requirements'' for exemption. Id.
    The Department proposes to update the HCE total annual compensation 
requirement with the same method and frequency used to update the 
standard salary level test--either by maintaining the required total 
annual compensation level at the annualized value of the 90th 
percentile of the weekly wages of all full-time salaried workers or by 
updating the total annual compensation requirement based on changes in 
the CPI-U. As discussed with regard to the standard salary level, 
either method for updating the required compensation would preserve 
what the Department has identified as the appropriate dividing line for 
the use of the minimal duties test. The Department also proposes to 
update the portion of the total annual compensation that employers are 
required to pay on a salary basis (proposed to be $921 per week) so 
that it continues to mirror the standard salary requirement as it is 
updated. The Department seeks comments on both methods of updating the 
HCE total annual compensation requirement, including comments on 
whether one method is better suited to maintaining the effectiveness of 
the compensation test.

F. Duties Requirements for Exemption

    While the Department has long viewed the salary level test as an 
initial bright-line test for white collar overtime eligibility, we have 
always recognized the salary level test works in tandem with the duties 
test. As previously explained, the part 541 regulations set forth three 
criteria that, in most instances, must be met for an employee to be 
excluded from the Act's minimum wage and overtime pay protections. 
Employees must (1) be paid on a salary basis, (2) be paid at least a 
fixed minimum salary per week, and (3) meet certain requirements as to 
their job duties.\41\ From the outset, examination of the duties 
performed by the employee was an integral part of the determination of 
exempt status, and employers must establish that the employee's 
``primary duty'' is the performance of exempt work in order for the 
exemption to apply. Each of the categories included in section 13(a)(1) 
has separate duties requirements. From 1949 until 2004 the regulations 
contained two different duties tests for executive, administrative, and 
professional employees depending on the salary level paid--a long 
duties test for employees paid a lower salary, and a short duties test 
for employees paid at a higher salary level. The long duties test 
included a 20 percent limit on the time spent on nonexempt tasks (40 
percent for employees in the retail or service industries). In the 2004 
Final Rule, the Department replaced the differing short and long duties 
tests with a single standard test for executive, administrative, and 
professional employees that did not include a cap on the amount of 
nonexempt work that could be performed.
---------------------------------------------------------------------------

    \41\ The exemptions for outside sales employees, doctors, 
lawyers, teachers, and computer employees are distinct from the 
other exemptions with respect to their salary requirements.
---------------------------------------------------------------------------

    The duties test has always worked in conjunction with the salary 
requirement to correctly identify exempt EAP employees. The Department 
has often noted that as salary levels rise a less robust examination of 
the duties is needed. This inverse correlation between the salary level 
and the need for an extensive duties analysis was the basis of the 
historical short and long duties tests. While the salary provides an 
initial bright-line test for EAP exemption, application of a duties 
test is imperative to ensure that overtime-eligible employees are not 
swept into the exemption. While the contours of the duties tests have 
evolved over time, the Department has steadfastly maintained that 
meeting a duties test remains a core requirement for the 
exemptions.\42\
---------------------------------------------------------------------------

    \42\ Over the years since the original EAP regulations were 
first implemented, commenters have repeatedly suggested that salary 
should be the sole basis for the exemption. For example, at a 1949 
hearing, ``some of the management witnesses were sufficiently 
convinced of the desirability of salary tests to propose the 
adoption of a salary level as the sole basis of exemption.'' Weiss 
Report at 9. The Department declined to use salary as the sole basis 
for exemption, stating that the ``Administrator would undoubtedly be 
exceeding his authority if he included within the definition of 
these terms craftsmen, such as mechanics, carpenters, or linotype 
operators, no matter how highly paid they might be.'' Weiss Report 
at 23. As recently as the 2004 Final Rule, the Department has 
maintained the view ``that the Secretary does not have authority 
under the FLSA to adopt a `salary only' test for exemption'' and 
rejected suggestions from employer groups to do so. 69 FR 22173.
---------------------------------------------------------------------------

    During the stakeholder listening sessions held in advance of this 
proposed rule, the Department heard from employer stakeholders, 
particularly in the retail and restaurant industries, who advocated for 
the need to maintain flexibility in the duties tests. These 
stakeholders stated that the ability of a store or restaurant manager 
or assistant manager to ``pitch in'' and help line employees when 
needed was a key part of their organizations' management culture and 
necessary to enhancing the customer experience. They emphasized that 
the employees in these entry-level management positions are critically 
important to their organizations and that the experience they gain in 
these positions will lead to higher level management opportunities. 
Employer stakeholders universally urged the Department not to consider 
any changes to the current duties tests, explaining that while the 
duties tests are sometimes difficult to apply and may not be perfect, 
employers have an understanding of the meaning and application of the 
current duties tests and any changes might engender costly litigation 
as parties try to adapt to and interpret the new rules.
    Employee stakeholders, on the other hand, stated that the current 
duties tests, particularly the 50 percent primary duty rule of thumb 
(Sec.  541.700(c)) and the concurrent duties doctrine for executives 
(Sec.  541.106), are insufficiently protective of employees. In 
particular, they expressed concern with cases in which the exemption 
has been applied to employees who have spent large amounts of time 
(sometimes more than 90 percent) performing nonexempt work. They 
asserted that some businesses, particularly in the retail industry, 
have built into their business model having exempt store managers 
perform significant amounts of nonexempt work in order to keep labor 
costs down. These employee stakeholders argued that where employees are 
essentially required to perform significant amounts of nonexempt work, 
the employees do not, in fact, have a primary duty of management in any 
meaningful sense.

[[Page 38543]]

In response to this concern, a few employer stakeholders argued that 
the concurrent duties regulation already addresses this issue by 
distinguishing between exempt executive employees who choose when to 
perform nonexempt duties and nonexempt employees who must perform 
duties as they are assigned. Sec.  541.106(a).
    The Department appreciates the views shared by employer and 
employee stakeholders on this important issue. The Department 
understands the importance of managers ``pitching in'' and leading by 
example. At the same time, the Department is concerned that employees 
in lower-level management positions may be classified as exempt and 
thus ineligible for overtime pay even though they are spending a 
significant amount of their work time performing nonexempt work. The 
Department believes that, at some point, a disproportionate amount of 
time spent on nonexempt duties may call into question whether an 
employee is, in fact, a bona fide EAP employee. The Department is 
concerned that the removal of the more protective long duties test in 
2004 has exacerbated these concerns and led to the inappropriate 
classification as EAP exempt of employees who pass the standard duties 
test but would have failed the long duties test. The issue sometimes 
arises when a manager is performing exempt duties less than 50 percent 
of the time, but it is argued that those duties are sufficiently 
important to nonetheless be considered the employee's primary duty. The 
issue also arises when a manager who is performing nonexempt duties 
much of the time is deemed to perform exempt duties concurrently with 
those nonexempt duties, and it is argued the employee is exempt on that 
basis. While the regulations provide that exempt executives can perform 
exempt duties concurrently with nonexempt duties, Sec.  541.106, this 
rule can be difficult to apply and can lead to varying results. Compare 
In re Family Dollar FLSA Litigation, 637 F.3d 508 (4th Cir. 2011) 
(manager of retail chain store considered an executive exempt from 
overtime pay requirements under the FLSA whether collecting cash, 
sweeping the floor, stocking shelves, working with employee schedules, 
or running a cash register); with Morgan v. Family Dollar Stores, Inc., 
551 F.3d 1233 (11th Cir. 2008) (store managers not exempt executives 
where they spent most of their time performing manual, not managerial, 
tasks). California has addressed this issue by requiring that exempt 
EAP employees spend at least 50 percent of their time performing their 
primary duty, and not counting time during which nonexempt work is 
performed concurrently. Cal. Lab. Code Sec. 515(a), (e); see Heyen v. 
Safeway Inc., 157 Cal. Rptr. 3d 280, 302 (Cal. Ct. App. 2013).
    Taking into account the views of stakeholders, the Department is 
seeking to determine whether, in light of our salary level proposal, 
changes to the duties tests are also warranted. The duties test must 
adequately protect overtime-eligible white collar employees who exceed 
the salary threshold from misclassification as exempt EAP employees.
    The Department is proposing to set the salary threshold at the 40th 
percentile of weekly earnings of full-time salaried employees. As 
previously discussed, because the standard duties test is based on the 
short duties test--which was intended to work with a higher salary 
level--and the proposed salary level is below the historic average for 
the short test salary, a salary level significantly below the 40th 
percentile would necessitate a more robust duties test to ensure proper 
application of the exemption. The Department believes that the salary 
level increase proposed in this NPRM, coupled with automatic updates to 
maintain the effectiveness of the salary level test, will address most 
of the concerns relating to the application of the EAP exemption. A 
regularly updated salary level will assist in screening out employees 
who spend significant amounts of time on nonexempt duties and for whom 
exempt work is not their primary duty. However, the Department invites 
comments on whether adjustments to the duties tests are necessary, 
particularly in light of the proposed change in the salary level test. 
The Department recognizes that duties remain a critical metric of 
exempt status and invites comment on the effectiveness of the duties 
tests found in the current regulations.
    While the Department is not proposing specific regulatory changes 
at this time, the Department is seeking additional information on the 
duties tests for consideration in the Final Rule. Specifically, the 
Department seeks comments on the following issues:
    A. What, if any, changes should be made to the duties tests?
    B. Should employees be required to spend a minimum amount of time 
performing work that is their primary duty in order to qualify for 
exemption? If so, what should that minimum amount be?
    C. Should the Department look to the State of California's law 
(requiring that 50 percent of an employee's time be spent exclusively 
on work that is the employee's primary duty) as a model? Is some other 
threshold that is less than 50 percent of an employee's time worked a 
better indicator of the realities of the workplace today?
    D. Does the single standard duties test for each exemption category 
appropriately distinguish between exempt and nonexempt employees? 
Should the Department reconsider our decision to eliminate the long/
short duties tests structure?
    E. Is the concurrent duties regulation for executive employees 
(allowing the performance of both exempt and nonexempt duties 
concurrently) working appropriately or does it need to be modified to 
avoid sweeping nonexempt employees into the exemption? Alternatively, 
should there be a limitation on the amount of nonexempt work? To what 
extent are exempt lower-level executive employees performing nonexempt 
work?
    In addition to seeking comments on the duties tests, the Department 
is also considering whether to add to the regulations examples of 
additional occupations to provide guidance in administering the EAP 
exemptions. Employer stakeholders have indicated that examples of how 
the exemptions may apply to specific jobs, such as those provided in 
current Sec. Sec.  541.203, 541.301(e), and 541.402, are useful in 
determining exempt status and should be expanded. The Department agrees 
that examples of how the general executive, administrative, and 
professional exemption criteria may apply to specific occupations are 
useful to the regulated community and seeks comments on what specific 
additional examples of nonexempt and exempt occupations would be most 
helpful to include.
Computer Related Occupations
    In further effort to provide effective guidance to the public on 
the administration of the EAP exemptions, the Department is considering 
the suggestions of employer stakeholders from the computer and 
information technology sectors to include additional examples of the 
application of the EAP exemptions to occupational categories in 
computer-related fields. The Department has, as a threshold matter, 
reviewed the authority by which it might include additional examples of 
computer-related occupations. For the reasons articulated in the 
preamble to the 2004 Final Rule, the Department continues to believe 
that we should not expand the EAP exemption beyond the computer 
exemption currently set forth in section 13(a)(17), given the clarity

[[Page 38544]]

with which Congress has set forth the scope of that exemption.\43\ 69 
FR 22160.
---------------------------------------------------------------------------

    \43\ Although the 1990 amendments to the FLSA afforded the 
Department some discretion to elaborate on computer-specific 
exemption criteria distinct from the standard EAP exemption criteria 
(Sec. 2, Pub. L. 101-583, 104 Stat 2871 (Nov. 15, 1990)), the 
Department concluded in the 2004 Final Rule that, because Congress 
subsequently codified the criteria for a computer employee exemption 
in FLSA section 13(a)(17) (Sec. 2105(a), Pub. L. 104-188, 110 Stat. 
1755 (Aug. 20, 1996)), it would be ``inappropriate'' to engage in 
further rulemaking after Congress had spoken on the issue. 69 FR 
22160.
---------------------------------------------------------------------------

    However, in the 2004 Final Rule, the Department did add additional 
examples of occupations within the computer industry such as systems 
analysts and computer programmers which, subject to a case-by-case 
duties analysis, might fall within the section 13(a)(1) administrative 
and executive exemptions. Sec.  541.402. In response to stakeholder 
input and as part of our broader effort to simplify part 541, the 
Department is again exploring the possibility of listing additional 
illustrative examples that typically do or do not fall within the 
general criteria for the three basic EAP exemptions (see Sec. Sec.  
541.100, .200, .300), as opposed to those falling within the computer-
specific exemption set forth in section 13(a)(17), to bring additional 
clarity to employers and employees within the computer and information 
technology industries.
    The Department continues to be cognizant of the ``tremendously 
rapid pace of significant changes occurring in the information 
technology industry'' (69 FR 22158), and therefore requests comments 
from employer and employee stakeholders in the computer and information 
technology sectors as to what additional occupational titles or 
categories should be included as examples in the part 541 regulations, 
along with what duties are typical of such categories and would thus 
cause them to generally meet or fail to meet the relevant EAP exemption 
criteria. To provide additional context, the Department, as an initial 
matter, expresses the view that a help desk operator whose responses to 
routine computer inquiries (such as requests to reset a user's password 
or address a system lock-out) are largely scripted or dictated by a 
manual that sets forth well-established techniques or procedures would 
not possess the discretion and independent judgment necessary for the 
administrative exemption, nor would that individual likely qualify for 
any other EAP exemption. On the other hand, an information technology 
specialist who, without supervision, routinely troubleshoots and 
repairs significant glitches in his company's point of sale software 
for the company's retail clients might be an example of an 
administrative employee pursuant to Sec.  541.200 as this employee's 
work appears to be directly related to the management or business 
operation of his employer or employer's customers and requires the use 
of discretion and independent judgment with respect to matters of 
significance.

VI. Paperwork Reduction Act

    The Paperwork Reduction Act of 1995 (PRA), 44 U.S.C. 3501 et seq., 
and its attendant regulations, 5 CFR part 1320, require the Department 
to consider the agency's need for its information collections, their 
practical utility, as well as the impact of paperwork and other 
information collection burdens imposed on the public, and how to 
minimize those burdens. The PRA typically requires an agency to provide 
notice and seek public comments on any proposed collection of 
information contained in a proposed rule. See 44 U.S.C. 3506(c)(2)(B); 
5 CFR 1320.8. Persons are not required to respond to the information 
collection requirements until they are approved by the Office of 
Management and Budget (OMB) under the PRA. This NPRM would revise the 
existing information collection requirements previously approved under 
OMB control number 1235-0018 (Records to be Kept by Employers--Fair 
Labor Standards Act) and OMB control number 1235-0021 (Employment 
Information Form) in that employers would need to maintain records of 
hours worked for more employees and more employees may file complaints 
to recover back wages under the overtime pay provision. As required by 
the PRA, the Department has submitted the information collection 
revisions to OMB for review in order to reflect changes that would 
result from this proposed rule were it to be adopted.
    Summary: FLSA section 11(c) requires all employers covered by the 
FLSA to make, keep, and preserve records of employees and of wages, 
hours, and other conditions of employment. A FLSA-covered employer must 
maintain the records for such period of time and make such reports as 
prescribed by regulations issued by the Secretary of Labor. The 
Department has promulgated regulations at 29 CFR part 516 to establish 
the basic FLSA recordkeeping requirements. No new information 
collection requirements would be imposed by the adoption of this NPRM; 
rather, burdens under existing requirements are expected to increase as 
more employees receive minimum wage and overtime protections due to the 
proposed increase in the salary level requirement. More specifically, 
the proposed changes in this NPRM may cause an increase in burden on 
the regulated community because employers will have additional 
employees to whom certain long-established recordkeeping requirements 
apply (e.g., maintaining daily records of hours worked by employees who 
are not exempt from the both minimum wage and overtime provisions). 
Additionally, the proposed changes in this NPRM may cause an initial 
increase in burden if more employees file a complaint with WHD to 
collect back wages under the overtime pay requirements. We anticipate 
this increased burden will wane over time as employers adjust to the 
new rule.
    Purpose and Use: WHD and employees use employer records to 
determine whether covered employers have complied with various FLSA 
requirements. Employers use the records to document compliance with the 
FLSA, including showing qualification for various FLSA exemptions. 
Additionally, WHD uses the Employment Information form to document 
allegations of non-compliance with labor standards the agency 
administers.
    Technology: The regulations prescribe no particular order or form 
of records and employers may preserve records in forms of their 
choosing provided that facilities are available for inspection and 
transcription of the records.
    Minimizing Small Entity Burden: Although the FLSA recordkeeping 
requirements do involve small businesses, including small state and 
local government agencies, the Department minimizes respondent burden 
by requiring no specific order or form of records in responding to this 
information collection. Burden is reduced on complainants by providing 
a template to guide answers.
    Public Comments: As part of its continuing effort to reduce 
paperwork and respondent burden, the Department conducts a preclearance 
consultation program to provide the general public and Federal agencies 
with an opportunity to comment on proposed and continuing collections 
of information in accordance with the PRA. This program helps to ensure 
that requested data can be provided in the desired format, reporting 
burden (time and financial resources) is minimized, collection 
instruments are clearly understood, and the impact of collection 
requirements on respondents can be properly assessed. The Department 
seeks public comments regarding the

[[Page 38545]]

burdens imposed by the information collections associated with this 
NPRM. Commenters may send their views about this information collection 
to the Department in the same manner as all other comments (e.g., 
through the regulations.gov Web site). All comments received will be 
made a matter of public record and posted without change to https://www.regulations.gov, including any personal information provided.
    As previously noted, an agency may not conduct an information 
collection unless it has a currently valid OMB approval, and the 
Department has submitted information collection requests under OMB 
control numbers 1235-0018 and 1235-0021 in order to update them to 
reflect this rulemaking and provide interested parties a specific 
opportunity to comment under the PRA. See 44 U.S.C. 3507(d); 5 CFR 
1320.11. Interested parties may receive a copy of the full supporting 
statements by sending a written request to the mail address shown in 
the ADDRESSES section at the beginning of this preamble. In addition to 
having an opportunity to file comments with the Department, comments 
about the paperwork implications may be addressed to OMB. Comments to 
OMB should be directed to: Office of Information and Regulatory 
Affairs, Attention OMB Desk Officer for the Wage and Hour Division, 
Office of Management and Budget, Room 10235, Washington, DC 20503; 
Telephone: 202-395-7316/Fax: 202-395-6974 (these are not toll free 
numbers). OMB will consider all written comments that the agency 
receives within 30 days of publication of this proposed rule. 
Commenters are encouraged, but not required, to send the Department a 
courtesy copy of any comments sent to OMB. The courtesy copy may be 
sent via the same channels as comments on the rule.
    OMB and the Department are particularly interested in comments 
that:
     Evaluate whether the proposed collections of information 
are necessary for the proper performance of the functions of the 
agency, including whether the information will have practical utility;
     Evaluate the accuracy of the agency's estimate of the 
burden of the proposed collection of information, including the 
validity of the methodology and assumptions used;
     Enhance the quality, utility, and clarity of the 
information to be collected; and
     Minimize the burden of the collection of information on 
those who are to respond, including through the use of appropriate 
automated, electronic, mechanical, or other technological collection 
techniques or other forms of information technology (e.g., permitting 
electronic submission of responses).
    Total annual burden estimates, which reflect both the existing and 
new responses for the recordkeeping and complaint process information 
collections at the proposed salary, are summarized as follows:
    Type of Review: Revisions to currently approved information 
collections.
    Agency: Wage and Hour Division, Department of Labor.
    Title: Records to be Kept by Employers--Fair Labor Standards Act.
    OMB Control Number: 1235-0018.
    Affected Public: Private sector businesses or other for-profits, 
farms, not-for-profit institutions, state, local and tribal 
governments, and individuals or households.
    Estimated Number of Respondents: 3,771,434 (unaffected by this 
rulemaking).
    Estimated Number of Responses: 50,467,523 (6,909,600 added by this 
rulemaking).
    Estimated Burden Hours: 1,235,161 hours (230,320 added by this 
rulemaking).
    Estimated Time per Response: Various (unaffected by this 
rulemaking).
    Frequency: Various (unaffected by this rulemaking).
    Other Burden Cost: 0.
    Title: Employment Information Form.
    OMB Control Number: 1235-0021.
    Affected Public: Businesses or other for-profit, farms, not-for-
profit institutions, state, local and tribal governments, and 
individuals or households.
    Total Respondents: 38,138 (2,788 added by this rulemaking).
    Estimated Number of Responses: 38,138 (2,788 added by this 
rulemaking).
    Estimated Burden Hours: 12,713 (930 hours added by this 
rulemaking).
    Estimated Time per Response: 20 minutes (unaffected by this 
rulemaking).
    Frequency: once.
    Other Burden Cost: 0.

VII. Analysis Conducted in Accordance With Executive Order 12866, 
Regulatory Planning and Review, and Executive Order 13563, Improving 
Regulation and Regulatory Review

    Executive Orders 12866 and 13563 direct agencies to assess all 
costs and benefits of available regulatory alternatives and, if the 
regulation is necessary, to select regulatory approaches that maximize 
net benefits (including potential economic, environmental, public 
health and safety effects, distributive impacts, and equity). Executive 
Order 13563 emphasizes the importance of quantifying both costs and 
benefits, of reducing costs, of harmonizing rules, and of promoting 
flexibility.
    Under Executive Order 12866, the Department must determine whether 
a regulatory action is economically ``significant,'' defined as having 
an annual effect on the economy of $100 million or more, and therefore 
subject to review by OMB and the requirements of the Executive Order. 
This proposed rule is economically significant within the meaning of 
Executive Order 12866; therefore, the Department has prepared a 
Preliminary Regulatory Impact Analysis (PRIA) in connection with this 
proposed rule as required under section 6(a)(3) of Executive Order 
12866, and OMB has reviewed the rule.

A. Introduction

i. Background
    The FLSA applies to all enterprises that have employees engaged in 
commerce or in the production of goods for commerce and have an annual 
gross volume of sales made or business done of at least $500,000 
(exclusive of excise taxes at the retail level that are separately 
stated); or are engaged in the operation of a hospital, an institution 
primarily engaged in the care of the sick, the aged, or individuals 
with intellectual disabilities who reside on the premises; a school for 
intellectually or physically disabled or gifted children; a preschool, 
elementary or secondary school, or an institution of higher education 
(without regard to whether such hospital, institution or school is 
public or private, or operated for profit or not); or are engaged in an 
activity of a public agency. See 29 U.S.C. 203(s).
    There are two ways an employee may be covered by the provisions of 
the FLSA: (1) Enterprise coverage, in which any employee of an 
enterprise covered by the FLSA is covered, and (2) individual coverage, 
in which even employees of non-covered enterprises may be covered if 
they are engaged in interstate commerce or in the production of goods 
for commerce, or are employed in domestic service. The FLSA requires 
employers to: (1) Pay employees who are covered and not exempt from the 
Act's requirements not less than the Federal minimum wage for all hours 
worked and overtime premium pay at a rate of not less than one and one-
half times the employee's regular rate of pay for all hours worked over 
40 in a workweek, and (2) make, keep, and preserve records of the 
persons

[[Page 38546]]

employed by the employer and of the wages, hours, and other conditions 
and practices of employment. It is widely recognized that the general 
requirement that employers pay a premium rate of pay for all hours 
worked over 40 in a workweek is a cornerstone of the Act, grounded in 
two policy objectives. The first is to spread employment by 
incentivizing employers to hire more employees rather than requiring 
existing employees to work longer hours, thereby reducing involuntary 
unemployment. The second policy objective is to reduce overwork and its 
detrimental effect on the health and well-being of workers.
    The FLSA provides a number of exemptions from the Act's minimum 
wage and overtime pay provisions, including one for bona fide 
executive, administrative, and professional employees. Such employees 
typically receive more monetary and non-monetary benefits than most 
blue collar and lower-level office workers and therefore are less 
likely to need the Act's protection. Thus, Congress created the 
exemption from the FLSA's minimum wage and overtime pay protections for 
employees employed in a bona fide executive, administrative, or 
professional capacity and for outside sales employees, as those terms 
are ``defined and delimited'' by the Department. 29 U.S.C. 213(a)(1). 
The Department's regulations implementing those exemptions are codified 
at 29 CFR part 541.
    For an employer to exclude an employee from minimum wage and 
overtime protection pursuant to the EAP exemptions, the employee 
generally must meet three criteria: (1) The employee must be paid a 
predetermined and fixed salary that is not subject to reduction because 
of variations in the quality or quantity of work performed (the 
``salary basis test''); (2) the amount of salary paid must meet a 
minimum specified amount (the ``salary level test''); and (3) the 
employee's job duties must primarily involve executive, administrative, 
or professional duties as defined by the regulations (the ``duties 
test''). The regulations governing these tests have been updated 
periodically since the FLSA's enactment in 1938, most recently in 2004 
when, among other revisions, the Department increased the salary level 
test to $455 per week.
    As a result of inflation and the low value of the salary threshold, 
the annual value of this salary level test, $23,660 ($455 per week for 
52 weeks), is now slightly below the 2014 poverty threshold for a 
family of four ($24,008),\44\ making it inconsistent with Congress' 
intent to exempt only bona fide EAP workers, who typically earn 
salaries well above those of any workers they may supervise and 
presumably enjoy other privileges of employment such as above average 
fringe benefits, greater job security, and better opportunities for 
advancement. Stein Report at 21-22.
---------------------------------------------------------------------------

    \44\ The 2014 poverty threshold for a family of four with two 
related people under 18 in the household. Available at: https://www.census.gov/hhes/www/poverty/data/threshld/.
---------------------------------------------------------------------------

    In the 2004 Final Rule, the Department also changed the structure 
of the duties test. Between 1949 and 2004, the EAP exemptions included 
two versions of the duties test. Assuming that a worker was paid on a 
salary basis, the exemptions would be met if a worker passed either a 
``long'' test, which involved a more rigorous set of duties criteria, 
paired with a lower salary level, or a ``short'' test, which imposed 
fewer duties requirements, paired with a higher salary level. In the 
1975 update, the last before the 2004 Final Rule, the Department set 
the long test salary levels at $155 per week for executive and 
administrative employees and $170 per week for professional employees. 
The short test salary level was set at $250 per week for all three EAP 
categories. In 2004, the Department replaced the two-test structure 
with a single ``standard'' duties test for each category, which closely 
resembles the former short test duties requirements, and a single 
salary level test of $455 per week based on an update of the 1975 long 
test salary level. The Department also introduced a highly compensated 
employee (HCE) exemption in the 2004 Final Rule, under which an 
employee may be exempt if he or she passes a very minimal duties test, 
receives at least $455 per week paid on a salary basis, and is highly 
compensated, defined in the 2004 Final Rule as earning total annual 
compensation, which may include commissions and nondiscretionary 
bonuses in addition to a salary, of at least $100,000. The HCE duties 
test is much more abbreviated than the historical short test duties 
requirements.
    The premise behind the salary level tests is that employers are 
more likely to pay higher salaries to workers in bona fide EAP jobs 
than to workers performing nonexempt duties. A high salary is 
considered a measure of an employer's good faith in classifying an 
employee as exempt, because an employer is less likely to have 
misclassified a worker as exempt if he or she is paid a high wage. 
Stein Report at 5; Weiss Report at 8.
    The salary level requirement was created to identify the dividing 
line distinguishing workers performing truly exempt duties from the 
nonexempt workers Congress intended to be protected by the FLSA's 
minimum wage and overtime provisions. Throughout the regulatory history 
of the FLSA, the Department has considered the salary level test the 
``best single test'' of exempt status. Stein Report at 19. This bright-
line test is easily observed, objective, and clear. Id.
ii. Need for Rulemaking
    The salary level test has been updated seven times since it was 
implemented in 1938. Table 1 presents the weekly salary levels 
associated with the EAP exemptions since 1938, organized by exemption 
and long/short/standard duties test.

                            Table 1--Historical Salary Levels for the EAP Exemptions
----------------------------------------------------------------------------------------------------------------
                                                                    Long test
                 Date enacted                  --------------------------------------------------   Short test
                                                   Executive     Administrative    Professional        (all)
----------------------------------------------------------------------------------------------------------------
1938..........................................             $30               $30  ..............  ..............
1940..........................................              30               $50             $50  ..............
1949..........................................              55                75              75            $100
1958..........................................              80                95              95             125
1963..........................................             100               100             115             150

[[Page 38547]]

 
1970..........................................             125               125             140             200
1975..........................................             155               155             170             250
----------------------------------------------------------------------------------------------------------------
                                                  Standard Test
----------------------------------------------------------------------------------------------------------------
2004..........................................                                $455
----------------------------------------------------------------------------------------------------------------

    The standard salary level was set at $455 per week in 2004. 
Following more than ten years of inflation, the purchasing power, or 
real value, of the standard salary level test has eroded substantially, 
and as a result increasingly more workers earn above the salary 
threshold. By 2013 the real value of the standard salary level had 
declined 18.9 percent since 2004, calculated using the Consumer Price 
Index for all urban consumers (CPI-U).\45\ Figure 1 demonstrates how 
the real values of the salary levels have changed since 1949, measured 
in 2013 dollars.
---------------------------------------------------------------------------

    \45\ CPI-U data available at: https://data.bls.gov/cgi-bin/cpicalc.pl.
[GRAPHIC] [TIFF OMITTED] TP06JY15.000

    As a result of the erosion of the real value of the standard salary 
level, more and more workers lack the clear protection the salary level 
test is meant to provide. Each year that the salary level is not 
updated, its utility as a distinguishing mechanism between exempt and 
nonexempt workers declines. The Department has revised the levels just 
once in the 40 years since 1975. In contrast, in the 37 years between 
1938 and 1975, salary test levels were increased approximately every 
five to nine years. In our 2004 rulemaking, the Department stated the 
intention to ``update the salary levels on a more regular basis, as it 
did prior to 1975,'' and added that ``the salary levels should be 
adjusted when wage survey data and other policy concerns support such a 
change.'' 69 FR 22171.
    The real value of the salary level test has fallen substantially 
both when measured against its 2004 level and the 1975 levels. If the 
standard EAP salary level established in 2004 had kept up with 
inflation (measured using the CPI-U), it would be $561 per week in 2013 
dollars, a 23.3 percent increase relative

[[Page 38548]]

to its current level. If the EAP salary level for the short test 
established in 1975 had kept up with inflation, it would be $1,083 per 
week, a 137.9 percent increase relative to the current salary level.
    In order to restore the value of the standard salary level as a 
line of demarcation between those workers for whom Congress intended to 
provide minimum wage and overtime protections and those workers who may 
be performing bona fide EAP duties, and to maintain its continued 
validity, the Department proposes to set the standard salary level 
equal to the 40th percentile of weekly earnings for all full-time 
salaried workers. Based on 2013 salary data,\46\ this is equivalent to 
a standard salary level of $921 per week. The Department also proposes 
to automatically update the standard salary level annually in the 
future. Furthermore, the Department proposes to set the HCE 
compensation level at the 90th percentile of annualized weekly earnings 
for full-time salaried workers, equivalent to $122,148, and to update 
the level annually in the future. Automatic updating would preserve the 
value of these earnings thresholds, eliminate the volatility associated 
with previous changes in the thresholds, and provide certainty for 
employers with respect to future changes. It would also simplify the 
updating process, as the Department would simply publish a notice in 
the Federal Register of the updated salary and compensation thresholds 
on an annual basis, and additional notice and comment rulemaking to 
adjust the salary and annual compensation thresholds would not be 
necessary unless the Department determined in the future that the 
methodology for setting the standard salary or the HCE total 
compensation levels needed to be adjusted.
---------------------------------------------------------------------------

    \46\ Unless otherwise noted, the Department relied upon 2013 
data in the development of the NPRM. The Department will update the 
data used in the Final Rule resulting from this proposal.
---------------------------------------------------------------------------

iii. Summary of Affected Workers, Costs, Benefits, and Transfers
    The Department estimated the number of affected workers and 
quantified costs and transfer payments associated with this proposed 
rulemaking.\47\ All estimates are based on analysis of the Current 
Population Survey (CPS), a monthly survey of 60,000 households 
conducted by the U.S. Census Bureau. In 2013, there were an estimated 
144.2 million wage and salary workers in the United States, of whom 
128.5 million were subject to the FLSA and the Department's 
regulations.48 49 Of these 128.5 million workers, the 
Department estimates that 43.0 million are white collar salaried 
employees who may be affected by a change to the Department's part 541 
regulations and are not covered by another (non-EAP) 
exemption.50 51 The remaining 85.5 million workers include 
blue collar workers, workers paid on an hourly basis, and workers 
eligible for another (non-EAP) overtime exemption. These workers were 
excluded because they will generally not be affected by this proposed 
rulemaking. Of the 43.0 million workers discussed above, the Department 
estimates that 28.5 million are exempt from the minimum wage and 
overtime pay provisions under the part 541 EAP exemptions, while 14.4 
million do not satisfy the duties tests for EAP exemption and/or earn 
less than $455 per week.\52\ However, of the 28.5 million EAP exempt 
workers, 7.1 million were in ``named occupations'' and thus need only 
pass the duties tests to be subject to the standard EAP exemptions.\53\ 
Therefore, these workers were not considered in the analysis, leaving 
21.4 million EAP exempt workers potentially affected by this proposed 
rule.
---------------------------------------------------------------------------

    \47\ Because the Department has not proposed specific changes to 
the duties tests, potential changes to the duties tests are not 
included in this RIA. However, the Department discusses a potential 
methodology for determining the impact of any changes to the duties 
test in section VII.F.
    \48\ Data on wage and salary workers are from the CPS, series 
ID: LNU02000000.
    \49\ Workers not covered as employees by the FLSA and/or the 
Department's regulations include: members of the military, unpaid 
volunteers, the self-employed, many religious workers, and most 
federal employees. The number of workers covered by the FLSA was 
estimated using the CPS Merged Outgoing Rotation Group (MORG) data.
    \50\ As discussed in more detail later, the Department used 
pooled data from 2011-2013 to represent the 2013 population in order 
to increase sample size, and thus the granularity of results.
    \51\ As discussed later, the Department excluded from this 
analysis certain workers for whom their employer could claim a non-
EAP exemption from the FLSA's minimum wage and overtime pay 
provisions, and certain workers for whom the employer could claim an 
overtime pay exemption. For simplicity, the Department refers to 
these exemptions as other (non-EAP) exemptions.
    \52\ Here and elsewhere in this analysis, numbers are reported 
at varying levels of aggregation, and are generally rounded to a 
single decimal point. However, calculations are performed using 
exact numbers. Therefore, as in this case, some numbers may not 
match the reported total or the calculation shown due to rounding of 
components.
    \53\ Workers not subject to the EAP salary level test include 
teachers, academic administrative personnel, physicians, lawyers, 
judges, and outside sales workers.
---------------------------------------------------------------------------

    The Department proposes to increase the standard salary level from 
$455 per week to the 40th percentile of weekly earnings for all full-
time salaried workers, which translates to $921 per week, an increase 
of $466 over the current level (Table 2).\54\ The Department also 
proposes to increase the HCE annual compensation level to the 90th 
percentile of annualized weekly earnings for full-time salaried 
workers, which translates to $122,148 annually.
---------------------------------------------------------------------------

    \54\ The BLS data set used for this rulemaking consists of 
earnings for all full-time (defined as at least 35 hours per week) 
non-hourly paid employees. For the purpose of this rulemaking, the 
Department considers data representing compensation paid to non-
hourly workers to be an appropriate proxy for compensation paid to 
salaried workers.

                                         Table 2--Proposed Salary Levels
----------------------------------------------------------------------------------------------------------------
                                                                                          Total increase
                  Salary level                    Current salary     Proposed    -------------------------------
                                                       level       salary level          $               %
----------------------------------------------------------------------------------------------------------------
Standard exemption..............................       $455/week       $921/week             466           102.4
HCE exemption...................................    100,000/year    122,148/year          22,148            22.1
----------------------------------------------------------------------------------------------------------------

    The Department also proposes to annually update the standard salary 
level to ensure the ongoing effectiveness of the salary level as a 
means of delimiting workers who should not fall within the EAP 
exemption. Similarly, the Department proposes to annually update the 
HCE total annual compensation level to ensure the effectiveness of the 
annual compensation requirement as a test for which employees should be 
subject to the minimal duties test for the HCE exemption.
    In Year 1, an estimated 4.6 million workers would be affected by 
the increase in the standard salary level test (Table 3). This figure 
consists of currently EAP-exempt workers who earn at least $455 per 
week but less than the 40th percentile ($921) of all full-

[[Page 38549]]

time salaried workers. Additionally, an estimated 36,000 workers would 
be affected by the increase in the HCE compensation test. In Year 10, 
with automatic updating, between 5.1 and 5.6 million workers are 
projected to be affected by the change in the standard salary level 
test and between 33,000 and 42,000 workers affected by the change in 
the HCE total annual compensation test, depending on the updating 
methodology used.

   Table 3--Estimated Number of Affected EAP Workers, Ten-Year Projections With and Without Automatic Updating
----------------------------------------------------------------------------------------------------------------
                                        Without automatic        Updated with fixed        Updated with CPI-U
                                            updating                 percentiles       -------------------------
 Affected EAP workers (1,000s) \a\ ----------------------------------------------------
                                       Year 1      Year 10       Year 1      Year 10       Year 1      Year 10
----------------------------------------------------------------------------------------------------------------
Standard exemption................        4,646        2,685        4,646        5,568        4,646        5,062
HCE exemption.....................           36            7           36           42           36           33
----------------------------------------------------------------------------------------------------------------
\a\ Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection
  under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels).

    Three direct costs to employers are quantified in this analysis: 
(1) Regulatory familiarization costs; (2) adjustment costs; and (3) 
managerial costs. Regulatory familiarization costs are the costs 
incurred to read and become familiar with the requirements of the rule. 
Adjustment costs are the costs accrued to determine workers' new 
exemption statuses, notify employees of policy changes, and update 
payroll systems. Managerial costs associated with this proposed 
rulemaking occur because hours of workers who are newly entitled to 
overtime may be more closely scheduled and monitored to minimize or 
avoid paying the overtime premium.
    The costs presented here are the combined costs for both the change 
in the standard salary level test and the HCE annual compensation level 
(these will be disaggregated in section VII.D.iv.). With updating, 
total average annualized direct employer costs were estimated to be 
between $239.6 and $255.3 million, assuming a 7 percent discount rate; 
hereafter, unless otherwise specified, average annualized values will 
be presented using the 7 percent real discount rate (Table 4). 
Deadweight loss (DWL) is also a cost but not a direct employer cost. 
DWL is a function of the difference between the wage employers are 
willing to pay for the hours lost, and the wage workers are willing to 
take for those hours. In other words, DWL represents the decrease in 
total economic surplus in the market arising from the change in the 
regulation. Average annualized DWL was estimated to be between $9.5 and 
$10.5 million, depending on updating methodology.
    In addition to the direct costs, this proposed rulemaking will also 
transfer income from employers to employees in the form of wages. 
Average annualized transfers were estimated to be between $1,178.0 and 
$1,271.4 million, depending on updating methodology. The majority of 
these transfers are from employers to affected EAP workers who become 
overtime protected due to changes in the EAP regulations.
    Employers may incur additional costs, such as hiring new workers. 
These other costs are discussed in section VII.D.iv.5. Another 
potential impact of the rule's proposed increase in the salary 
threshold is a reduction in litigation costs. Other unquantified 
transfers, costs, and benefits are discussed in section VII.D.vii.

   Table 4--Summary of Regulatory Costs and Transfers, Standard and HCE Salary Levels With Automatic Updating
                                                [Millions 2013$]
----------------------------------------------------------------------------------------------------------------
                                                                  Future years \c\      Average annualized value
                                   Automatic                 ---------------------------------------------------
      Cost/transfer \a\         updating method     Year 1                                3% real      7% real
                                      \b\                        Year 2      Year 10        rate         rate
----------------------------------------------------------------------------------------------------------------
Direct employer costs........  Percentile......       $592.7       $188.8       $225.3       $248.8       $255.3
                               CPI-U...........        592.7        181.1        198.6        232.3        239.6
Transfers \d\................  Percentile......      1,482.5      1,160.2      1,339.6      1,271.9      1,271.4
                               CPI-U...........      1,482.5      1,126.4      1,191.4      1,173.7      1,178.0
DWL..........................  Percentile......          7.4         10.8         11.2         10.5         10.5
                               CPI-U...........          7.4         10.3          9.7          9.6          9.5
----------------------------------------------------------------------------------------------------------------
\a\ Costs and transfers for affected workers passing the standard and HCE tests are combined.
\b\ The percentile method sets the standard salary level at the 40th percentile of weekly earnings for full-time
  salaried workers and the HCE compensation level at the 90th percentile. The CPI-U method adjusts both levels
  based on the annual percent change in the CPI-U.
\c\ These costs/transfers represent a range over the nine-year span.
\d\ This is the net transfer from employers to workers. There may also be transfers of hours and income from
  some workers to other workers.

iv. Terminology and Abbreviations
    The following terminology and abbreviations will be used throughout 
this Regulatory Impact Analysis (RIA).
    Affected EAP workers: The population of potentially affected EAP 
workers who either earn between $455 and the proposed salary level of 
the 40th percentile of weekly earnings ($921) or qualify for the HCE 
exemption and earn between $100,000 and the 90th percentile of earnings 
($122,148 annually). This is estimated to be 4.7 million workers.\55\
---------------------------------------------------------------------------

    \55\ Setting the standard salary level at the 40th percentile is 
estimated to affect 4,646,000 workers. See Table 3. Additionally, 
36,000 workers are potentially affected by the change in the HCE 
exemption's total compensation level. Id. Accordingly, throughout 
this NPRM we refer to the total affected workers as 4.7 million 
(4,646,000 + 36,000, rounded to the nearest 100,000 workers). 
However, when discussing only those workers affected by the change 
in the standard salary level test, the number decreases to 4.6 
million (4,646,000, similarly rounded).

---------------------------------------------------------------------------

[[Page 38550]]

    BLS: Bureau of Labor Statistics.
    CPI-U: Consumer Price Index for all urban consumers.
    CPS: Current Population Survey.
    Duties test: To be exempt from the FLSA's minimum wage and overtime 
requirements under section 13(a)(1), the employee's primary job duty 
must involve bona fide executive, administrative, or professional 
duties as defined by the regulations. The Department distinguishes 
among four such tests:
    Standard duties test: The duties test used in conjunction with the 
standard salary level test, as set in 2004 and applied to date, to 
determine eligibility for the EAP exemptions. It replaced the short and 
long tests in effect from 1949 to 2004, but its criteria closely follow 
those of the former short test.
    HCE duties test: The duties test used in conjunction with the HCE 
compensation level test, as set in 2004 and applied to date, to 
determine eligibility for the HCE exemption. It is much less stringent 
than the standard and short duties tests to reflect that very highly 
paid employees are much more likely to be properly classified as 
exempt.
    Long duties test: One of two duties tests used from 1949 until 
2004; this more restrictive duties test had a greater number of 
requirements, including a limit on the amount of nonexempt work that 
could be performed, and was used in conjunction with a lower salary 
level test to determine eligibility for the EAP exemptions (see Table 
1).
    Short duties test: One of two duties tests used from 1949 to 2004; 
this less restrictive duties test had fewer requirements and was used 
in conjunction with a higher salary level test to determine eligibility 
for the EAP exemptions (see Table 1).
    DWL: Deadweight loss; the loss of economic efficiency that can 
occur when equilibrium in a market for a good or service is not 
achieved.
    EAP: Executive, administrative, and professional.
    HCE: Highly compensated employee; a category of EAP-exempt 
employee, established in 2004 and characterized by high earnings and a 
minimal duties test.
    MORG: Merged Outgoing Rotation Group supplement to the CPS.
    Named occupations: Workers in named occupations are not subject to 
the salary level or salary basis tests. These occupations include 
teachers, academic administrative personnel,\56\ physicians,\57\ 
lawyers, and judges.\58\
---------------------------------------------------------------------------

    \56\ Academic administrative personnel (including admissions 
counselors and academic counselors) need to be paid either (1) the 
salary level or (2) a salary that is at least equal to the entrance 
salary for teachers in the educational establishment at which they 
are employed (see Sec.  541.204). Entrance salaries at the 
educational establishment of employment cannot be distinguished in 
the data and so this alternative is not considered (thus these 
employees were excluded from the analysis, the same as was done in 
the 2004 Final Rule).
    \57\ The term physician includes medical doctors including 
general practitioners and specialists, osteopathic physicians 
(doctors of osteopathy), podiatrists, dentists (doctors of dental 
medicine), and optometrists (doctors of optometry or with a Bachelor 
of Science in optometry). Sec.  541.304(b).
    \58\ Judges may not be considered ``employees'' under the FLSA 
definition. However, since this distinction cannot be made in the 
data, all judges are excluded from the analysis (the same as was 
done in the 2004 Final Rule).
---------------------------------------------------------------------------

Overtime Workers
    Occasional overtime workers: Workers who report they usually work 
40 hours or less per week (identified with variable PEHRUSL1 in CPS 
MORG) but in the survey week worked more than 40 hours (variable 
PEHRACT1 in CPS MORG).
    Regular overtime workers: Workers who report they usually work more 
than 40 hours per week (identified with variable PEHRUSL1 in CPS MORG).
    Pooled data for 2011-2013: CPS MORG data from 2011-2013 with 
earnings inflated to 2013 dollars and sample observations weighted to 
reflect the population in 2013; used to increase sample size.
    Potentially affected EAP workers: EAP exempt workers who are not in 
named occupations and are included in the analysis (i.e., white collar, 
salaried, not eligible for another (non-EAP) overtime pay exemption). 
This is estimated to be 21.4 million workers.
    Price elasticity of demand (with respect to wage): The percentage 
change in labor hours demanded in response to a one percent change in 
wages.
    Real dollars (2013$): Dollars adjusted using the CPI-U to reflect 
the purchasing power they would have in 2013.
    Salary basis test: The EAP exemptions' requirement that workers be 
paid on a salary basis, that is, a pre-determined amount that cannot be 
reduced because of variations in the quality or quantity of the 
employee's work.
    Salary level test: The salary a worker must earn in order to be 
subject to the EAP exemptions. The Department distinguishes among four 
such tests:
    Standard salary level: The weekly salary level associated with the 
standard duties test that determines (in part) eligibility for the EAP 
exemptions. The standard salary level was set at $455 per week in the 
2004 Final Rule.
    HCE compensation level: Workers who meet the standard salary level 
requirement but not the standard duties test nevertheless are exempt if 
they pass a minimal duties test and earn at least the HCE total annual 
compensation required amount. The HCE required compensation level was 
set at $100,000 per year in the 2004 Final Rule, of which at least $455 
per week must be paid on a salary or fee basis.
    Short test salary level: The weekly salary level associated with 
the short duties test (eliminated in 2004).
    Long test salary level: The weekly salary level associated with the 
long duties test (eliminated in 2004).
    Workers covered by the FLSA and subject to the Department's 
regulations: Includes all workers except those excluded from the 
analysis because they are not covered by the FLSA or subject to the 
Department's requirements. Excluded workers include: members of the 
military, unpaid volunteers, the self-employed, many religious workers, 
and federal employees (with a few exceptions).\59\
---------------------------------------------------------------------------

    \59\ Employees of firms with annual revenue less than $500,000 
who are not engaged in interstate commerce are also not covered by 
the FLSA. However, these workers are not excluded from this analysis 
because the Department has no reliable way of estimating the size of 
this worker population, although the Department believes it composes 
a small percent of workers. These workers were also not excluded 
from the 2004 Final Rule.
---------------------------------------------------------------------------

    The Department also notes that the terms employee and worker are 
used interchangeably throughout this analysis.

B. Methodology To Determine the Number of Potentially Affected EAP 
Workers

i. Overview
    This section explains the methodology used to estimate the number 
and characteristics of workers who are subject to the EAP exemptions. 
In this proposed rule, as in the 2004 Final Rule, the Department 
estimated the number of EAP exempt workers because there is no data 
source that identifies workers as EAP exempt. Employers are not 
required to report EAP exempt workers to any central agency or as part 
of any employee or establishment survey. The methodology described here 
is largely based on the approach the Department used in the 2004 Final 
Rule. 69 FR 22196-209. All tables include estimates for 2013. Some 
tables also include estimates for 2005

[[Page 38551]]

(the first full calendar year after the most recent increase to the 
salary level was implemented) to demonstrate how the prevalence of the 
EAP exemption has changed from 2005 through 2013. Figure 2 illustrates 
how the U.S. civilian workforce was analyzed through successive stages 
to estimate the number of potentially affected EAP workers.
[GRAPHIC] [TIFF OMITTED] TP06JY15.001

ii. Data
    The estimates of EAP exempt workers are based on data drawn from 
the CPS MORG, which is sponsored jointly by the U.S. Census Bureau and 
the BLS. The CPS is a large, nationally representative sample of the 
labor force. Households are surveyed for four months, excluded from the 
survey for eight months, surveyed for an additional four months, then 
permanently dropped from the sample. During the last month of each 
rotation in the sample (month 4 and month 16), employed respondents 
complete a supplementary questionnaire (the MORG) in addition to the 
regular survey. This supplement contains the detailed information on 
earnings necessary to estimate a worker's exemption status.
    Although the CPS is a large scale survey, administered to 60,000 
households representing the entire nation, it is still possible to have 
relatively few observations when looking at subsets of employees, such 
as exempt workers in a specific occupation employed in a specific 
industry, or workers in a specific region. To increase the sample size, 
the Department pooled together three years of CPS MORG data (2011 
through 2013). Earnings for each 2011 and 2012 observation were 
inflated to 2013 dollars using the CPI-U, and the weight of each 
observation was adjusted so that the total number of potentially 
affected EAP workers in the pooled sample remained the same as the 
number represented by the 2013 CPS MORG. Thus, the pooled CPS MORG 
sample uses roughly three times as many observations to represent the 
same total number of workers in 2013. The additional observations allow 
the Department to better estimate certain attributes of the potentially 
affected labor force.
    Some assumptions had to be made to use these data as the basis for 
the analysis. For example, the Department eliminated workers who 
reported that their weekly hours vary and provided no additional 
information on hours worked. This was done because the Department 
cannot estimate impacts for these workers since it is unknown whether 
they work overtime and therefore unknown whether there would be any 
need to pay for overtime if their status changed from exempt to 
nonexempt. The Department reweighted the rest of the sample to account 
for this change (to keep the same total population estimates). This 
adjustment assumes that the distribution of hours worked by workers 
whose hours do not vary is representative of hours worked by workers 
whose hours do vary. The Department believes that without more 
information this is an appropriate assumption.\60\ To the extent these 
excluded workers are exempt, if they tend to work more overtime than 
other workers, then transfer payments, costs, and DWL may be 
underestimated. Conversely, if they work fewer overtime hours then 
transfer payments, costs, and DWL may be overestimated.
---------------------------------------------------------------------------

    \60\ This is justifiable because other employment 
characteristics are similar across these two populations. The share 
of all workers whose hours vary is 6.3 percent.

---------------------------------------------------------------------------

[[Page 38552]]

iii. Number of Workers Covered by the Department's Part 541 Regulations
    To estimate the number of workers covered by the FLSA and subject 
to the Department's part 541 regulations, the Department first excluded 
workers who are not protected by the FLSA or are not subject to the 
Department's regulations for a variety of reasons--for instance, they 
may not be covered by, or considered to be employees under, the FLSA. 
These workers include:
     Military personnel,
     unpaid volunteers,
     self-employed individuals,
     clergy and other religious workers, and
     federal employees (with a few exceptions described below).
    Many of these workers are excluded from the CPS MORG: members of 
the military on active duty, unpaid volunteers, and the self-employed. 
For other categories that are not automatically excluded from the CPS 
data, such as unpaid workers, that is, workers with zero wages and 
earnings but who report being employed, the Department has implemented 
measures to screen them out.
    Religious workers were excluded from the analysis after being 
identified by their occupation codes: `clergy' (Census occupational 
code 2040), `directors, religious activities and education' (2050), and 
`religious workers, all other' (2060). Most employees of the federal 
government are covered by the FLSA but are not subject to the 
Department's part 541 regulations because their minimum wage and 
overtime pay are regulated by the Office of Personnel Management 
(OPM).\61\ See 29 U.S.C. 204(f). Exceptions exist for U.S. Postal 
Service employees, Tennessee Valley Authority employees, and Library of 
Congress employees. See 29 U.S.C. 203(e)(2)(A). These covered federal 
workers were identified and included in the analysis using occupation 
and/or industry codes.\62\ Employees of firms that have annual revenue 
of less than $500,000 and who are not engaged in interstate commerce 
are also not covered by the FLSA. The Department does not exclude them 
from the analysis because it has no reliable way of estimating the size 
of this worker population, although the Department believes it is a 
small percentage of workers. The 2004 Final Rule analysis similarly did 
not adjust for these workers.
---------------------------------------------------------------------------

    \61\ Federal workers are identified in the CPS MORG with the 
class of worker variable PEIO1COW.
    \62\ Postal Service employees were identified with Census 
industry code 6370. Tennessee Valley Authority employees were 
identified as federal workers employed in the electric power 
generation, transmission, and distribution industry (570) and in 
Kentucky, Tennessee, Mississippi, Alabama, Georgia, North Carolina, 
or Virginia. Library of Congress employees were identified as 
federal workers under Census industry `libraries and archives' 
(6770) and residing in Washington, DC.
---------------------------------------------------------------------------

    Table 5 presents the Department's estimates of the total number of 
workers, and the number of workers covered by the FLSA and subject to 
the Department's part 541 regulations in 2005 and 2013. The Department 
estimated that in 2013 there were 144.2 million wage and salary workers 
in the United States. Of these, 128.5 million were covered by the FLSA 
and subject to the Department's regulations (89.1 percent). The 
remaining 15.7 million workers were excluded from coverage by the FLSA 
for the reasons described above and delineated in Table 6.

 Table 5--Estimated Number of Workers Covered by the FLSA and Subject to the Department's Part 541 Regulations,
                                                  2005 and 2013
----------------------------------------------------------------------------------------------------------------
                                                                                    Subject to the Department's
                                                                     Civilian               regulations
                              Year                                  employment   -------------------------------
                                                                     (1,000s)         Number
                                                                                     (1,000s)         Percent
----------------------------------------------------------------------------------------------------------------
2005............................................................         142,126         122,716            86.3
2013............................................................         144,214     \a\ 128,511            89.1
----------------------------------------------------------------------------------------------------------------
\a\ Estimate uses pooled data for 2011-2013.


  Table 6--Reason Not Subject to the Department's Part 541 Regulations,
                                  2013
------------------------------------------------------------------------
                                                                Number
                           Reason                              (1,000s)
------------------------------------------------------------------------
Total......................................................       15,703
Self-employed and unpaid workers...........................       12,130
Religious workers..........................................          518
Federal employees \a\......................................        3,057
------------------------------------------------------------------------
Note: 2013 estimates use pooled data for 2011-2013.
\a\ Most employees of the federal government are covered by the FLSA but
  are not covered by part 541. Exceptions are for U.S. Postal Service
  employees, Tennessee Valley Authority employees, and Library of
  Congress employees.

iv. Number of Workers in the Analysis
    After limiting the analysis to workers covered by the FLSA and 
subject to the Department's regulations, several other groups of 
workers are identified and excluded from further analysis since they 
are unlikely to be affected by this proposed rule. These include:
     Blue collar workers,
     workers paid hourly, and
     workers who are exempt under certain other (non-EAP) 
exemptions.
    In 2013 there were 46.6 million blue collar workers (Table 7). 
These workers were identified in the CPS MORG data using data from the 
U.S. Government Accountability Office's (GAO) 1999 white collar 
exemptions report \63\ and the Department's 2004 regulatory impact 
analysis. Supervisors in traditionally blue collar industries are 
classified as white collar workers because their duties are generally 
managerial or administrative, and therefore they were not excluded as 
blue collar workers. In 2013, 76.1 million workers were paid on an 
hourly basis.\64\
---------------------------------------------------------------------------

    \63\ Fair Labor Standards Act: White Collar Exemptions in the 
Modern Work Place, GAO/HEHS-99-164, p. 40-41.
    \64\ The CPS MORG variable PEERNHRY is used to determine hourly 
status.
---------------------------------------------------------------------------

    Also excluded from further analysis were workers who are exempt 
under certain other (non-EAP) exemptions. Although some of these 
workers may also be exempt under the EAP exemptions, even if these 
workers lost their EAP exempt status they would remain exempt from the 
minimum wage and/or overtime pay provisions and thus were excluded from 
the analysis. In 2013 an estimated 4.2 million workers, including some 
agricultural and transportation workers, were excluded from further 
analysis because they were subject to another (non-EAP) overtime 
exemption. See Appendix A: Methodology for Estimating Exemption Status, 
for details on how this population was identified.

[[Page 38553]]



              Table 7--Estimated Number of Workers Covered by the FLSA and Subject to the Department's Regulations, 2005 and 2013 (1,000s)
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                              Reason Excluded \b\
                                                Subject to   Workers in    Excluded  -------------------------------------------------------------------
                     Year                       DOL's Part      the          from                                         Another exemption \c\
                                                 541 Reg.     analysis     analysis   Blue collar     Hourly   -----------------------------------------
                                                                \a\                     workers      workers    Agriculture  Transportation     Other
--------------------------------------------------------------------------------------------------------------------------------------------------------
2005.........................................      122,716       39,689       83,027       46,245       74,192          773          1,944         1,006
2013.........................................      128,511       42,970       85,541       46,644       76,113          911          1,827         1,484
--------------------------------------------------------------------------------------------------------------------------------------------------------
Note: 2013 estimates use pooled data for 2011-2013.
\a\ Wage and salary workers who are white collar, salaried, and not eligible for another (non-EAP) overtime exemption.
\b\ Numbers do not add to total due to overlap.
\c\ Eligible for another (non-EAP) overtime pay exemption.

    The Department excluded some of these workers from the population 
of potentially affected EAP workers in the 2004 Final Rule, but not all 
of them. Agricultural and transportation workers are two of the largest 
groups of workers excluded from this analysis, and they were similarly 
excluded in 2004. Agricultural workers were identified by occupational-
industry combination.\65\ Transportation workers were defined as those 
who are subject to the following FLSA exemptions: section 13(b)(1), 
section 13(b)(2), section 13(b)(3), section 13(b)(6), or section 
13(b)(10). This methodology is the same as in the 2004 Final Rule and 
is explained in Appendix A. The Department excluded 911,000 
agricultural workers and 1.8 million transportation workers from the 
analysis. The remaining 1.5 million excluded workers are included in 
multiple FLSA minimum wage and overtime exemptions and are detailed in 
Appendix A. However, of these 1.5 million workers, all but 28,000 are 
either blue collar or hourly and thus the impact of excluding these 
workers is negligible.
---------------------------------------------------------------------------

    \65\ In the 2004 Final Rule all workers in agricultural 
industries were excluded. 69 FR 22197. Here only workers also in 
select occupations were excluded since not all workers in 
agricultural industries qualify for the agricultural overtime pay 
exemptions. This method better approximates the true number of 
exempt agricultural workers and provides a more conservative--i.e., 
greater--estimate of the number of affected workers.
---------------------------------------------------------------------------

    For 2013 there were a total of 85.5 million workers excluded from 
the analysis for the reasons denoted above. These eliminations left 
43.0 million workers covered by the FLSA and potentially affected by 
this proposed rulemaking.
v. Number of Potentially Affected EAP Workers
    After excluding workers not subject to the Department's FLSA 
regulations and workers who are unlikely to be affected by this 
proposed rulemaking (i.e., blue collar workers, workers paid hourly, 
workers who are subject to another (non-EAP) overtime exemption), the 
Department estimated the number of workers for whom employers might 
claim the EAP exemptions. There are two ways a worker can lose overtime 
protection pursuant to the EAP exemptions: the standard EAP test and 
the HCE test. To be exempt under the standard EAP test the employee 
must:
     Be paid a predetermined and fixed salary that is not 
subject to reductions because of variations in the quality or quantity 
of work performed (the salary basis test),
     earn at least a designated salary amount; the salary level 
has been set at $455 per week since 2004 (the salary level test), and
     perform work activities that primarily involve executive, 
administrative, or professional duties as defined by the regulations 
(the duties test).
    The HCE test requires the employee to pass the same standard salary 
basis and salary level tests. However, the HCE duties test is much less 
restrictive than the standard duties test, and the employee must earn 
at least $100,000 in total annual compensation, including at least $455 
per week paid on a salary or fee basis, while the balance may be paid 
as nondiscretionary bonuses and commissions.
    Hourly computer employees who earn at least $27.63 per hour and 
perform certain duties are exempt under section 13(a)(17) of the FLSA. 
These workers are considered part of the EAP exemptions but were 
excluded from the analysis because they are paid hourly and will not be 
affected by this proposed rulemaking (these workers were similarly 
excluded in the 2004 analysis). Salaried computer workers are exempt if 
they meet the salary and duties tests applicable to the EAP exemptions, 
and are included in the analysis since they will be impacted by this 
proposed rulemaking.
    Additionally, administrative and professional employees may be paid 
on a fee basis,\66\ as opposed to a salary basis, at a rate of at least 
the amount specified by the Department in the regulations. However, the 
CPS MORG does not identify workers paid on a fee basis (only hourly or 
non-hourly). Thus in the analysis, workers paid on a fee basis are 
considered with non-hourly workers and consequently classified as 
``salaried'' (as was done in the 2004 Final Rule).
---------------------------------------------------------------------------

    \66\ Payment on a ``fee basis'' occurs where an employee is paid 
an agreed sum for a single job regardless of the time required for 
its completion. Sec.  541.605(a). Salary level test compliance for 
fee basis employees is assessed by determining whether the hourly 
rate for work performed (i.e., the fee payment divided by the number 
of hours worked) would total at least $455 per week if the employee 
worked 40 hours. Sec.  541.605(b).
---------------------------------------------------------------------------

    Weekly earnings are also available in the data, which allowed the 
Department to identify which workers passed the salary level tests.\67\ 
The CPS MORG data do not capture information about job duties. 
Therefore, to determine whether a worker met the duties test, the 
Department used an analysis performed by officials from the WHD in 1998 
in response to a request from the GAO. Because WHD enforces the FLSA's 
overtime requirements and regularly assesses workers' exempt status, 
WHD's representatives were uniquely qualified to provide the analysis. 
The analysis was used in both the GAO's 1999 white collar exemptions 
report \68\ and the Department's 2004 regulatory impact analysis. See 
69 FR 22198.
---------------------------------------------------------------------------

    \67\ The CPS MORG variable PRERNWA, which measures weekly 
earnings, is used to identify weekly salary. The CPS variable 
includes nondiscretionary bonuses and commissions, which do not 
count toward the standard salary level test. This discrepancy 
between the earnings variable used and the FLSA definition of salary 
may cause a slight overestimate of the number of workers estimated 
to meet the standard test.
    \68\ Fair Labor Standards Act: White Collar Exemptions in the 
Modern Work Place. (1999). GAO/HEHS-99-164, p. 40-41.
---------------------------------------------------------------------------

    WHD's representatives examined 499 occupational codes, excluding 
nine that

[[Page 38554]]

were not relevant to the analysis for various reasons (one code was 
assigned to unemployed persons whose last job was in the Armed Forces, 
some codes were assigned to workers who are not FLSA covered, others 
had no observations). Of the remaining occupational codes, WHD's 
representatives determined that 251 occupational codes likely included 
EAP exempt workers and assigned one of four probability codes 
reflecting the estimated likelihood, expressed as ranges, that a worker 
in a specific occupation would perform duties required to meet the EAP 
duties tests. The Department supplemented this analysis in the 2004 
Final Rule regulatory impact analysis when the HCE exemption was 
introduced. The Department modified the four probability codes for 
highly paid workers based upon our analysis of the provisions of the 
highly compensated test relative to the standard duties test (Table 8). 
To illustrate, WHD representatives assigned exempt probability code 3 
to the occupation ``first-line supervisors/managers of construction 
trades and extraction workers'' (Census code 6200), which indicates 
that a worker in this occupation has a 10 to 50 percent likelihood of 
meeting the standard EAP duties test. However, if that worker earns at 
least $100,000 annually, he or she has between a 58.4 percent and 60 
percent probability of being exempt under the shorter HCE test.
    The occupations identified by the GAO in 1999 and used by the 
Department in the 2004 Final Rule map to an earlier occupational 
classification scheme (the 1990 Census Occupational Codes). Therefore, 
for this proposed rule an occupational crosswalk was used to map the 
previous occupational codes to the 2002 Census Occupational Codes which 
are used in the CPS MORG 2002 through 2010 data, and to the 2010 Census 
Occupational Codes which are used in the CPS MORG 2011 through 2013 
data.\69\ If a new occupation is comprised of more than one previous 
occupation, then the new occupation's probability code is the weighted 
average of the previous occupations' probability codes, rounded to the 
closest probability code.
---------------------------------------------------------------------------

    \69\ Crosswalks and methodology available at: https://www.census.gov/people/io/methodology/.

                         Table 8--Probability Worker in Category Passes the Duties Test
----------------------------------------------------------------------------------------------------------------
                                                       The Standard EAP Test               The HCE Test
                                                 ---------------------------------------------------------------
                Probability code                    Lower bound     Upper bound     Lower bound     Upper bound
                                                     (percent)       (percent)       (percent)       (percent)
----------------------------------------------------------------------------------------------------------------
0...............................................               0               0               0               0
1...............................................              90             100             100             100
2...............................................              50              90              94              96
3...............................................              10              50            58.4              60
4...............................................               0              10              15              15
----------------------------------------------------------------------------------------------------------------

    These codes provide information on the likelihood an employee met 
the duties test but they do not identify the workers in the CPS MORG 
who actually passed the test. Therefore, the Department designated 
workers as exempt or nonexempt based on the probabilities. For example, 
for every ten public relations managers, between five and nine were 
estimated to pass the standard duties test (based on probability 
category 2). However, it is unknown which of these ten workers are 
exempt; therefore, the Department must determine the status for these 
workers. Exemption status could be randomly assigned with equal 
probability, but this would ignore the earnings of the worker as a 
factor in determining the probability of exemption. The probability of 
qualifying for the exemption increases with earnings because higher 
paid workers are more likely to perform the required duties, an 
assumption adhered to by both the Department in the 2004 Final Rule and 
the GAO in its 1999 Report.\70\ The Department estimated the 
probability of exemption for each worker as a function of both earnings 
and the occupation's exempt probability category using a gamma 
distribution.\71\ Based on these revised probabilities, each worker was 
assigned exempt or nonexempt status based on a random draw from a 
binomial distribution using the worker's revised probability as the 
probability of success. Thus, if this method is applied to ten workers 
who each have a 60 percent probability of being exempt, six workers 
would be expected to be designated as exempt.\72\ However, which 
particular workers are designated as exempt may vary with each set of 
ten random draws. For details see Appendix A.
---------------------------------------------------------------------------

    \70\ For the EAP exemptions, the relationship between earnings 
and exemption status is not linear and is better represented with a 
gamma distribution. For the HCE exemption, the relationship between 
earnings and exemption can be well represented with a linear 
function because the relationship is linear at high salary levels 
(as determined by the Department in the 2004 Final Rule). Therefore, 
the gamma model and the linear model would produce similar results. 
See 69 FR 22204-08, 22215-16.
    \71\ The gamma distribution was chosen because, during the 2004 
revision, this non-linear distribution best fit the data compared to 
the other non-linear distributions considered (i.e., normal and 
lognormal). A gamma distribution is a general statistical 
distribution that is based on two parameters that control the scale 
(alpha) and shape (in this context, called the rate parameter, 
beta).
    \72\ A binominal distribution is frequently used for a 
dichotomous variable where there are two possible outcomes; for 
example, whether one owns a home (outcome of 1) or does not own a 
home (outcome of 0). Taking a random draw from a binomial 
distribution results in either a zero or a one based on a 
probability of ``success'' (outcome of 1). This methodology assigns 
exempt status to the appropriate share of workers without biasing 
the results with manual assignment.
---------------------------------------------------------------------------

    The Department estimated that of the 43.0 million workers 
considered in the analysis, 28.5 million qualified for the EAP 
exemptions (Table 9). However, some of these workers were excluded from 
further analysis because they would not be affected by the proposed 
rule. This excluded group contains workers in named occupations who are 
not required to pass the salary requirements (although they must still 
pass the duties tests) and therefore whose exemption status is not 
dependent on their earnings. These occupations include physicians 
(identified with Census occupation codes 3010, 3040, 3060, 3120), 
lawyers (2100), teachers (occupations 2200-2550 and industries 7860 or 
7870), academic administrative personnel (school counselors (occupation 
2000 and industries 7860 or 7870) and educational administrators 
(occupation 0230 and industries 7860 or 7870)), and outside sales 
workers (a subset of occupation 4950). Out of the 28.5

[[Page 38555]]

million workers who are EAP exempt, 7.1 million, or 25.1 percent, were 
in named occupations in 2013. Thus these workers would be unaffected by 
changes in the standard salary level test. The 21.4 million EAP exempt 
workers remaining in the analysis are referred to in this proposed 
rulemaking as ``potentially affected.'' In addition to the 21.4 million 
potentially affected EAP exempt workers, the Department estimates that 
an additional 6.3 million salaried white collar workers who do not 
satisfy the duties test and who currently earn at least $455 per week 
but less than the proposed salary level will have their overtime 
protection strengthened because their exemption status will be clear 
based on the salary test alone without the need to examine their 
duties.

            Table 9--Estimated Percentages of EAP Exempt Workers in Named Occupations, 2005 and 2013
----------------------------------------------------------------------------------------------------------------
                                                                                   EAP exempt in     % of EAP
                                                  Workers in the    EAP exempt         named         exempt in
                      Year                           analysis       (millions)      occupations        named
                                                  (millions) \a\                  (millions) \b\    occupations
----------------------------------------------------------------------------------------------------------------
2005............................................            39.7            25.0             6.4            25.7
2013............................................            43.0            28.5             7.1            25.1
----------------------------------------------------------------------------------------------------------------
Note: 2013 estimates use pooled data for 2011-2013.
\a\ Wage and salary workers who are white collar, salaried, and not eligible for another (non-EAP) overtime
  exemption.
\b\ Workers not subject to a salary level test includes, but is not limited to, teachers, academic
  administrative personnel, physicians, lawyers, and judges.

    There are three groups of workers who lose minimum wage and 
overtime protections under the EAP exemptions: (1) Those passing just 
the standard EAP tests (i.e., passing the standard duties test, the 
salary basis test, and the standard salary level test and not passing 
the HCE tests); (2) those passing just the HCE tests (i.e., passing the 
HCE duties test, salary basis test, and the total compensation test and 
not passing the standard duties tests); and (3) those passing both 
tests. Based on analysis of the occupational codes and CPS earnings 
data, the Department has concluded that in 2013, of the 21.4 million 
potentially affected EAP workers, approximately 15.7 million passed 
only the standard EAP test, 5.6 million passed both the standard and 
the HCE tests, and approximately 75,000 passed only the HCE test (Table 
10). When impacts are discussed in section VII.D., workers who pass 
both tests will be considered with those who pass only the standard 
salary level test because this salary level test is more restrictive 
(i.e., the worker may continue to pass the standard salary level test 
even if he or she no longer passes the HCE compensation test).

        Table 10--Estimated Number of Workers Exempt Under the EAP Exemptions by Test Type, 2005 and 2013
----------------------------------------------------------------------------------------------------------------
                                                            Potentially affected EAP workers (millions)
                                                 ---------------------------------------------------------------
                      Year                                         Pass standard     Pass both     Pass HCE test
                                                       Total         test only         tests           only
----------------------------------------------------------------------------------------------------------------
2005............................................            18.6            15.8             2.8            0.03
2013............................................            21.4            15.7             5.6            0.08
----------------------------------------------------------------------------------------------------------------
Note: 2013 estimates use pooled data for 2011-2013.

vi. Characteristics of Potentially Affected EAP Workers
    After estimating the population of workers who are subject to the 
EAP exemptions and potentially affected by this proposed rulemaking, 
the Department tabulated the characteristics of these workers. The 
characteristics considered and presented here include: industry of 
employment, occupation, and Metropolitan Statistical Area (MSA) status. 
As previously noted, the Department estimated 2013 values using CPS 
MORG data pooled from 2011-2013 in order to increase the sample size.
    Table 11 presents the estimated number of potentially affected EAP 
workers broken down into 13 major industry groups.\73\ The industry 
with the most potentially affected EAP workers was professional and 
business services, with 4.2 million potentially affected EAP workers. 
Other industries where a large number of workers were potentially 
affected are education and health services (3.4 million), financial 
activities (3.3 million), and manufacturing (3.3 million). The industry 
with the smallest number of potentially affected EAP workers was 
agriculture, forestry, fishing, and hunting (33,000).
---------------------------------------------------------------------------

    \73\ See Appendix B: Additional Tables, for potentially affected 
workers categorized into the more detailed 51 industry group 
classifications.
---------------------------------------------------------------------------

    Looking at exemption status by occupation, 10.8 million workers 
employed in the management, business, and financial occupations were 
potentially affected; this occupation category accounts for roughly 
half of all potentially affected EAP workers. Professional and related 
occupations also employed many of the potentially affected EAP workers 
(7.0 million, which is 32.9 percent of all potentially affected EAP 
workers).
    The Department considered MSA status because workers in cities and 
suburban areas tend to be paid more than workers in rural areas. The 
percentage of potentially affected EAP workers (92.0 percent) who live 
in MSAs is larger than the percentage of the total workforce (85.8 
percent) who live in MSAs.

[[Page 38556]]



 Table 11--Potentially Affected EAP Workers by Industry, Occupation, and
            MSA Status, Number and as Percent of Total, 2013
------------------------------------------------------------------------
                                            Potentially    As percent of
                                           affected EAP     potentially
    Industry, occupation, MSA status          workers      affected EAP
                                            (millions)        workers
------------------------------------------------------------------------
Total...................................            21.4           100.0
------------------------------------------------------------------------
                               By Industry
------------------------------------------------------------------------
Agriculture, forestry, fishing, &                   0.03             0.2
 hunting................................
Mining..................................            0.18             0.8
Construction............................            0.76             3.6
Manufacturing...........................            3.27            15.3
Wholesale & retail trade................            2.42            11.3
Transportation & utilities..............            0.80             3.7
Information.............................            0.90             4.2
Financial activities....................            3.30            15.4
Professional & business services........            4.20            19.6
Education & health services.............            3.41            15.9
Leisure & hospitality...................            0.75             3.5
Other services..........................            0.55             2.6
Public administration...................            0.83             3.9
------------------------------------------------------------------------
                              By Occupation
------------------------------------------------------------------------
Management, business, & financial.......           10.79            50.4
Professional & related..................            7.04            32.9
Services................................            0.19             0.9
Sales & related.........................            2.19            10.2
Office & administrative support.........            0.97             4.5
Farming, fishing, & forestry............            0.00             0.0
Construction & extraction...............            0.02             0.1
Installation, maintenance, & repair.....            0.05             0.2
Production..............................            0.10             0.5
Transportation & material moving........            0.04             0.2
------------------------------------------------------------------------
                              By MSA Status
------------------------------------------------------------------------
MSA.....................................           19.67           92.0%
Non-MSA.................................            1.62            7.6%
Not Identified..........................            0.09            0.4%
------------------------------------------------------------------------
Note: Pooled data for 2011-2013.

C. Determining the Revised Salary Level Test Values

i. Background
    The Department proposes to set the EAP standard salary level at the 
40th percentile of the weekly earnings distribution for all full-time 
salaried workers and to set the HCE compensation test equal to the 90th 
percentile (at an annual salary equivalent) of this distribution. These 
methods were used because they generate salary levels that (1) were 
deemed to be appropriate in distinguishing between workers who should 
and should not be exempt; (2) are easy to calculate and thus easy to 
replicate, creating transparency through simplicity; and (3) generate 
consistent salary levels.\74\ The Department believes that setting the 
standard salary level at the 40th percentile earnings ($921 per week) 
allows for reliance on the current standard duties test without an 
unacceptably high risk of overtime-eligible employees being 
misclassified as EAP exempt and denied overtime protection. 
Additionally, the Department believes that setting the standard salary 
level at the 40th percentile earnings will not result in an 
unacceptably high risk that employees performing bona fide EAP duties 
will become entitled to overtime protection by virtue of the salary 
test.
---------------------------------------------------------------------------

    \74\ On a quarterly basis, BLS publishes a table of deciles of 
the weekly wages of full-time salaried workers, calculated using CPS 
data, which will provide employers with information on changes in 
salary levels prior to the annual updates. https://www.bls.gov/cps/research_series_earnings_nonhourly_workers.htm.
---------------------------------------------------------------------------

    The methodologies used to revise the EAP salary levels have varied 
somewhat across the seven updates to the salary level test since it was 
implemented in 1938. To guide the determination of the proposed salary 
level, the Department considered methodologies used previously to 
revise the EAP salary levels. In particular, the Department focused on 
the 1958 revisions and the most recent revisions in 2004. The 1958 
methodology is particularly instructive in that it synthesized previous 
approaches to setting the salary level, and the basic structures it 
adopted have been a touchstone in subsequent rulemakings (with the 
exception of 1975).
The 1958 Revisions
    In 1958, the Department updated the salary levels based on a 1958 
Report and Recommendations on Proposed Revision of Regulations, Part 
541, by Harry S. Kantor (Kantor Report). To determine the revised 
salary levels the Department looked at data collected during WHD 
investigations on actual salaries paid to exempt EAP employees, grouped 
by geographic region, industry groups, number of employees, and size of 
city. The Department then set the salary level so that no more than 
about 10 percent of those in the lowest-wage

[[Page 38557]]

region, lowest-wage industry, smallest establishment group, or smallest 
city group would fail to meet the test. Kantor Report at 
6.75 76 This methodology is referred to as the Kantor method 
and the Department followed a similar methodology in setting the salary 
levels in 1963 and 1970.
---------------------------------------------------------------------------

    \75\ The Kantor method was based on an analysis of a survey of 
exempt workers as determined by investigations conducted by WHD. 
Subsequent analyses, including both the 2004 rulemaking and this 
proposed rule, have estimated exempt status using multiple data 
sources.
    \76\ Because the salary level test is likely to have the largest 
impact on the low-wage categories of the economy (e.g., low-wage 
regions and industries), salaries in those regions/industries were 
selected as the basis for the required salary level under the Kantor 
method.
---------------------------------------------------------------------------

The 2004 Revisions
    A significant change in 2004 from the Kantor method was that the 
salaries of both exempt and nonexempt full-time salaried workers in the 
South and retail industry were used to determine levels (hereafter 
referred to as the 2004 method), rather than the salaries of exempt 
workers only. However, because the salaries of exempt workers on 
average are higher than the salaries of all full-time salaried workers, 
the Department selected a higher earnings percentile for full-time 
salaried workers. Based on the Department's 2004 analysis, the 20th 
percentile of earnings for exempt and nonexempt full-time salaried 
workers in the South and retail industry achieved a result very similar 
to the 10th percentile for workers in the lowest-wage regions and 
industries who were estimated to be exempt. 69 FR 22169.
ii. Proposed Methodology for the Standard Salary Level
    The Department proposes to set the standard salary level at the 
40th percentile of the distribution of weekly earnings for all full-
time salaried workers nationwide. For the purposes of this proposed 
rulemaking, the Department relied on BLS calculations of the dollar 
value of the 40th earnings percentile from the CPS MORG data. BLS 
limited the population to salaried workers who work at least 35 hours 
per week and determined the specified percentile of the resulting 
weighted weekly earnings distribution.\77\
---------------------------------------------------------------------------

    \77\ The Census Bureau publishes a public-use version of the CPS 
MORG data, which is very similar to the data used by BLS but 
involves a few changes to protect respondents' confidentiality. The 
salary level found with the public-use files is only very slightly 
different from that obtained with the confidential data.
---------------------------------------------------------------------------

    This methodology differs somewhat in specifics from previous 
revisions to the salary levels but the general concept holds: define a 
relevant population of workers, estimate an earnings distribution for 
that population, then set the salary level to a designated percentile 
of that distribution in order for the salary to serve as a meaningful 
line of demarcation between those Congress intended to protect and 
those who may qualify for exemption. The proposed method continues the 
evolution of the Department's approach from the Kantor method to the 
2004 method.
    The Department spent considerable time evaluating the previous 
methodologies. Where the proposed methodology differs from past 
methodologies, the Department believes the proposed methodology is an 
improvement. The Department compared the proposed method with the past 
methods, and the reasons for selecting the proposed method are detailed 
in the rest of this section.
The Kantor and 2004 Methods
    The Department replicated the Kantor method and the 2004 method to 
evaluate and compare them to the proposed methodology.\78\ Although the 
Department was able to replicate the 1958 and 2004 methods reasonably 
well, we could not completely replicate those methods due to changes in 
data availability, occupation classification systems, and incomplete 
documentation. In general, there are four steps in the process:
---------------------------------------------------------------------------

    \78\ The Department followed the same methodology used in the 
2004 Final Rule for estimating the Kantor method with minor 
adjustments. In an attempt to more accurately estimate the Kantor 
method, for example, this analysis included non-MSAs as a low-wage 
sector as Kantor did but the 2004 revisions did not.
---------------------------------------------------------------------------

    1. Identify workers likely to be members of the population of 
interest.
    2. Further narrow the population of interest by distinguishing that 
sub-population employed in low-wage categories.
    3. Estimate the distribution of earnings for these workers.
    4. Identify the salary level that is equal to a pre-determined 
percentile of the distribution.
    The population of workers considered for purposes of setting the 
salary level depends on whether the 2004 method or the Kantor method is 
used. In replicating both methods, we limited the population to workers 
subject to the FLSA and covered by the Department's part 541 
provisions, and excluded EAP exempt workers in named occupations, and 
those exempt under another (non-EAP) exemption. For the 2004 method, 
the Department further limited the population to full-time salaried 
workers, and for the Kantor method we further limited the population of 
interest by only including those workers determined as likely to be EAP 
exempt (see more detailed methodology explanations in section VII.B. 
and Appendix A).
    During the 2004 revisions the Department identified two low-wage 
categories: The South (low-wage geographic region), and the retail 
industry (low-wage industry). For this proposed rule the Department 
identified low-wage categories by comparing average weekly earnings 
across categories for the populations of workers used in the Kantor 
method and the 2004 method. The South was determined to be the lowest-
wage region and was used for the 2004 method; however, the Department 
chose to use a more detailed geographical break-down for the Kantor 
method to reflect the geographic categories Kantor used. Therefore, for 
the Kantor method the East South Central Division is considered the 
lowest-wage geographical area.\79\ The Department found that the 
industry with the lowest mean weekly earnings depends on whether the 
Kantor method or the 2004 method's population was used. Therefore, 
three industries are considered low-wage: Leisure and hospitality, 
other services, and public administration. The Department also 
considered non-MSAs as a low-wage sector in the Kantor method. The 2004 
revision did not consider population density but the Kantor method 
examined earnings across population size groups. In conclusion, the 
2004 method looks at workers in the South and low-wage industries 
whereas the Kantor method looks at workers in the East South Central 
Division, non-MSAs, and the three low-wage industries.
---------------------------------------------------------------------------

    \79\ The East South Central Division is a subset of the South 
and includes Alabama, Kentucky, Mississippi, and Tennessee. If the 
South is used instead, the resulting salary levels would increase 
slightly.
---------------------------------------------------------------------------

    Next, the Department estimated the distributions of weekly earnings 
of two populations: (1) Workers who are in at least one of the low-wage 
categories and in the Kantor population, and (2) workers who are in at 
least one of the low-wage categories and in the 2004 population. From 
these distributions, alternate salary levels were identified based on 
pre-determined percentiles. For the Kantor method, the salary level for 
the long duties test is identified based on the 10th percentile of 
weekly earnings for the relevant population of likely EAP exempt 
workers, while the 2004 method salary level is identified based on the 
20th percentile of weekly

[[Page 38558]]

earnings for the relevant population of both exempt and nonexempt 
salaried workers. Using 2013 CPS MORG data, the 2004 method resulted in 
a salary level of $577 per week and the Kantor method resulted in a 
salary level of $657 per week. Table 12 presents the distribution of 
weekly earnings used to estimate the salary levels under the proposed 
method, the 2004 method, and the Kantor method.

                                  Table 12--Weekly Earnings Distribution, 2013
----------------------------------------------------------------------------------------------------------------
                                       Distribution of weekly earnings      Distribution of annual earnings \a\
                                   -----------------------------------------------------------------------------
            Percentile               Full-Time       2004        Kantor     Full-Time       2004        Kantor
                                      Salaried    Method \b\   Method \c\    Salaried    Method \b\   Method \c\
----------------------------------------------------------------------------------------------------------------
5.................................         $378         $330         $577      $19,656      $17,148      $30,000
10................................          490          416          657       25,480       21,632       34,176
15................................          586          500          721       30,472       26,000       37,500
20................................          645          577          780       33,540       30,000       40,586
25................................          726          634          850       37,752       32,968       44,200
30................................          773          697          913       40,196       36,247       47,486
35................................          852          769          976       44,304       39,988       50,732
40................................          921          812        1,035       47,892       42,209       53,817
45................................          981          878        1,095       51,012       45,659       56,960
50................................        1,065          961        1,171       55,380       49,972       60,879
55................................        1,154        1,015        1,250       60,008       52,762       65,000
60................................        1,248        1,095        1,346       64,896       56,960       69,992
65................................        1,363        1,194        1,434       70,876       62,093       74,566
70................................        1,478        1,295        1,538       76,856       67,317       80,000
75................................        1,626        1,433        1,659       84,552       74,533       86,245
80................................        1,828        1,576        1,827       95,056       81,952       95,000
85................................        2,000        1,792        1,999      104,000       93,208      103,958
90................................        2,349        2,071        2,341      122,148      107,707      121,721
95................................        3,077        2,732        2,885      160,004      142,050      150,000
----------------------------------------------------------------------------------------------------------------
Note: Estimates for the full-time salaried percentiles are from BLS. Estimates for the 2004 method and the
  Kantor method are based on pooled CPS MORG public-use data for 2011-2013. The use of pooled data allows us to
  better represent both earnings distributions and the characteristics of affected EAP workers.
\a\ Weekly earnings multiplied by 52.
\b\ Full-time salaried workers in the South or employed in a low-wage industry (excludes workers not subject to
  the FLSA, not subject to the salary level test, and in agriculture or transportation).
\c\ Salaried, white collar workers who earn at least $455 per week, pass the EAP duties test, and either live in
  the East South Central Division or a non-MSA or are employed in a low-wage industry (excludes workers not
  subject to FLSA, not subject to the salary level test, and in agriculture or transportation).

iii. Rationale for the Methodology Chosen
    The salary level test has historically been intended to serve as an 
initial bright-line test for overtime eligibility for white collar 
employees. As discussed previously, however, there will always be white 
collar overtime-eligible employees who are paid above the salary 
threshold. A low salary level increases the number of these employees. 
The necessity of applying the duties test to these overtime-protected 
employees consumes employer resources, may result in misclassification 
(which imposes additional costs to employers and society in the form of 
litigation), and is an indicator of the effectiveness of the salary 
level. Similarly, there will always be employees performing bona fide 
EAP duties who are paid below the salary threshold; the inability of 
employers to claim the EAP exemption for these employees is also an 
indicator of the effectiveness of the salary level. Selecting the 
standard salary level will inevitably affect the number of workers 
falling into each of these two categories. The Kantor method sought to 
minimize the number of white collar employees who pass the duties test 
but were excluded from the exemption by the salary threshold and 
therefore set the salary level at the bottom 10 percent of exempt EAP 
employees in low wage regions and industries so as to prevent 
``disqualifying any substantial number of such employees.'' Kantor 
Report at 5; see Weiss Report at 9. This method was based on the long/
short test structure, in which employees paid at lower salary levels 
were protected by significantly more rigorous duties requirements than 
are part of the current standard duties test. This approach, however, 
does not take into sufficient account the inefficiencies of applying 
the duties test to large numbers of overtime-eligible white collar 
employees and the possibility of misclassification of those employees 
as exempt.
    In this rulemaking, the Department wants to correct for the 
elimination of the long duties test and set a salary level that 
appropriately classifies white collar workers as entitled to minimum 
wage and overtime protection or potentially exempt. Thus the 
Department's proposed standard salary level is higher than the level 
the Kantor or 2004 methods would generate but still lower than the 
historical average for the short test. Setting the salary level at the 
40th percentile of weekly earnings for full-time salaried workers will 
reduce the number of employees subject to the standard duties test by 
raising the salary threshold; the Department believes that this will 
simplify the determination of exemption status for employers and will 
result in reduced misclassification of overtime-eligible white collar 
workers as exempt and reduced litigation. At the 40th percentile, 10.6 
million white collar employees would no longer be subject to potential 
litigation over the duties they perform (4.6 million currently EAP 
exempt employees who would be newly entitled to overtime due to the 
increase in the salary threshold and 6.0 million overtime-eligible 
white collar employees who are paid between $455 and $921 per week 
whose exemption status would no longer depend on the application of the 
duties test). The proposed salary level will therefore more efficiently 
distinguish between employees who may meet the duties requirement of 
the

[[Page 38559]]

EAP exemption and those who do not, without necessitating a return to 
the more detailed long duties test.
    The proposed salary level also affects the likelihood of workers 
being misclassified as exempt from overtime pay. This provides an 
additional measure of the effectiveness of the salary level as a 
bright-line test delineating exempt and nonexempt workers. The 
Department estimated the number of workers misclassified as exempt as 
the number of salaried white collar workers who: Earn at least $455 per 
week; do not satisfy the EAP duties tests; are not in a named 
occupation (or exempt under another (non-EAP) exemption); usually work 
overtime; and do not usually receive overtime pay.\80\ The Department 
estimates that almost 20 percent of the 11.6 million salaried white 
collar workers who fail the duties test are misclassified as exempt. 
The Department estimates that at the proposed salary level, the number 
of overtime-eligible white collar workers earning at or above the 
salary level will decrease by 6.0 million, and that approximately 
806,562 (13.5 percent) of these workers are currently misclassified as 
exempt.
---------------------------------------------------------------------------

    \80\ Based on workers' response to the CPS-MORG question 
concerning whether they receive overtime pay, tips, or commissions 
at their job (``PEERNUOT'' variable).
---------------------------------------------------------------------------

    In this section the Department assesses the impact of the standard 
salary level as a bright-line test for the EAP exemptions by examining: 
(1) The number of white collar workers who pass the salary level test 
but not the duties test and (2) the number of white collar workers who 
pass the duties test but not the salary level test. The Department 
makes this assessment at the current salary level and the proposed 
salary level, while holding all other factors determining exempt status 
constant (e.g., not considering whether the duties test is correctly 
applied or potential employer response to the change in the salary 
level test). Examining the impact of the salary threshold in isolation 
from the application of the duties test or employer adjustments to pay 
or hours does not provide a complete picture of the impact of a new 
salary threshold. It does, however, allow the Department to evaluate 
the effectiveness of the salary level in protecting overtime-eligible 
white collar employees without unduly excluding from the exemption 
employees performing EAP duties.
    In order to calculate the potential impact on the two groups of 
workers, the Department estimated: (1) The number of salaried white 
collar workers who are eligible for overtime pay because they do not 
pass the standard EAP duties test, but earn above a specific salary 
level; and (2) the number of salaried white collar workers who satisfy 
the standard duties test but earn less than a specific standard salary 
level.\81\ These numbers were estimated at the current salary level 
($455) and the proposed standard salary level of the 40th percentile of 
weekly wages of all full-time salaried workers ($921).
---------------------------------------------------------------------------

    \81\ These populations are limited to salaried, white collar 
workers subject to the FLSA and the Department's part 541 
regulations, and not eligible for another (non-EAP) exemption, not 
in a named occupation, and not HCE only.
---------------------------------------------------------------------------

    As a benchmark, the Department estimates that at the current 
standard salary threshold, there are 11.6 million salaried white collar 
workers who fail the standard duties test and are therefore overtime 
eligible, but earn at least the $455 threshold, while there are only 
845,500 salaried white collar workers who pass the standard duties test 
but earn less than the $455 level. Thus the number of white collar 
workers who pass the current salary threshold test but not the duties 
test is nearly 14 times the number of white collar workers who pass the 
duties test but are paid below the salary threshold. This underscores 
the large number of overtime-eligible workers for whom employers must 
perform a duties analysis, and who may be at risk of misclassification 
as EAP exempt. If the salary threshold were raised to the 40th 
percentile, the number of overtime-eligible salaried white collar 
workers who would earn at least the threshold but do not pass the 
duties test would be reduced to 5.6 million. At the 40th percentile, 
the number of salaried white collar workers who would pass the standard 
duties test but earn less than the 40th percentile would be 4.6 million 
(approximately 25 percent of all white collar salaried employees who 
pass the standard duties test). While this number is higher than the 
number of such employees under the Kantor method, it includes employees 
who would not have passed the more rigorous long duties test and 
therefore were not included under that approach.

[[Page 38560]]

[GRAPHIC] [TIFF OMITTED] TP06JY15.002

    As illustrated in Figure 3, as the salary level increases there is 
a decrease in the share of overtime-eligible white collar workers for 
whom employers would be required to make an assessment under the duties 
test and who would be subject to possible misclassification. At the 
same time, as the salary level increases there is an increase in the 
share of white collar workers who pass the duties test but are screened 
from exemption by the salary threshold. At the current salary level, 
there is a very large gap between white collar workers who are overtime 
eligible but earn at least the threshold (about 85 percent of all 
salaried white collar workers who fail the duties test are paid at 
least $455 per week) and white collar workers who pass the standard 
duties test but do not meet the current salary level (about 4 percent 
of all salaried white collar workers who pass the duties test are paid 
less than $455 per week). At the proposed salary level of the 40th 
percentile of weekly earnings of full-time salaried workers, the 
percentage of overtime-eligible white collar workers who earn above the 
threshold (and thus would be at risk of misclassification) remains 
substantially higher than the percentage of white collar workers who 
pass the duties test but earn less than the salary threshold (and would 
become overtime protected).\82\ The salary threshold would have to be 
considerably higher (at a salary level of approximately $1,015, 
approximately the 50th percentile level of full-time salaried workers) 
before the percentage of white collar workers who earn less than the 
threshold but pass the duties test would equal the percentage who are 
overtime eligible but earn at least the salary threshold.
---------------------------------------------------------------------------

    \82\ Approximately 41 percent of white collar salaried workers 
who do not pass the duties test earn at least the proposed salary 
level ($921 per week). Conversely, approximately 25 percent of 
employees who pass the standard duties test (and 22 percent of 
employees who are currently exempt) earn less than the proposed 
salary level.
---------------------------------------------------------------------------

    The Department has also looked at the impact of the proposed salary 
level on these two groups of workers in low-wage (East South Central) 
and high-wage (Pacific) regions in addition to nationally.\83\ For the 
East South Central region, the salary level at which the percentages of 
the two groups are about equal is approximately $914 per week, while in 
the Pacific region, the salary at which the percentages of the two 
groups are equal is approximately $1,154 per week. The Department's 
proposed salary level of the 40th percentile of weekly earnings of 
full-time salaried workers ($921 per week) falls appropriately within 
this range. This supports the Department's use of nationwide data to 
set a salary level that is appropriate for classifying workers as 
entitled to minimum wage and overtime pay or potentially exempt, and 
takes into account the impact on employers in low-wage regions.
---------------------------------------------------------------------------

    \83\ Of the nine Census Region Divisions, the East South Central 
and Pacific divisions correspond to the divisions with the lowest 
and highest earnings using the Kantor method. The East South Central 
includes Alabama, Kentucky, Mississippi, and Tennessee. The Pacific 
includes Alaska, California, Hawaii, Oregon, and Washington.
---------------------------------------------------------------------------

    Appropriateness. The standard salary level serves as a bright-line 
test for employers, intended to assist in identifying those workers 
with duties that may make them truly bona fide executive, 
administrative, or professional employees. As explained in the 
preceding analysis, the Department has determined that setting the 
proposed standard salary level at the 40th percentile of earnings for 
full-time salaried workers ($921) appropriately balances the tradeoff 
between denying the exemption for employees who are currently exempt 
and exposing workers who fail to meet the duties test to the risk of 
misclassification as exempt. In the absence of a long duties test which 
limits the amount of nonexempt work that can be performed, the 
Department believes a salary level at or above the proposed salary 
level appropriately distinguishes between overtime-protected and 
potentially exempt employees. Of employees currently

[[Page 38561]]

exempt under the part 541 regulations, that is, those who are paid on a 
salary basis of at least $455 and meet the duties test, approximately 
78 percent earn at least the proposed level of $921 per week. 
Conversely, among overtime-eligible white collar employees (both 
salaried and hourly), approximately 75 percent earn less than the 
proposed salary level.
    Simplicity. The proposed method of basing the standard salary 
threshold on a particular percentile of weekly earnings for full-time 
salaried employees involves less estimation than previous updates, 
making it easier to implement, less prone to error, and more 
transparent than before. The proposed method reduces computation by 
simplifying the classification of workers to just two criteria: Wage or 
salaried, and full-time or part-time. Application of the Kantor method, 
in particular, would involve significant work to replicate since one 
would need to identify likely EAP exempt workers, a process which 
requires applying the standard duties test to determine the population 
of workers used in the earnings distribution. The proposed method is 
easier for stakeholders to replicate and understand because the 
standard duties test does not need to be applied to determine the 
population of workers used in the earnings distribution.
    Consistency. A method that produces very different salary levels in 
consecutive years will reduce confidence that the salary levels in any 
given year are optimal. Since 2003, the 40th percentile of full-time 
salaried workers' weekly earnings has increased by an average of 2.6 
percent annually. Similarly, the salary levels that would have been 
generated by the 2004 method increased 2.4 percent annually on average 
between 2003 and 2013. Conversely, since 2003 the salary levels that 
would have been generated by the Kantor method increased 3.6 percent on 
average annually. The larger growth rate for the Kantor method explains 
why despite the Kantor method and 2004 method generating very similar 
salary levels for the 2004 rulemaking, by 2013 these levels differ 
significantly (Kantor = $657; 2004 = $577). The primary reason the 
Kantor method generates a larger salary level than the 2004 method in 
2013 is because the Kantor method uses the value of the current salary 
level test to identify the population of workers from which the 
earnings distribution is determined. Therefore, the Kantor method 
limits the pool of workers in the sample to those who meet the required 
salary level before evaluating the salaries of workers in low-wage 
regions and industries, while the 2004 method looks to all salaried 
workers in the South and retail industry but does not exclude workers 
with salaries below the current salary level. For example, in 2003 the 
Kantor method population of interest was limited to workers earning at 
least $155 per week (the 1975 long test salary level); in this proposed 
rule the Kantor method's population was restricted to workers earning 
at least $455 per week. Therefore the population considered in Kantor's 
method changes each time the salary level is changed. The Department's 
proposed method, like the 2004 method, considers all full-time salaried 
workers and does not limit the pool to only those workers who meet the 
current salary level test, thus avoiding this potential shortcoming of 
the Kantor method.
    Based on the comparison of the characteristics of the methods 
reviewed in this section, the Department has determined that the 
proposed method, for the reasons identified, meets the objectives of 
appropriateness, simplicity, and consistency.
iv. Standard Salary Levels With Alternative Methodologies
    When assessing the effects of the proposed standard salary level on 
the U.S. economy, the Department also evaluated several alternatives. 
This section presents the alternative salary levels considered and the 
bases for identifying those alternative levels. As shown in Table 13, 
the alternative salary levels evaluated are:
     Alternative 1: Calculate the salary level by adjusting the 
2004 salary level of $455 for inflation from 2004 to 2013 as measured 
by the CPI-U. This results in a salary level of $561 per week.
     Alternative 2: Use the 2004 method to set the salary level 
at $577 per week.
     Alternative 3: Use the Kantor method to set the salary 
level at $657 per week.
     Alternative 4: Use the 50th earnings percentile of full-
time hourly and salaried workers. This results in a salary level of 
$776 per week.
     Alternative 5: Adjust the salary level from the Kantor 
method to reflect the historical ratio between the long and short test 
salary levels. This results in a salary level of $979 per week.
     Alternative 6: Use the 50th earnings percentile of full-
time salaried workers. This results in a salary level of $1,065 per 
week.
     Alternative 7: Adjust the 1975 short test salary level of 
$250 for inflation from 1975 to 2013. This results in a salary level of 
$1,083 per week.

     Table 13--Proposed Standard Salary Level and Alternatives, 2013
------------------------------------------------------------------------
                                 Salary level      Total increase \a\
          Alternative              (weekly/    -------------------------
                                   annually)         $            %
------------------------------------------------------------------------
Alternative #1: Inflate 2004      $561/$29,178          106         23.3
 levels.......................
Alternative #2: 2004 method...      577/30,000          122         26.8
Alternative #3: Kantor method.      657/34,176          202         44.4
Alternative #4: Median full-        776/40,352          321         70.5
 time hourly and salaried
 workers......................
Proposed (40th percentile full-     921/47,892          466        102.4
 time salaried)...............
Alternative #5: Kantor short        979/50,922          524        115.2
 test.........................
Alternative #6: Median full-      1,065/55,380          610        134.1
 time salaried................
Alternative #7: Inflate 1975      1,083/56,291          628        137.9
 short test level.............
------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Average weekly change between proposed/alternative salary level and
  the salary level set in 2004 ($455 per week).

    Alternatives 2 (2004 method) and 3 (Kantor method) were already 
discussed. Alternative 5 (Kantor short test) is also based on the 
Kantor method but, whereas alternative 3 generates the salary level 
associated with the long duties test, alternative 5 generates a level 
more closely resembling the salary associated with the short duties 
test. In the 2004 Final Rule, the Department replaced the structure of 
a short and a long duties test with a single standard

[[Page 38562]]

duties test based on the less restrictive short duties test, which had 
historically been paired with a higher salary level test. However, the 
Department set the standard salary level in 2004 at a level that was 
equivalent to the Kantor long test salary level, which was associated 
with the long duties test and limited the amount of nonexempt work that 
the employee could perform. In alternative 5, the Department therefore 
considered revising the standard salary level to approximate the short 
test salary that better matches the standard duties test. On average, 
the salary levels set in 1949 through 1975 were 149 percent higher for 
the short test than the long test. Therefore, the Department inflated 
the 2013 Kantor estimate of $657 by 149 percent, which generated a 
short salary level equivalent of $979.\84\ While the Department used 
the average difference between the Kantor long and short tests for this 
alternative, the ratio of the short to long salary tests ranged from 
approximately 130 percent to 180 percent between 1949 and 2004. The low 
end of this range would result in a salary of $854; the high end would 
result in a salary of $1,183.
---------------------------------------------------------------------------

    \84\ The Department estimated the average historic ratio of 149 
percent as the simple average of the fifteen historical ratios of 
the short duties salary level to the long duties salary level 
(salary levels were set in 5 years and in each year the salary level 
varied between the three exemptions: executive, administrative, and 
professional). If the Department had weighted the average ratio 
based on the length of time the historic salary levels were in 
effect, this would have yielded an average historic ratio of 152 
percent and a salary level of $999.
---------------------------------------------------------------------------

    Alternatives 1 (inflating the 2004 level) and 7 (inflating the 1975 
short test level) use similar approaches to each other. Both begin with 
an exemption salary level set in an earlier rulemaking, and use the 
CPI-U to adjust that salary level to account for inflation between the 
year it was set and 2013. Where the two approaches differ is in the 
selection of the starting point. Alternative 1 assumes the 2004 
standard salary level was set at an appropriate level, and that changes 
in earnings since that time can be reflected well by changes in prices. 
Alternative 1 is inappropriate because the salary level set in 2004 
does not fully account for changes in the sample and the change from 
long and short duties tests to a single standard test that is 
comparable to the old short duties test. Alternative 7 assumes that the 
1975 salary levels were set to a more appropriate level than the 2004 
levels; inflating the 1975 short duties test salary level to 2013 
results in a salary level of $1,083 per week. This alternative is 
inappropriate because it is based on interim salary rates. 40 FR 7091. 
Additionally, the Department thinks the salary level generated with 
this method is too high in light of the fact that there no longer is a 
long duties test with an associated lower salary level that employers 
may use to claim that employees are exempt.
    Alternatives 4 and 6 set the standard salary equal to the 50th 
percentile, or median, of weekly earnings for two groups of workers: 
full-time hourly and salaried workers and full-time salaried workers, 
respectively. These approaches are similar to the proposed method in 
that they set the salary level equal to a percentile of an earnings 
distribution. The 50th earnings percentile of all full-time hourly and 
salaried workers results in a salary level of $776. The Department 
concluded, however, that it would not be appropriate to include the 
wages of hourly workers in setting the EAP salary threshold and that 
the resulting salary level was too low to work effectively with the 
standard duties test. Selecting the 50th earnings percentile of full-
time salaried workers results in a standard salary level of $1,065, 
which is only $18 per week less than alternative 7. Like alternative 7, 
the Department believes that the salary level generated with this 
method is too high because there is no longer a long duties test with 
an associated lower salary level that employers may use to claim that 
employees are exempt.
    Section VII.D. will detail the transfers, costs, and benefits of 
the proposed salary levels and alternatives. A comparison of the costs 
and benefits justifies the Department's decision to propose a standard 
salary level of the 40th percentile of weekly earnings for all full-
time salaried workers ($921 per week).
v. Proposed Methodology for the HCE Total Annual Compensation Level
    The Department proposes to set the HCE compensation level equal to 
the annual equivalent of the 90th percentile of the distribution of 
earnings for all full-time salaried workers. BLS calculated the salary 
level from the CPS MORG data by limiting the population to non-hourly 
workers who work full-time (i.e., at least 35 hours per week) and 
determining the 90th percentile of the resulting weighted weekly 
earnings distribution. The 90th percentile of weekly earnings ($2,349) 
was then multiplied by 52 to determine the annual earnings equivalent 
($122,148). This mirrors the method used to set the standard salary 
level but uses a percentile towards the top of the earnings 
distribution to reflect the minimal duties criteria associated with the 
highly compensated exemption.
    The Department also evaluated the following alternative HCE 
compensation levels:
     HCE alternative 1: Leave the HCE compensation level 
unchanged at $100,000 per year.
     HCE alternative 2: Set the HCE compensation level at 
$150,000 per year, which is approximately the annualized level of 
weekly earnings exceeded by 6.3 percent of full-time salaried workers. 
This is the same percent of such workers that exceeded the HCE 
compensation level in 2004.
    The Department concluded that HCE alternative 1 was inappropriate 
because leaving the HCE compensation level unchanged at $100,000 per 
year would ignore more than 10 years of wage growth. In 2013, 
approximately 17 percent of full-time salaried workers earned at least 
$100,000 annually, more than twice the share who earned that amount in 
2004. Conversely, HCE alternative 2 would set the annual compensation 
level at $150,000.\85\ The Department believes this salary level would 
be too high to provide a meaningful alternative test for exemption. 
Thus, the Department believes its proposal to adjust the HCE total 
annual compensation to reflect the 90th percentile of earnings of full-
time salaried workers strikes the appropriate balance.
---------------------------------------------------------------------------

    \85\ This compensation level corresponds to the annual value of 
the highest weekly earnings reported in the CPS MORG public-use 
data.
---------------------------------------------------------------------------

D. Impacts of Revised Salary and Compensation Level Test Values

i. Overview
    Impacts due to the proposed increases in the EAP salary levels will 
depend on how employers respond. Employer response is expected to vary 
by the characteristics of the affected EAP workers. For workers who 
usually work 40 hours a week or less, the Department assumes that 
employers will reclassify these workers as overtime-eligible and will 
pay the same weekly earnings for the same number of hours worked. While 
these employees will become overtime eligible, employers can continue 
to pay their current salaries and need make no adjustments as long as 
the employees' hours do not exceed 40 hours in a workweek.\86\ For 
employees who work overtime, employers may: (1) Pay the required 
overtime premium for the current number of overtime hours based upon 
the current implicit regular rate of pay;

[[Page 38563]]

(2) reduce the regular rate of pay so total weekly earnings and hours 
do not change after overtime is paid; (3) eliminate overtime hours; (4) 
increase employees' salaries to the proposed salary level; or (5) use 
some combination of these responses. Transfers from employers to 
employees, direct employer costs, and DWL depend on how employers 
respond to the proposed rulemaking.
---------------------------------------------------------------------------

    \86\ Assuming the worker earns the minimum wage. Otherwise, 
wages and hours will be adjusted to reflect compliance with minimum 
wage requirements.
---------------------------------------------------------------------------

    The cost, benefit and transfer estimates appearing throughout this 
section represent nationwide aggregates. Given the potential for this 
proposed rule to have impacts that differ by region or industry, the 
Department invites detailed comment, data and analysis that would allow 
for estimation of impacts on a regional or industry basis.
ii. Summary of Quantified Impacts
    Table 14 presents the aggregated projected costs, transfers, and 
DWL associated with increasing the standard EAP salary level from $455 
per week to the 40th earnings percentile, $921 per week, and the HCE 
compensation level from $100,000 to the 90th earnings percentile, 
$122,148 annually (without automatic updating). The Department 
estimated that the direct employer costs of this proposal will total 
$592.7 million in the first year, with average annualized direct costs 
of $194.2 million per year over 10 years. In addition to the direct 
costs, this proposed rulemaking would also transfer income from 
employers to employees. Year 1 transfers would equal $1,482.5 million, 
with average annualized transfers estimated at $872.9 million per year 
over 10 years. Finally, the 10-year average annualized DWL was 
estimated to be $7.2 million.
    In order to increase the sample size and the reliability and 
granularity of results in this analysis, the Department used three 
years (2011-2013) of CPS MORG data to represent the 2013 labor market. 
Monetary values in 2011 and 2012 were inflated to 2013 dollars and the 
sample was reweighted to reflect the population of potentially affected 
workers in 2013. The potential employer costs due to reduced profits 
and additional hiring were not quantified but are discussed in section 
VII.D.iv.5.

 Table 14--Summary of Regulatory Costs and Transfers, Standard and HCE Salary Levels, Without Automatic Updating
                                                (millions 2013$)
----------------------------------------------------------------------------------------------------------------
                                                                  Future years \b\      Average annualized value
                                                             ---------------------------------------------------
               Cost/Transfer \a\                    Year 1                                3% Real      7% Real
                                                                 Year 2      Year 10        rate         rate
----------------------------------------------------------------------------------------------------------------
Direct Employer Costs:
    Regulatory familiarization.................       $254.5         $0.0         $0.0        $29.0        $33.9
    Adjustment.................................        160.1          1.1          0.1         18.4         21.5
    Managerial.................................        178.1        169.0         93.1        135.9        138.9
                                                ----------------------------------------------------------------
        Total direct costs \c\.................        592.7        170.0         93.1        183.2        194.2
----------------------------------------------------------------------------------------------------------------
Transfers from Employers to Workers \d\
    Due to minimum wage........................         46.7         44.0          9.9         27.9         29.3
    Due to overtime pay........................      1,435.8      1,017.1        490.2        815.7        843.6
                                                ----------------------------------------------------------------
        Total transfers \d\....................      1,482.5      1,061.2        500.1        843.6        872.9
----------------------------------------------------------------------------------------------------------------
DWL \e\                                          ...........  ...........  ...........  ...........  ...........
    DWL........................................          7.4          9.8          4.3          7.0          7.2
----------------------------------------------------------------------------------------------------------------
\a\ Additional costs and benefits of the rule that could not be quantified or monetized are discussed in the
  text.
\b\ These costs/transfers represent a range over the nine-year span.
\c\ Components may not add to total due to rounding.
\d\ This is the net transfer from employers to workers. There may also be transfers of hours and income from
  some workers to others.
\e\ DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. Since
  the transfer associated with the minimum wage is negligible compared to the transfer associated with overtime
  pay, the vast majority of this cost is attributed to the overtime pay provision.

iii. Affected EAP Workers
1. Overview
    Costs, transfer payments, DWL, and benefits of this proposed 
rulemaking depend on the number of affected EAP workers and labor 
market adjustments made by employers. The Department estimated there 
were 21.4 million potentially affected EAP workers, that is EAP workers 
who either (1) passed the salary basis test, the standard salary level 
test, and the standard duties test, or (2) passed the salary basis 
test, passed the standard salary level test, the HCE total compensation 
level test, and the HCE duties test. This number excludes workers in 
named occupations who are not subject to the salary tests or who 
qualify for another (non-EAP) exemption.
    The Department estimated that increasing the standard salary level 
from $455 per week to the 40th earnings percentile of all full-time 
salaried workers ($921 per week) would directly affect 4.6 million 
workers (i.e., the number of potentially affected workers who earn at 
least $455 per week but less than $921 per week). These affected 
workers compose 21.7 percent of potentially affected EAP workers. The 
Department also estimated that 36,000 workers would be directly 
affected by an increase in the HCE compensation level from $100,000 to 
the 90th earnings percentile (the number of potentially affected 
workers who earn between $100,000 and $122,148 annually and pass the 
minimal duties test but not the standard duties test; about 0.2 percent 
of the pool of potentially affected EAP workers).
    Table 15 presents the number of affected EAP workers, the mean 
number of overtime hours they work per week, and their average weekly 
earnings. The 4.6 million workers affected by the increase in the 
standard salary level average 1.6 hours of overtime per week and earn 
an average of $731 per week.

[[Page 38564]]

The average number of overtime hours is low because most of these 
workers (3.7 million) do not usually work overtime.\87\ However, the 
estimated 988,000 affected workers who regularly work overtime average 
11.1 hours of overtime per week. The 36,000 EAP workers affected by the 
proposed change in the HCE annual compensation level average 5.8 hours 
of overtime per week and earn an average of $2,103 per week.
---------------------------------------------------------------------------

    \87\ That is, workers who report they usually work 40 hours or 
less per week (identified with variable PEHRUSL1 in CPS MORG).
---------------------------------------------------------------------------

    Although most affected EAP workers who typically do not work 
overtime might experience little or no change in their daily work 
routine, those who regularly work overtime may experience significant 
changes. The Department expects that workers who routinely work some 
overtime or who earn less than the minimum wage are most likely to be 
tangibly impacted by the revised salary level.\88\ Employers might 
respond by: converting such employees to overtime eligible, paying at 
least the minimum wage, and paying the overtime premium; reducing 
overtime hours; reducing workers' regular wage rates (where the rate 
exceeds the minimum wage); increasing the employees' salary to the 
proposed salary level; or use some combination of these responses.
---------------------------------------------------------------------------

    \88\ A small proportion (0.3 percent) of affected EAP workers 
earns implicit hourly wages that are less than the applicable 
minimum wage (the higher of the state or federal minimum wage). The 
implicit hourly wage is calculated as an affected EAP employee's 
total weekly earnings divided by total weekly hours worked.

          Table 15--Number of Affected EAP Workers, Mean Overtime Hours, and Mean Weekly Earnings, 2013
----------------------------------------------------------------------------------------------------------------
                                                     Affected EAP workers \a\
                                                 --------------------------------  Mean overtime    Mean usual
           Type of affected EAP worker                Number                           hours          weekly
                                                     (1,000s)       % of total                       earnings
----------------------------------------------------------------------------------------------------------------
                                              Standard Salary Level
----------------------------------------------------------------------------------------------------------------
All affected EAP workers........................           4,646             100             1.6            $731
Earn less than the minimum wage \b\.............              12             0.3            36.4             529
Regularly work overtime.........................             988            21.3            11.1             743
Occasionally work overtime \c\..................             180             3.9             8.0             729
----------------------------------------------------------------------------------------------------------------
                                             HCE Compensation Level
----------------------------------------------------------------------------------------------------------------
All affected EAP workers........................            36.2             100             5.8           2,103
Earn less than the minimum wage \b\.............  ..............  ..............  ..............  ..............
Regularly work overtime.........................            14.5            40.1            14.3           2,119
Occasionally work overtime \c\..................             1.0             2.6             6.5           2,120
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection
  under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels).
\b\ The applicable minimum wage is the higher of the federal minimum wage and the state minimum wage. HCE
  workers will not be impacted by the minimum wage provision.
\c\ Workers who do not usually work overtime but did in the survey week. Mean overtime hours are actual overtime
  hours in the survey week.

    The Department considered two types of overtime workers in this 
analysis: regular overtime workers and occasional overtime workers.\89\ 
Regular overtime workers typically worked more than 40 hours per week. 
Occasional overtime workers typically worked 40 hours or less per week, 
but they worked more than 40 hours in the week they were surveyed. The 
Department considers these two populations separately in the analysis 
because labor market responses to overtime pay requirements may differ 
for these two types of workers.
---------------------------------------------------------------------------

    \89\ Regular overtime workers were identified in the CPS MORG 
with variable PEHRUSL1. Occasional overtime workers were identified 
in the CPS MORG with variables PEHRUSL1 and PEHRACT1.
---------------------------------------------------------------------------

    An estimated 181,000 occasional overtime workers will be affected 
by either the standard salary level or the HCE total annual 
compensation level increase in any given week (3.9 percent of all 
affected EAP workers). They averaged 8.0 hours of overtime per week. 
This group represents the number of workers with occasional overtime 
hours in the week the CPS MORG survey was conducted. In other weeks, 
these specific individuals may not work overtime but other workers, who 
did not work overtime in the survey week, may work overtime. Because 
the survey week is a representative week, the Department believes the 
prevalence of occasional overtime in the survey week, and the 
characteristics of these workers, is representative of other weeks 
(even though a different group of workers would be identified as 
occasional overtime workers in a different week).\90\
---------------------------------------------------------------------------

    \90\ The Department can estimate the average number of 
occasional overtime workers in any given week but cannot estimate 
the total number of individuals working occasional overtime in the 
year since the Department does not know how many weeks in a year a 
specific worker works overtime.
---------------------------------------------------------------------------

2. Characteristics of Affected EAP Workers
    In this section the Department examines the characteristics of 
affected EAP workers. Table 16 presents the distribution of affected 
workers across industries, occupations, and MSA status. The industry 
with the largest number of affected EAP workers was education and 
health services (1.0 million). The management, business, and financial 
occupation category accounted for the most affected EAP workers by 
occupation (2.1 million). A substantial majority of affected EAP 
workers resided in MSAs (4.1 million). Employers in non-MSAs and low-
wage industries may perceive a greater impact due to the lower wages 
and salaries typically paid in those areas and industries. However, 
because the vast majority of potentially affected workers reside in 
MSAs and do not work in low-wage industries, the Department believes 
that the proposed salary level is appropriate.

[[Page 38565]]



     Table 16--Estimated Number of Exempt Workers With the Current and Proposed Salary Levels, by Industry,
                                        Occupation, and MSA Status, 2013
----------------------------------------------------------------------------------------------------------------
                                                                    Potentially affected EAP workers (millions)
                                                                                        \a\
                                                                 -----------------------------------------------
                                                                                   With updated standard and HCE
              Industry, occupation, and MSA status                                            levels
                                                                    At current   -------------------------------
                                                                   salary levels                     Reduction
                                                                                    Number \b\       (affected
                                                                                                   workers) \c\
----------------------------------------------------------------------------------------------------------------
Total...........................................................           21.38           16.70            4.68
----------------------------------------------------------------------------------------------------------------
                                                   By Industry
----------------------------------------------------------------------------------------------------------------
Agriculture, forestry, fishing, & hunting.......................            0.03            0.03            0.01
Mining..........................................................            0.18            0.16            0.02
Construction....................................................            0.76            0.61            0.16
Manufacturing...................................................            3.27            2.86            0.41
Wholesale & retail trade........................................            2.42            1.76            0.66
Transportation & utilities......................................            0.80            0.64            0.16
Information.....................................................            0.90            0.71            0.18
Financial activities............................................            3.30            2.61            0.68
Professional & business services................................            4.20            3.46            0.73
Education & health services.....................................            3.41            2.41            0.99
Leisure & hospitality...........................................            0.75            0.49            0.26
Other services..................................................            0.55            0.36            0.18
Public administration...........................................            0.83            0.59            0.24
----------------------------------------------------------------------------------------------------------------
                                                  By Occupation
----------------------------------------------------------------------------------------------------------------
Management, business, & financial...............................           10.79            8.69            2.10
Professional & related..........................................            7.04            5.63            1.40
Services........................................................            0.19            0.11            0.08
Sales & related.................................................            2.19            1.57            0.62
Office & administrative support.................................            0.97            0.53            0.44
Farming, fishing, & forestry....................................            0.00            0.00            0.00
Construction & extraction.......................................            0.02            0.02            0.01
Installation, maintenance, & repair.............................            0.05            0.04            0.01
Production......................................................            0.10            0.08            0.02
Transportation & material moving................................            0.04            0.03            0.01
----------------------------------------------------------------------------------------------------------------
                                                  By MSA Status
----------------------------------------------------------------------------------------------------------------
MSA.............................................................           19.67           15.53            4.14
Non-MSA.........................................................            1.62            1.11            0.52
Not identified..................................................            0.09            0.06            0.02
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Workers who are white collar, salaried, not eligible for another (non-EAP) overtime exemption, and not in a
  named occupation.
\b\ Workers who continue to be exempt after the proposed increases in the salary levels (assuming affected
  workers' weekly earnings do not increase to the proposed salary level).
\c\ Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection
  under the proposed salary levels (if their weekly earnings do not increase to the proposed salary levels).

iv. Costs
1. Summary
    Three direct costs to employers were quantified in this analysis: 
(1) Regulatory familiarization costs; (2) adjustment costs; and (3) 
managerial costs. Regulatory familiarization costs are costs to learn 
about the change in the regulation and only occur in Year 1. Adjustment 
costs are costs incurred by firms to determine workers' exemption 
statuses, notify employees of policy changes, and update payroll 
systems. Managerial costs associated with this proposed rulemaking 
occur because employers may spend more time scheduling newly nonexempt 
employees and more closely monitor their hours to minimize or avoid 
paying the overtime premium.
    The Department estimated costs in Year 1 assuming the first year of 
the analysis was 2013. The Department estimated that in Year 1 
regulatory familiarization costs would equal $254.5 million, Year 1 
adjustment costs would sum to $160.1 million, and Year 1 managerial 
costs would total $178.1 million (Table 17). Total direct employer 
costs in Year 1 were estimated to equal $592.7 million. Adjustment 
costs and management costs are ongoing and will need to be projected 
for future years (section VII.D.x.).
    Costs that are not quantified are discussed in section VII.D.iv.5. 
Adjustment costs and managerial costs associated with automatically 
updating the standard salary level are discussed in section VII.E.iii.

[[Page 38566]]



               Table 17--Summary of Year 1 Direct Employer Costs of This Proposed Rule (Millions)
----------------------------------------------------------------------------------------------------------------
                                                                                        HCE
                      Direct employer costs                          Standard      Compensation        Total
                                                                   salary level        level
----------------------------------------------------------------------------------------------------------------
Regulatory familiarization \a\..................................  ..............  ..............          $254.5
Adjustment......................................................           158.8            $1.2           160.1
Managerial......................................................           176.0             2.1           178.1
                                                                 -----------------------------------------------
    Total direct costs..........................................           334.8             3.3           592.7
----------------------------------------------------------------------------------------------------------------
\a\ Regulatory familiarization costs are assessed jointly for the change in the standard salary level and the
  HCE compensation level.

2. Regulatory Familiarization Costs
    A change in the standard EAP weekly salary test to the proposed 
level would impose direct costs on businesses by requiring them to 
review the regulation. It is not clear whether regulatory 
familiarization costs are a function of the number of establishments or 
the number of firms. The Department believes that generally the 
headquarters of a firm will conduct the regulatory review for the 
entire company; however, some firms provide more autonomy to their 
establishments, and in such cases regulatory familiarization may occur 
at the establishment level. To be conservative, the Department uses the 
number of establishments in its cost estimate because this provides a 
larger cost estimate.
    The Department believes that all establishments will incur 
regulatory familiarization costs, even if they do not employ exempt 
workers, because all establishments will need to confirm whether this 
proposed rulemaking includes any provisions that may impact their 
workers. Firms with more affected EAP workers will likely spend more 
time reviewing the regulation than firms with fewer or no affected EAP 
workers (since a careful reading of the regulations will probably 
follow the initial decision that the firm is affected). However, the 
Department does not know the distribution of affected EAP workers 
across firms and so an average cost per establishment is used.
    No data were identified from which to estimate the amount of time 
required to review the regulation. The Department requests that 
commenters provide data if possible. For this NPRM, the Department 
estimated establishments will use on average one hour of time because 
the proposed regulation is narrowly focused on the salary level tests.
    To estimate the total regulatory familiarization costs, three 
pieces of information must be estimated: (1) A wage level for the 
employees reviewing the rule; (2) the number of hours each employee 
spends reviewing the rule; and (3) the number of establishments 
employing workers. The Department's analysis assumed that mid-level 
human resource workers with a median wage of $23.63 per hour will 
review the proposed rule.\91\ Assuming benefits are paid at a rate of 
45 percent of the base wage and one hour of time is required for 
regulatory familiarization, the average cost per establishment is 
$34.19.\92\ The number of establishments with paid employees in 2011 
was 7.44 million.\93\
---------------------------------------------------------------------------

    \91\ Calculated as the median wage in the CPS for workers with 
the occupation ``human resources, training, and labor relations 
specialists'' (0620) in 2013. The Department determined this 
occupation includes most of the workers who would conduct these 
tasks. Bureau of Labor Statistics, U.S. Department of Labor, 
Occupational Outlook Handbook, 2014-15 Edition, Human Resources 
Specialists and Labor Relations Specialists, available at: https://www.bls.gov/ooh/business-and-financial/human-resources-specialists-and-labor-relations-specialists.htm.
    \92\ The benefits-earnings ratio is derived from the BLS's 
Employer Costs for Employee Compensation (ECEC) data.
    \93\ Data for 2011 was the most recent available at the time of 
writing. Survey of U.S. Businesses 2011. Available at: https://www.census.gov/econ/susb/. Also included in the number of 
establishments incurring regulatory familiarization costs are the 
90,106 state and local governments reported in the 2012 Census of 
Governments: Employment Summary Report. Available at: https://www2.census.gov/govs/cog/g12_org.pdf.
---------------------------------------------------------------------------

    Regulatory familiarization costs in Year 1 were estimated to be 
$254.5 million ($34.19 per establishment x 1 hour x 7.44 million 
establishments).\94\ In future years, new firms will be formed and may 
incur regulatory familiarization costs. However, the Department 
believes the incremental cost of this regulation will be zero since new 
firms will only need to familiarize themselves with the updated law, 
instead of the old law.
---------------------------------------------------------------------------

    \94\ As previously noted, the Department chose to use the number 
of establishments rather than the number of firms to provide a more 
conservative estimate of the regulatory familiarization cost. Using 
the number of firms, 5.8 million, would result in a reduced 
regulatory familiarization cost estimate of $197.4 million in Year 
1.
---------------------------------------------------------------------------

3. Adjustment Costs
    A change in the EAP salary test to the proposed level will impose 
direct costs on firms by requiring them to re-determine the exemption 
status of employees, update and adapt overtime policies, notify 
employees of policy changes, and adjust their payroll systems. The 
Department believes the size of these costs will depend on the number 
of affected EAP workers and will occur in any year when the salary 
level is raised and exemption status is changed for some workers. To 
estimate adjustment costs three pieces of information must be 
estimated: (1) A wage level for the employees making the adjustments; 
(2) the amount of time spent making the adjustments; and (3) the 
estimated number of newly affected EAP workers. The Department again 
estimated that the average wage with benefits for human resources, 
training, and labor relations specialists is $34.19 per hour (as 
explained above). No applicable data were identified from which to 
estimate the amount of time required to make these adjustments.\95\ The 
Department requests that commenters provide any applicable data. For 
this NPRM, the Department chose to use one hour of time per affected 
worker. The estimated number of affected EAP workers in Year 1 is 4.682 
million (as discussed in section VII.D.iii.). Therefore, total Year 1 
adjustment costs were estimated to equal $160.1 million ($34.19 x 1 
hour x 4.682 million workers).
---------------------------------------------------------------------------

    \95\ Costs in the 2004 Final Rule were considered but because 
that revision included changes to the duties test the cost estimates 
are not directly applicable.
---------------------------------------------------------------------------

    Adjustment costs may be partially offset by a reduction in the cost 
to employers of determining employees' exempt status. Currently, to 
determine whether an employee is exempt firms must apply the duties 
test to salaried workers who earn at least $455 per week. Following 
this rulemaking, firms will no longer be required to apply the 
potentially time consuming duties test to employees earning less than 
the proposed salary level. This will be a clear cost savings to 
employers for employees who do not pass the duties test and earn at 
least $455 per week but less than the proposed salary level. The 
Department did not estimate the potential size of this cost savings.

[[Page 38567]]

4. Managerial Costs
    If employers reclassify employees as overtime eligible due to the 
changes in the salary levels, then firms may incur ongoing managerial 
costs associated with this proposed rulemaking because the employer may 
schedule and more closely monitor an employee's hours to minimize or 
avoid paying the overtime premium. These costs are in addition to the 
one-time regulatory familiarization and adjustment costs described 
above. For example, when scheduling hours the manager may have to 
assess whether the marginal benefit of scheduling the worker for more 
than 40 hours exceeds the marginal cost of paying the overtime premium. 
Additionally, the manager may have to spend more time monitoring the 
employee's work and productivity since the marginal cost of employing 
the worker per hour has increased.
    Because there was little precedent or data to aid in evaluating 
these costs, the Department examined several sources to estimate costs. 
First, prior part 541 rulemakings were reviewed to determine whether 
managerial costs were estimated. No estimates were found. This cost was 
not quantified for the 2004 rulemaking. Second, a literature review was 
conducted in an effort to identify information to help guide the cost 
estimates; again, no estimates were found. The Department requests data 
from the public applicable to this cost estimate. Despite a lack of 
available data, the Department chose to include estimated managerial 
costs to produce as full and accurate a cost estimate to employers as 
possible.
    To provide a sense of the potential magnitude of these costs, the 
Department estimated these costs assuming that management spends an 
additional five minutes per week scheduling and monitoring each 
affected worker expected to be reclassified as overtime eligible as a 
result of this NPRM, and whose hours are adjusted (1,022,000 affected 
EAP workers as calculated in section VII.D.vi.). As will be discussed 
in detail below, most affected workers do not currently work overtime, 
and there is no reason to expect their hours worked to change when 
their status changes from exempt to nonexempt. Similarly, employers are 
likely to find that it is less costly to give some workers a raise in 
order to maintain their exempt status. For both these groups of 
workers, management will have little or no need to increase their 
monitoring of hours worked. Under these assumptions, the additional 
managerial hours worked per week were estimated to be 85,200 hours 
rounded ((5 minutes/60 minutes) x 1,022,000 workers).
    The median hourly wage in 2013 for a manager was $27.78 and 
benefits were paid at a rate of 45 percent of the base wage, which 
totaled $40.20 per hour.96 97 Multiplying the additional 
85,200 weekly managerial hours by the hourly wage of $40.20 and 52 
weeks per year, the Year 1 costs were estimated to total $178.1 million 
for the proposed standard salary level. Although the exact magnitude 
would vary with the number of affected EAP workers each year, these 
costs would be incurred annually.
---------------------------------------------------------------------------

    \96\ Calculated as the median wage in the CPS for workers in 
management occupations (excluding chief executives) in 2013.
    \97\ The adjustment ratio is derived from the BLS's Employer 
Costs for Employee Compensation (ECEC) data using variables 
CMU1020000000000D and CMU1030000000000D.
---------------------------------------------------------------------------

5. Other Potential Costs
    In addition to the costs discussed above, there may be additional 
costs that have not been quantified. Other categories of unquantified 
costs are discussed in section VII.D.vii and immediately below.
Reduced Profits
    The increase in worker earnings' resulting from the revised salary 
level is a transfer of income from firms to workers, not a cost, and is 
thus neutral concerning its primary effect on welfare and gross 
domestic product (GDP). However, there are potential secondary effects 
(both costs and benefits) of the transfer due to the potential 
difference in the marginal utility of income and the marginal 
propensity to consume between workers and business owners. The transfer 
may result in societal gain during periods when the economy is 
operating below potential to the extent that transferring income to 
workers with a relatively high marginal propensity to consume results 
in a larger multiplier effect and impact on GDP. Conversely, this 
transfer may also reduce the profits available to firms for business 
investment.
Hiring Costs
    One of Congress' goals in enacting the FLSA in 1938 was to spread 
employment to a greater number of workers by effectively raising the 
wages of employees working more than 40 hours per week. To the extent 
that firms respond to an update to the salary level test by reducing 
overtime, they may do so by spreading hours to other workers, 
including: Current workers employed for less than 40 hours per week by 
that employer, current workers who retain their exempt status, and 
newly hired workers. If new workers are hired to absorb these 
transferred hours, then the associated hiring costs are a cost of this 
proposed rulemaking. The reduction in hours is considered in more 
detail in section VII.D.v.
v. Transfers
1. Overview
    Transfer payments occur when income is redistributed from one party 
to another. The Department has quantified two possible transfers likely 
to result from this proposed update to the salary level tests: (1) 
Transfers to employees from employers to ensure compliance with the 
FLSA minimum wage provision; and (2) transfers to employees from 
employers to ensure compliance with the FLSA overtime pay provision. 
Transfers in Year 1 to workers from employers due to the minimum wage 
provision were estimated to equal $46.7 million. The proposed increase 
in the HCE exemption compensation level does not affect minimum wage 
transfers because workers eligible for the HCE exemption earn well 
above the minimum wage. Transfers to employees from employers due to 
the overtime pay provision were estimated to be $1,435.8 million, 
$1,394.2 million of which is from the increased standard salary level, 
while the remainder is attributable to the increased HCE compensation 
level. Total Year 1 transfers were estimated to be $1,482.5 million 
(Table 18).

                                Table 18--Summary of Year 1 Regulatory Transfers
                                                   (Millions)
----------------------------------------------------------------------------------------------------------------
                                                                                        HCE
               Transfer from employers to workers                    Standard      Compensation        Total
                                                                   salary level        level
----------------------------------------------------------------------------------------------------------------
Due to minimum wage.............................................           $46.7            $0.0           $46.7

[[Page 38568]]

 
Due to overtime pay.............................................         1,394.2            41.7         1,435.8
                                                                 -----------------------------------------------
    Total transfers.............................................         1,440.8            41.7         1,482.5
----------------------------------------------------------------------------------------------------------------

    Because the overtime premium depends on the base wage, the 
estimates of minimum wage transfers and overtime transfers are linked. 
This can be considered a two-step approach. The Department first 
identified affected EAP workers with an implicit regular hourly wage 
lower than the minimum wage, and then calculated the wage increase 
necessary to reach the minimum wage. The implicit regular rate of pay 
is calculated as usual weekly earnings divided by usual weekly hours 
worked. For those employees whose implicit regular rate of pay is below 
the minimum wage, the overtime premium was based on the minimum wage as 
the regular rate of pay.
2. Transfers Due to the Minimum Wage Provision
    Transfers from employers to workers to ensure compliance with the 
federal minimum wage are small compared to the transfers attributed to 
overtime pay and are only associated with the change in the standard 
salary level (workers currently eligible for the HCE test earn well 
above the minimum wage). For purposes of this analysis, the hourly rate 
of pay is calculated as usual weekly earnings divided by usual weekly 
hours worked. In addition to earning low wages, this set of workers 
earns an hourly rate below the federal minimum wage and also works many 
hours per week. To demonstrate, in order to earn less than the federal 
minimum wage of $7.25 per hour, but at least $455 per week, these 
workers must regularly work significant amounts of overtime (since 
$455/$7.25 = 62.8 hours). The applicable minimum wage is the higher of 
the federal minimum wage and the state minimum wage. Most affected EAP 
workers already receive at least the minimum wage; an estimated 12,000 
affected EAP workers (less than 0.3 percent of all affected EAP 
workers) currently earn an implicit hourly rate of pay less than the 
minimum wage. The Department estimated transfers due to payment of the 
minimum wage by calculating the change in earnings if wages rose to the 
minimum wage for workers who become nonexempt and thus would have to be 
paid the minimum wage.\98\
---------------------------------------------------------------------------

    \98\ Because these workers' hourly wages will be set at the 
minimum wage after the proposed rule, their employers will not be 
able to adjust their wages downward to offset part of the cost of 
paying the overtime pay premium (which will be discussed in the 
following section). Therefore, these workers will generally receive 
larger transfers attributed to the overtime pay provision than other 
workers.
---------------------------------------------------------------------------

    In response to an increase in the regular rate of pay to the 
minimum wage, employers may reduce the workers' hours, which must be 
considered when estimating transfers attributed to payment of the 
minimum wage to newly overtime-eligible workers. In theory, because the 
quantity of labor hours demanded is inversely related to wages, a 
higher mandated wage could result in fewer hours of labor demanded. 
However, the weight of the empirical evidence finds that increases in 
the minimum wage have caused little or no significant job loss.\99\ 
Thus, in the case of this proposed regulation, the Department believes 
that any disemployment effect due to the effect of the minimum wage 
provision would be negligible. This is partially due to the small 
number of workers affected by this provision. The Department estimated 
the potential disemployment effects (i.e., the estimated reduction in 
hours) of the transfer attributed to the minimum wage by multiplying 
the percent change in the regular rate of pay by a labor demand 
elasticity of -0.075.\100\
---------------------------------------------------------------------------

    \99\ Belman, D., and P.J. Wolfson (2014). What Does the Minimum 
Wage Do? Kalamazoo, MI: W.E. Upjohn Institute for Employment 
Research. Dube, A., T.W. Lester, and M. Reich. (2010). Minimum Wage 
Effects Across State Borders: Estimates Using Contiguous Counties. 
IRLE Working Paper No. 157-07. https://irle.berkeley.edu/workingpapers/157-07.pdf. Schmitt, J. (2013). Why Does the Minimum 
Wage Have No Discernible Effect on Employment? Center for Economic 
and Policy Research.
    \100\ This is based on the estimated impact of a change in the 
minimum wage from $7.25 to $9.00 per hour on the employment of 
teenagers from Congressional Budget Office. (2014). The Effects of a 
Minimum-Wage Increase on Employment and Family Income. While an 
elasticity estimate for adult workers would be more appropriate, the 
report stated that the elasticity for adults was ``about one-third 
of the elasticity'' for teenagers, without providing a specific 
value. In addition, the literature for adults is more limited.
---------------------------------------------------------------------------

    At the proposed salary level ($921 per week), the Department 
estimated that 12,000 affected EAP workers will on average see an 
hourly wage increase of $0.98, work 1.0 fewer hour per week, and 
receive an increase in weekly earnings of $74.0 as a result of coverage 
by the minimum wage provisions (Table 19). Thus, the total change in 
weekly earnings due to the payment of the minimum wage was estimated to 
be $897,300 per week ($74.0 x 12,000) or $46.7 million in Year 1.

   Table 19--Minimum Wage Only: Mean Hourly Wages, Usual Overtime Hours, and Weekly Earnings for Affected EAP
                                                  Workers, 2013
----------------------------------------------------------------------------------------------------------------
                                                                                                   Total weekly
                                                    Hourly wage    Usual weekly    Usual weekly      transfer
                                                        \a\            hours         earnings      (1,000s) \b\
----------------------------------------------------------------------------------------------------------------
Before proposed regulation......................           $7.09            76.4          $529.1              --
After proposed regulation.......................            8.07            75.4           603.1              --
Change..........................................            0.98            -1.0            74.0          $897.3
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
 \a\The applicable minimum wage is the higher of the federal minimum wage and the state minimum wage.
 \b\Usual weekly earnings multiplied by the 12,000 exempt workers with an implicit regular rate of pay below the
  minimum wage who would lose their exemption status under the proposed rulemaking if weekly earnings did not
  change.


[[Page 38569]]

3. Transfers Due to the Overtime Pay Provision
    The proposed rule will also transfer income to affected EAP workers 
working in excess of 40 hours per week through payment of overtime to 
workers earning between the current and proposed salary levels. The 
size of the transfers will depend largely on how employers respond to 
the proposed salary level for affected EAP workers who work overtime. 
Employers may respond by: (1) Paying the required overtime premium to 
affected workers for the same number of overtime hours at the same 
implicit regular rate of pay; (2) reducing the regular rate of pay for 
workers working overtime; (3) eliminating overtime hours and 
potentially transferring some of these hours to other workers; (4) 
increasing workers' salary to the proposed salary level; or (5) using 
some combination of these responses. How employers will respond depends 
on the relative costs of each of these alternatives; in turn, the 
relative costs of each of these alternatives are a function of workers' 
earnings and hours worked.
    The simplest approach to estimating these transfer payments would 
be to multiply an employee's regular rate of pay (after compliance with 
the minimum wage) by 1.5 for all overtime hours; this is referred to as 
the ``full overtime premium'' model.\101\ However, due to expected wage 
and hour adjustments by employers, this would likely overestimate the 
size of the transfer. Therefore, the Department used a methodology that 
allows for employer adjustments, such as changes in the regular rate of 
pay or hours worked. The size of these adjustments is likely to vary 
depending on the affected worker's salary and work patterns.
---------------------------------------------------------------------------

    \101\ The implicit regular rate of pay is calculated as usual 
weekly earnings divided by usual weekly hours worked. For example, 
the regular rate of pay for an employee previously ineligible for 
overtime whose usual weekly earnings was $600 and usual weekly hours 
was 50 would be $12. Under the full overtime premium model, this 
employee would receive $660 (40 hours x $12) + (10 hours x $12 x 
1.5).
---------------------------------------------------------------------------

Employer Adjustments to the Regular Rate of Pay
    This section focuses on evaluating employers' responses to affected 
EAP workers who work regular overtime (usually work more than 40 hours 
in a week). The requirement that employers pay newly nonexempt 
employees in accordance with minimum wage and overtime requirements may 
result in changes in employment conditions; requiring an overtime 
premium increases the marginal cost of labor, which employers will 
likely try to offset by adjusting wages or hours. How employers respond 
to a new salary level and the ensuing changes in employment conditions 
will depend on the demand for labor, current wages, employer and 
employee bargaining power, and other factors. To model employer 
responses, the Department used a method that reflects the average 
response among all employers for all affected workers. However, 
individual employer responses will vary.
    Two conceptual models are useful for thinking about how employers 
may respond to reclassifying certain employees as overtime eligible: 
The ``full overtime premium'' model and the ``employment contract'' 
model.\102\ These models make different assumptions about the demand 
for overtime hours and the structure of the employment agreement which 
result in different implications for predicting employer responses.
---------------------------------------------------------------------------

    \102\ The employment contract model is also known as the fixed-
job model. See Trejo, S.J. (1991). The Effects of Overtime Pay 
Regulation on Worker Compensation. American Economic Review, 81(4), 
719-740 and Barkume, A. (2010). The Structure of Labor Costs with 
Overtime Work in U.S. Jobs. Industrial and Labor Relations Review, 
64(1), 128-142.
---------------------------------------------------------------------------

    The full overtime premium model is based on the traditional ``labor 
demand'' model of determining wage and hour conditions. In the labor 
demand model, employers and employees negotiate fixed hourly wages and 
then subsequently negotiate hours worked, rather than determining both 
hours and pay simultaneously. This model assumes employees are aware of 
the hourly wage rate they negotiated and may be more reluctant to 
accept downward adjustments. The labor demand model would apply if 
employees had a contract to be paid at an hourly rate, meaning that 
employers could not reduce the regular rate of pay in response to the 
requirement to pay a 50 percent premium on hours worked beyond 40 in a 
week. However, the increase in the cost of labor would lead to a 
reduction in the hours of labor demanded as long as labor demand is not 
completely inelastic. The full overtime premium model is a particular 
scenario of the labor demand model in which the demand for labor is 
completely inelastic, that is employers will demand the same number of 
hours worked regardless of the cost.
    In the employment contract model, employers and employees negotiate 
total pay and hours simultaneously, rather than negotiating a fixed 
hourly wage and then determining hours. Under this model, when 
employers are required to pay employees an overtime premium, they 
adjust the employees' implicit hourly rate of pay downward so that when 
the overtime premium is paid total employee earnings (and thus total 
employer cost) remain constant, along with the employees' hours. The 
employer does not experience a change in cost and the employee does not 
experience a change in earnings or hours. The employment contract model 
would hold if the workers who are reclassified as overtime protected 
had an employment agreement specifying set total earnings and hours of 
work.
    The employment contract model tends to be more applicable to 
salaried workers while the labor demand model is generally more 
applicable to workers paid hourly. Since all affected EAP workers in 
this analysis are salaried, the Department believes the employment 
contract model may be more appropriate for estimating employer response 
to the proposed salary increase. However, the employment contract model 
may not always hold true due to market constraints, employer 
incentives, or workers' bargaining power. Four examples are provided.
     Employers are constrained because they cannot reduce an 
employee's implicit hourly rate of pay below the minimum wage. If the 
employee's implicit hourly rate of pay before the change is at or below 
the minimum wage, then employers will not be able to reduce the rate of 
pay to offset the cost of paying the overtime premium.
     Employees generally have some, albeit limited, bargaining 
power which may prevent employers from reducing the employee's implicit 
hourly rate of pay to fully offset increased costs.
     Employers may be hesitant to reduce the employee's 
implicit hourly rate of pay by the entire amount predicted by the 
employment contract model because it may hurt employee morale and 
consequently productivity.
     Employers are often limited in their ability to pay 
different regular rates of pay to different employees who perform the 
same work and have the same qualifications. In order to keep wages 
constant across employees and reduce wages for overtime workers, 
employers would need to reduce the implicit hourly rate of pay for 
employees who do not work overtime as well as those who do work 
overtime. This would reduce total earnings for these non-overtime 
employees (potentially causing retention problems, productivity losses, 
and morale concerns).
    Therefore, the likely outcome will fall somewhere between the 
conditions predicted by the full overtime premium and employment 
contract models. For

[[Page 38570]]

example, the implicit hourly rate of pay may fall, but not all the way 
to the wage predicted by the employment contract model, and overtime 
hours may fall but not be eliminated since the implicit hourly rate of 
pay has fallen. The Department conducted a literature review to 
evaluate how the market would adjust to a change in the requirement to 
pay overtime.
    Barkume (2010) and Trejo (1991) empirically tested for evidence of 
these two competing models by measuring labor market responses to the 
application of FLSA overtime pay regulations.\103\ Both concluded that 
wages partially adjust toward the level consistent with the employment 
contract model in response to the overtime pay provision.\104\ Barkume 
found that employee wage rates were adjusted downward by 40 to 80 
percent of the amount the employment contract model predicted, 
depending on modeling assumptions. Earlier research had demonstrated 
that in the absence of regulation some employers may voluntarily pay 
workers some overtime premium to entice them to work longer hours, to 
compensate workers for unexpected changes in their schedules, or as a 
result of collective bargaining.\105\ Thus Barkume assumed that workers 
would receive an average voluntary overtime pay premium of 28 percent 
in the absence of an overtime pay regulation. Including this voluntary 
overtime pay from employers, he estimated that in response to overtime 
pay regulation, the wage adjusted downward by 80 percent of the amount 
that would occur with the employment contract model. Conversely, when 
Barkume assumed workers would receive no voluntary overtime pay premium 
in the absence of an overtime pay regulation, wages adjusted downward 
40 percent of the amount the employment contract model 
predicted.106 107 However, while it seemed reasonable that 
some premium was paid for overtime in the absence of regulation, 
Barkume's assumption of a 28 percent initial overtime premium is likely 
too high for the salaried workers potentially affected by a change in 
the salary and compensation level requirements for the EAP 
exemptions.\108\
---------------------------------------------------------------------------

    \103\ Barkume, A. (2010). The Structure of Labor Costs with 
Overtime Work in U.S. Jobs. Industrial and Labor Relations Review, 
64(1), 128-142. Trejo, S.J. (1991). The Effects of Overtime Pay 
Regulation on Worker Compensation. American Economic Review, 81(4), 
719-740.
    \104\ Since both papers were based on cross-sectional data, 
findings were assumed to be at the final equilibrium wages. Studies 
showing wage contracts are likely to be stickier in the short run 
than in the long run have limited applicability here since this 
analysis deals exclusively with salaried workers who are less likely 
to be aware of their implicit hourly wage rate. The Department has 
modeled a sticky adjustment process by assuming the wage elasticity 
of demand for labor is smaller in Year 1 than in subsequent years.
    \105\ Barzel, Y. (1973). The Determination of Daily Hours and 
Wages. The Quarterly Journal of Economics, 87(2), 220-238 
demonstrated that modest fluctuations in labor demand could justify 
substantial overtime premiums in the employment contract model. 
Hart, R.A. and Yue, M. (2000). Why Do Firms Pay an Overtime Premium? 
IZA Discussion Paper No. 163, showed that establishing an overtime 
premium in an employment contract can reduce inefficiencies.
    \106\ Barkume's estimates are consistent with Trejo's 1991 
finding that the wage adjustment when there is no overtime premium 
was only about 40 percent of the full employment contract model 
adjustment. Trejo's estimates range from 25 percent to 49 percent 
and average 40 percent. Trejo, S.J. (1991). The Effects of Overtime 
Pay Regulation on Worker Compensation. American Economic Review, 
81(4), 719-740.
    \107\ Consider a worker earning $500 and working 50 hours per 
week. Assuming no overtime premium is paid the imputed hourly rate 
of pay is $10. Assuming a 28 percent overtime premium, the hourly 
rate of pay is $9.47 (($9.47 x 40) + (($9.47 x 1.28) x 10)) = $500. 
If the hourly rate of pay was fully adjusted to the employment 
contract model level when overtime pay is newly required, the hourly 
rate of pay would be $9.09 (($9.09 x 40) + (($9.09 x 1.5) x 10)) = 
$500. Forty percent of the adjustment from $10 to $9.09 results in 
an adjusted regular rate of pay of $9.64. Eighty percent of the 
adjustment from $9.47 to $9.09 results in an adjusted hourly rate of 
pay of $9.17. The Department took the average of these two adjusted 
wages to estimate that the resulting hourly rate of pay would be 
$9.40.
    \108\ Barkume (2010) based this assumption on the findings of 
Bell, D. and Hart, R. (2003). Wages, Hours, and Overtime Premia: 
Evidence from the British Labor Market. Industrial and Labor 
Relations Review, 56(3), 470-480. This study used 1998 data on male, 
non-managerial full-time workers in Britain. British workers were 
likely paid a larger voluntary overtime premium than American 
workers because Britain did not have a required overtime pay 
regulation and so collective bargaining played a larger role in 
implementing overtime pay.
---------------------------------------------------------------------------

Modeling Employer Adjustments to the Hourly Rate of Pay and Overtime 
Hours
    In practice, employers do not seem to adjust wages of regular 
overtime workers to the full extent indicated by the employment 
contract model, and thus employees appear to get a small but 
significant increase in weekly earnings due to coverage by overtime pay 
regulations. Barkume and Trejo found evidence partially supporting both 
the employment contract model and the full overtime premium model in 
response to a 50 percent overtime premium requirement: A decrease in 
the regular rate of pay for workers with overtime (but not the full 
decrease to the employment contract model level) and a decrease in the 
probability of working overtime. Therefore, when modeling employer 
responses with respect to the adjustment to the regular rate of pay, 
the Department used a method that falls somewhere between the 
employment contract model and the full overtime premium model (i.e., 
the partial employment contract model).
    Barkume reported two methods to estimate this partial employment 
contract wage, depending on the amount of overtime pay assumed to be 
paid in the absence of regulation. As noted above, the Department 
believes both the model assuming a voluntary 28 percent overtime 
premium and the model assuming no voluntary overtime premium are 
unrealistic for the affected population. Therefore, lacking more 
information, the Department determined that an appropriate estimate of 
the impact on the implicit hourly rate of pay for regular overtime 
workers after the proposed rule should be determined using the average 
of Barkume's two estimates of partial employment contract model 
adjustments: A wage change that is 40 percent of the wage change 
assuming an initial zero overtime pay premium, and a wage change that 
is 80 percent of the wage change assuming an initial 28 percent 
overtime pay premium.\109\ This is approximately equivalent to assuming 
that overtime workers received a 14 percent overtime premium in the 
absence of regulation (the mid-point between 0 and 28 percent).
---------------------------------------------------------------------------

    \109\ Both studies considered a population that included hourly 
workers. Evidence is not available on how the adjustment towards the 
employment contract model differs between salaried and hourly 
workers. The employment contract model may be more likely to hold 
for salaried workers than for hourly workers since salaried workers 
directly observe their weekly total earnings, not their implicit 
equivalent hourly wage. Thus, applying the partial adjustment to the 
employment contract model as estimated by these studies may 
overestimate the transfers from employers to workers who are 
salaried.
---------------------------------------------------------------------------

    How employers adjust workers' wages and hours depends on employment 
conditions. The discussion begins with a description of how employment 
conditions affect employers' wage adjustments depending on the 
differing work characteristics of their employees. However, changing 
employees' earnings is also likely to result in adjustments to hours 
worked. Thus, after estimating wage adjustments the Department 
calculated the adjustments to hours worked as a function of the new 
wage. Finally, transfers from employers to employees were estimated as 
a function of the changes in wages and the changes in hours.
    The Department identified four types of workers whose work 
characteristics impact how employers were modeled to respond to the 
proposed changes in both the standard and HCE salary levels:

[[Page 38571]]

     Type 1: Workers who do not work overtime. These workers 
will not experience any adjustment in their hourly rate of pay.
     Type 2: Workers who do not regularly work overtime but 
occasionally work overtime.\110\ Some of these workers' implicit hourly 
rate of pay will fall.\111\ Others will have no change in their hourly 
rate of pay.
---------------------------------------------------------------------------

    \110\ Type 2 workers are those who worked overtime in the survey 
week (the week referred to in the CPS MORG questionnaire). If a 
different week was chosen as the survey week then likely some of 
these workers would not have worked overtime. However, because the 
data are representative of both the population and all twelve months 
in a year, the Department believes the share of Type 2 workers in 
the given week is representative of an average week in the year.
    \111\ The Department assumes that Type 2 workers are currently 
paid additional wages for overtime hours worked at the usual hourly 
wage rate. Specifically, Type 2 workers' actual earnings for the 
week are calculated as (usual weekly earnings/usual hours worked) x 
(actual hours worked last week).
---------------------------------------------------------------------------

     Type 3: Workers who regularly work overtime. These 
workers' implicit hourly rate of pay falls to reflect the partial 
employment contract model adjustment.\112\
---------------------------------------------------------------------------

    \112\ The reduction in the regular hourly wage is restricted by 
the minimum wage; the wage cannot fall below the minimum wage.
---------------------------------------------------------------------------

     Type 4: Workers who regularly work overtime. These workers 
differ from the Type 3 workers in that once wages and hours are 
adjusted, weekly earnings are greater than the proposed salary level, 
so employers will increase these workers' earnings to the proposed 
salary level so they can continue to claim the EAP exemption for 
them.\113\
---------------------------------------------------------------------------

    \113\ It is possible that employers will increase the salaries 
paid to some ``occasional'' overtime workers to maintain the 
exemption for the worker, but the Department has no way of 
identifying these workers.
---------------------------------------------------------------------------

    Type 1 affected EAP workers will become overtime eligible, but 
since they do not work overtime, they will see no change in their 
weekly earnings. Type 2 and Type 3 affected EAP workers will become 
overtime eligible and must be paid the overtime premium for any 
overtime hours worked and may see changes in their regular rate of pay, 
and/or hours, and thus weekly earnings. As explained in more detail 
below, Type 2 and Type 3 affected workers were modeled differently due 
to the difference in the nature of the overtime hours worked. Type 3 
workers receive wages adjusted for partial compliance with the 
employment contract model and their hours adjust in response. Type 4 
workers are those who regularly work overtime, but will remain exempt 
because their weekly earnings will be raised to the proposed EAP salary 
level (either the standard salary level or HCE compensation level 
depending on which test the worker passed). How employers respond to 
workers who work overtime hours is described in more detail in the 
following paragraphs for Type 2 and Type 3 workers.
    The Department distinguishes those who regularly work overtime 
(Type 3 workers) from those who occasionally, or irregularly, work 
overtime (Type 2 workers) because employer adjustment to the proposed 
rule may differ accordingly. The Department believes that employers are 
more likely to adjust hours worked and wages for regular overtime 
workers because their hours are predictable. Conversely, it may be more 
difficult to adjust hours and wages for occasional overtime workers 
because employers may be responding to a transient, perhaps 
unpredicted, shift in market demand for the good or service they 
provide. In this case it is likely advantageous for the employer to pay 
for this occasional overtime rather than to adjust permanent staffing. 
Additionally, the transient and possibly unpredicted nature of the 
change may make it difficult to adjust wages for these workers.
    The Department treats Type 2 affected workers in two ways due to 
the uncertainty of the nature of these occasional overtime hours 
worked. If these workers work extra hours on an unforeseen, short-term, 
as-needed basis (e.g., to adjust to unanticipated increases in demand), 
then there may be less opportunity for employers to adjust straight-
time wages downward.\114\ However, if these workers work extra hours on 
a foreseen, periodic basis (e.g., work a few extra hours one week each 
month, but workers do not consider it ``regular overtime'' because they 
do not work overtime during three weeks each month), then there may be 
some opportunity for employers to adjust straight-time wages downward 
(e.g., so pre- and post-revision monthly income is more similar). That 
this overtime is periodic and predictable is what makes it much more 
similar to that worked by Type 3 workers, and provides employers with 
more opportunity to adjust hours and wages. Since in reality there is 
likely a mix of these two occasional overtime scenarios, the Department 
combines models representing these two scenarios when estimating 
impacts.\115\
---------------------------------------------------------------------------

    \114\ Employers may be reluctant to reset hourly wage rates to 
respond to unexpected changes to the need for overtime because the 
negative impact on worker morale may outweigh the gains from 
adjusting wages to unexpected shifts in demand. Of relevance is the 
well-established literature that shows employers do not quickly 
adjust wages downward in regard to downturns in the economy; the 
same logic applies to our approach to unexpected changes in demand. 
See, for example: Bewley, T. (1999). Why Wages Don't Fall During a 
Recession. Cambridge, MA: Harvard University Press. See also Barzel, 
Y. (1973). The Determination of Daily Hours and Wages. The Quarterly 
Journal of Economics, 87(2), 220-238.
    \115\ Trejo and Barkume's adjustments are averages; excluding 
some workers (i.e., half of Type 2 workers) from these adjustments 
could potentially bias the size of the adjustment for the workers 
who continue to receive the adjustment. This bias would exist if 
Barkume and Trejo estimated the average adjustment for a sample of 
workers including irregular overtime workers and the size of the 
adjustment for these workers differs from other workers. It is not 
clear whether Trejo's and Barkume's samples include both occasional 
and regular overtime workers; however, the Department's 
interpretation is that Trejo includes only workers who usually work 
overtime and Barkume includes both. If these assumptions are 
correct, the magnitude of this RIA's adjustment made for the workers 
whose wages and hours are adjusted would be appropriate if it were 
applying Trejo's results but may, due to applying Barkume's, result 
in an underestimate of the average fall in base wages. We believe 
the magnitude of any potential bias will be small because the half 
of Type 2 workers who are occasional overtime workers (and thus 
treated differently) compose only 8 percent of Type 2 and Type 3 
workers.
---------------------------------------------------------------------------

    Our estimate for how Type 2 workers are affected is based on the 
assumption that 50 percent of these workers who worked occasional 
overtime worked expected overtime hours and the other 50 percent worked 
unexpected overtime. Workers were randomly assigned to these two 
groups. Workers with expected occasional overtime hours were treated 
like Type 3 affected workers (partial employment contract model 
adjustments). Workers with unexpected occasional overtime hours were 
assumed to receive a 50 percent pay premium for the overtime hours 
worked (full overtime premium model).
    Since affected Type 2 and Type 3 EAP workers work more than 40 
hours per week, whether routinely or occasionally, they will now be 
overtime protected. These workers will receive an overtime premium 
based on their implicit hourly wage adjusted as described above. 
Because employers must now pay more for the same number of labor hours, 
they will seek to reduce those hours; in economics, this is described 
as a decrease in the quantity of labor hours demanded (a movement to 
the left along the labor demand curve). It is the net effect of these 
two changes that will determine the final weekly earnings for affected 
EAP workers. Next we describe how these workers' hours adjust in 
response to the change in their implicit hourly wage and the 
requirement to pay an overtime premium on that wage for each hour 
worked in excess of 40 hours per week.
    The reduction in hours is calculated using the elasticity of labor 
demand with respect to wages. The Department used a short-run demand 
elasticity of

[[Page 38572]]

-0.20 to estimate the percentage decrease in hours worked resulting 
from the increase in average hourly wages in Year 1 calculated using 
the adjusted base wage and the overtime wage premium.\116\ The 
interpretation of the short run demand elasticity in this context is 
that a 10 percent increase in wages will result in a 2 percent decrease 
in hours worked. Transfers projected for years 2 through 10 used a 
long-run elasticity; this will be discussed in section VII.D.x.1.\117\
---------------------------------------------------------------------------

    \116\ This elasticity estimate is based on the Department's 
analysis of Lichter, A., Peichl, A. & Siegloch, A. (2014). The Own-
Wage Elasticity of Labor Demand: A Meta-Regression Analysis. IZA DP 
No. 7958. Some researchers have estimated larger impacts from own 
wage changes on the number of overtime hours worked (Hamermesh, D. 
and S. Trejo. (2000). The Demand for Hours of Labor: Direct Evidence 
from California. The Review of Economics and Statistics, 82(1), 38-
47 conclude the price elasticity of demand for overtime hours is at 
least -0.5). The Department decided to use a general measure of 
elasticity applied to the average change in wages since the increase 
in the overtime wage is somewhat offset by a decrease in the non-
overtime wage as indicated in the employment contract model, and 
welcomes comments on the appropriate elasticity to be used in this 
analysis.
    \117\ In the short-run not all factors of production can be 
changed and so the change in hours demanded is smaller than in the 
long run, when all factors are flexible.
---------------------------------------------------------------------------

    The Department calculates the percent increase in hourly wages 
since it must be used with the elasticity of labor demand to determine 
the change in hours. This is equal to workers' new average hourly wage 
(including overtime pay) divided by their original implicit hourly 
wage. For Type 3 affected workers, and the 50 percent of Type 2 
affected workers who worked expected overtime, we estimate adjusted 
total hours worked after making wage adjustments using the partial 
employment contract model. To estimate adjusted hours worked, we set 
the percent change in total hours worked equal to the percent change in 
average wages multiplied by the wage elasticity of labor demand.\118\ 
The wage elasticity of labor demand was determined from a review of 
published econometric studies. The percent change in average wages is 
equal to the adjusted implicit average hourly wage minus the original 
implicit average hourly wage divided by the original implicit average 
hourly wage. The original implicit average hourly wage is equal to 
original weekly earnings divided by original hours worked. The adjusted 
implicit average hourly wage is equal to adjusted weekly earnings 
divided by adjusted total hours worked. Adjusted weekly earnings equals 
the adjusted hourly wage (i.e., after the partial employment contract 
model adjustment) multiplied by 40 hours plus adjusted hours worked in 
excess of 40 multiplied by 1.5 times the adjusted hourly wage.
---------------------------------------------------------------------------

    \118\ In this equation, the only unknown is adjusted total hours 
worked. Since adjusted total hours worked is in the denominator of 
the left side of the equation and is also in the numerator of the 
right side of the equation, solving for adjusted total hours worked 
requires solving a quadratic equation.
---------------------------------------------------------------------------

    Figure 4 is a flow chart summarizing the four types of affected EAP 
workers. Also shown are the impacts on exempt status, weekly earnings, 
and hours worked for each type of affected worker.

[[Page 38573]]

[GRAPHIC] [TIFF OMITTED] TP06JY15.003

    \a\ Affected EAP workers are those who are exempt under the 
current EAP exemptions and would gain minimum wage and overtime 
protection or receive a raise to the proposed increased salary 
level.
    \b\ Depending on how employers respond to this rule, some 
workers may experience adverse consequences due to a reduction in 
their hours of work, potentially necessitating a second job to 
maintain their pre-rule earnings level.
    \c\ Occasional overtime workers are those who responded that 
they (1) do not usually work overtime and (2) worked overtime in the 
survey week. In any given week different workers may be working 
occasional overtime but the Department assumes the total number of 
occasional overtime workers and occasional overtime hours are 
similar across weeks.
    \d\ The amount wages are adjusted downwards depends on whether 
the employment contract model or the labor demand model holds. The 
Department's preferred method uses a combination of the two. 
Employers reduce the regular hourly wage rate somewhat in response 
to overtime pay requirements, but the wage is not reduced enough to 
keep total compensation constant.
    \e\ Based on hourly wage and weekly hours it is more cost 
efficient for the employer to increase the worker's weekly salary to 
the updated salary level than to pay overtime pay.
    \f\ The Department assumed hours would not change due to lack of 
data and relevant literature; however, it is possible employers will 
increase these workers' hours in response to paying them a higher 
salary.
Estimates of the Number of and Impacts on Affected EAP workers
    The Department projects 4.7 million workers will be affected by 
either (1) an increase in the standard salary level to the 40th 
earnings percentile because they earn salaries between $455 per week 
and $921 per week or (2) an increase in the HCE compensation level to 
the 90th earnings percentile, $122,148 annually. These workers are 
categorized into the four ``types'' identified previously. There are 
3.5

[[Page 38574]]

million Type 1 workers (74.7 percent of all affected EAP workers), 
those who work 40 hours per week or less and thus will not be paid an 
overtime premium despite their expected change in status to overtime 
protected (Table 20). Type 2 workers, those who are expected to become 
overtime eligible and do not usually work overtime but did work 
overtime in the survey week (i.e., occasional overtime workers), total 
181,000 workers (3.9 percent of all affected EAP workers). Type 3 
workers, those who are expected to become overtime eligible and be paid 
the overtime premium, are composed of an estimated 931,000 workers 
(19.9 percent of all affected EAP workers). The number of affected Type 
4 workers was estimated to be 71,000 workers (1.5 percent of all 
affected workers); these are workers who the Department believes will 
remain exempt because firms will have a financial incentive to increase 
their weekly salaries to the proposed salary level so that they remain 
exempt, rather than pay a premium for overtime hours.\119\
---------------------------------------------------------------------------

    \119\ As previously described, the Department calculated a wage 
and hour adjustment for all regular overtime workers. Consider, by 
way of example, a worker who initially earned $900 and worked 70 
hours per week. Suppose the partial employment contract adjustment 
results in a regular rate of pay of $11.94 and 69.5 hours worked per 
week. After the partial employment contract adjustments, this worker 
would receive approximately $1,006 per week ((40 x $11.94) + (29.5 x 
($11.94 x 1.5)). Since this is greater than the proposed standard 
salary level, the Department estimated that this worker would have 
his salary increased to $921 and remain exempt at that threshold.

                                                  Table 20--Affected EAP Workers by Type (1,000s), 2013
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                   Regular OT
                                                                                        No overtime     Occasional OT  ---------------------------------
                                                                        Total \a\       worked (T1)          (T2)            Newly        Remain exempt
                                                                                                                         nonexempt (T3)        (T4)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Standard salary level..............................................          4,646            3,478              180              920               67
HCE compensation level.............................................             36.2             20.7              1.0             11.1              3.4
                                                                    ------------------------------------------------------------------------------------
    Total..........................................................          4,682            3,499              181              931               71
--------------------------------------------------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Estimated number of workers exempt under the EAP exemptions who would be entitled to overtime protection under the proposed salary levels (if their
  weekly earnings do not increase to the proposed salary levels).
*Type 1: Workers without regular OT and without occasional OT.
*Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly earnings increase, but regular rate of pay and
  hours fall for 50 percent of workers.
*Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours increase, but regular rate of pay and hours
  fall.
*Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level).

    The proposed rulemaking will likely impact affected workers' wages, 
hours, and earnings. How these will change depends on the type of 
worker. Predicted changes in implicit wage rates are outlined in Table 
21; changes in hours in Table 22; and changes in weekly earnings in 
Table 23. Type 1 workers will have no change in wages, hours, or 
earnings.\120\
---------------------------------------------------------------------------

    \120\ It is possible that these workers may experience an 
increase in hours and weekly earnings because of transfers of hours 
from overtime workers. Due to the high level of uncertainty in 
employers' responses regarding the transfer of hours, the Department 
did not have credible evidence to support an estimation of the 
number of hours transferred to other workers.
---------------------------------------------------------------------------

    Estimating changes in the regular rate of pay for Type 3 workers 
and the 50 percent of Type 2 workers who regularly work occasional 
overtime requires application of the partial employment contract model, 
which predicts a decrease in their average regular rates of pay. The 
Department estimates that employers would decrease these workers' 
regular hourly rates of pay to the amount predicted by the partial 
employment contract model adjustment. Employers would not be able to 
adjust the regular rate of pay for the occasional overtime workers 
whose overtime is irregularly scheduled and unpredictable (the 
remaining 50 percent of Type 2 workers). As a group, Type 2 workers 
currently exempt under the standard test would see a decrease in their 
average regular hourly wage (i.e., excluding the overtime premium) from 
$18.30 to $17.88, a decrease of 2.3 percent. Type 2 workers paid 
between $100,000 and the proposed HCE compensation level would see an 
average decrease in their regular hourly wage from $52.99 to $50.85, a 
decrease of 4.0 percent. However, because workers will now receive a 50 
percent premium on their regular hourly wage for each hour worked in 
excess of 40 hours per week, average weekly earnings for Type 2 workers 
would increase.
    Type 3 workers will also receive decreases in their regular hourly 
wage as predicted by the partial employment contract model. Type 3 
affected workers paid below the proposed standard salary level would 
have their regular hourly rate of pay decrease on average from $14.71 
to $13.93 per hour, a decrease of 5.3 percent. Type 3 workers paid 
between $100,000 and the proposed HCE compensation level would have 
their regular rate of pay decrease on average from $39.23 to $36.66 per 
hour, a decrease of 6.5 percent. Again, although regular hourly rates 
decline, weekly earnings will increase on average because these workers 
are now eligible for the overtime premium.
    Type 4 workers' implicit hourly rates of pay would increase in 
order for their earnings to meet the proposed standard salary level 
($921 per week) or the proposed HCE annual compensation level (122,148 
annually). The implicit hourly rate for Type 4 affected EAP workers who 
had earned between $455 and $921 per week would increase on average 
from $16.40 to $16.72 (a 2.0 percent increase) (Table 21). The implicit 
hourly rate of pay for Type 4 workers who had earned between $100,000 
and $122,148 annually would increase on average from $41.87 to $42.32 
(a 1.1 percent increase).

[[Page 38575]]



                   Table 21--Average Regular Rate of Pay by Type of Affected EAP Worker, 2013
----------------------------------------------------------------------------------------------------------------
                                                                                            Regular OT
                                                    No overtime    Occasional OT -------------------------------
                                       Total        worked (T1)        (T2)            Newly      Remain  exempt
                                                                                  nonexempt (T3)       (T4)
----------------------------------------------------------------------------------------------------------------
                                              Standard Salary Level
----------------------------------------------------------------------------------------------------------------
Before proposed rule............          $18.38          $19.39          $18.30          $14.71          $16.40
After proposed rule.............           18.21           19.39           17.88           13.93           16.72
Change..........................           -0.17            0.00           -0.42           -0.78            0.33
----------------------------------------------------------------------------------------------------------------
                                             HCE Compensation Level
----------------------------------------------------------------------------------------------------------------
Before proposed rule............          $47.26          $52.18          $52.99          $39.23          $41.87
After proposed rule.............           46.46           52.18           50.85           36.66           42.32
Change..........................           -0.80            0.00           -2.14           -2.57            0.45
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
*Type 1: Workers without regular OT and without occasional OT.
*Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly
  earnings increase, but regular rate of pay and hours fall for 50 percent of workers.
*Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours
  increase, but regular rate of pay and hours fall.
*Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level).

    Type 1 and Type 4 workers would have no change in hours. Type 1 
workers' hours would not change because they do not work overtime and 
thus the requirement to pay an overtime premium does not affect them. 
Type 4 workers' hours would not change because they continue to be 
exempt, and therefore are not paid a premium for overtime hours. Type 2 
and Type 3 workers would see a small decrease in their hours of 
overtime worked. This reduction in hours is relatively small and is due 
to the effect on labor demand of the increase in the average hourly 
base wage as predicted by the employment contract model.\121\
---------------------------------------------------------------------------

    \121\ The Department estimates that half of Type 2 workers will 
not see a reduction in their hours; however as a group, Type 2 
workers are expected to experience a reduction in their hours of 
work.
---------------------------------------------------------------------------

    Type 2 workers who would be newly overtime eligible would see a 
decrease in average weekly hours in weeks where occasional overtime is 
worked, from 48.0 to 47.9 hours (0.3 percent) (Table 22).\122\ Type 2 
workers who would no longer earn the HCE compensation level would see a 
decrease in average weekly hours in applicable weeks from 46.5 to 46.3 
(0.4 percent).
---------------------------------------------------------------------------

    \122\ Type 2 workers do not see increases in regular earnings to 
the new salary level (as Type 4 workers do) even if their new 
earnings exceed that new level. This is because the estimated new 
earnings only reflect their earnings in that week; their earnings 
for the entire year do not necessarily exceed the salary level.
---------------------------------------------------------------------------

    Type 3 workers affected by the increase in the standard salary 
level would see a decrease in hours worked from 50.7 to 50.3 hours per 
week (0.8 percent). Type 3 workers affected by the increase in the HCE 
compensation level would see an average decrease from 53.7 to 53.3 
hours per week (0.8 percent).

                      Table 22--Average Weekly Hours for Affected EAP Workers by Type, 2013
----------------------------------------------------------------------------------------------------------------
                                                                                            Regular OT
                                                    No overtime    Occasional OT -------------------------------
                                       Total        worked (T1)        (T2)            Newly      Remain  exempt
                                                                                  nonexempt (T3)       (T4)
----------------------------------------------------------------------------------------------------------------
                                            Standard Salary Level \a\
----------------------------------------------------------------------------------------------------------------
Before proposed rule............            41.6            38.6            48.0            50.7            56.9
After proposed rule.............            41.5            38.6            47.9            50.3            56.9
Change..........................            -0.1             0.0            -0.1            -0.4             0.0
----------------------------------------------------------------------------------------------------------------
                                           HCE Compensation Level \a\
----------------------------------------------------------------------------------------------------------------
Before proposed rule............            45.8            39.8            46.5            53.7            56.4
After proposed rule.............            45.7            39.8            46.3            53.3            56.4
Change..........................            -0.1             0.0            -0.2            -0.4             0.0
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Usual hours for Types 1, 3, and 4 but actual hours for Type 2.
*Type 1: Workers without regular OT and without occasional OT.
*Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly
  earnings increase, but regular rate of pay and hours fall for 50 percent of workers.
*Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours
  increase, but regular rate of pay and hours fall.
*Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level).


[[Page 38576]]

    Because Type 1 workers do not experience a change in their regular 
rate of pay or hours they would have no change in earnings due to the 
proposed rule (Table 23). While their hours are not expected to change, 
Type 4 workers' salaries would increase to the proposed standard salary 
level or HCE compensation level (depending on which test they pass). 
Thus, Type 4 workers' average weekly earnings would increase by $20.47 
(2.3 percent) for those affected by the change in the standard salary 
level and by $27.36 per week (1.2 percent) for those affected by the 
HCE compensation level.
    Although both Type 2 and Type 3 workers on average experience a 
decrease in both their regular rate of pay and hours worked, their 
weekly earnings are expected to increase as a result of the overtime 
premium. Based on a standard salary level of $921 per week, Type 2 
workers' average weekly earnings increase from $879.35 to $925.33, a 
5.2 percent increase.\123\ The average weekly earnings of Type 2 
workers affected by the change in the HCE compensation level were 
estimated to increase from $2,470.77 to $2,514.22, a 1.8 percent 
increase.
---------------------------------------------------------------------------

    \123\ For these calculations, the Department assumed Type 2 
workers are paid their regular rate of pay for all occasional 
overtime hours. For example, if a Type 2 worker earned $700 per week 
and normally worked a 40 hour workweek then his or her regular rate 
of pay would be $17.50 per hour. If that person worked 10 hours of 
overtime in some week, he or she would earn $875 ($700 + $17.50 x 
10) in that week. This is why baseline average weekly earnings are 
higher than for other types of workers. These workers do not see 
increases in regular earnings to the new salary level since their 
earnings only exceed the salary level in some weeks. If instead, the 
Department assumed Type 2 workers received no additional pay for 
occasional overtime hours, but merely received their usual weekly 
salary, then estimated baseline earnings would be smaller, and 
estimated transfers would be larger for these workers.
---------------------------------------------------------------------------

    Average weekly earnings of Type 3 workers also increase. For Type 3 
workers affected by the standard salary level, average weekly earnings 
would increase from $731.54 to $751.13, an increase of 2.7 percent. 
Type 3 workers affected by the change in the HCE compensation level 
have an increase in average weekly earnings from $2,057.41 to 
$2,117.56, an increase of 2.9 percent.

                    Table 23--Average Weekly Earnings for Affected EAP Workers by Type, 2013
----------------------------------------------------------------------------------------------------------------
                                                                                            Regular OT
                                                                                 -------------------------------
                                       Total        No overtime    Occasional OT       Newly
                                                   worked  (T1)        (T2)          Nonexempt    Remain  exempt
                                                                                       (T3)             (T4)
----------------------------------------------------------------------------------------------------------------
                                           Standard Salary Level \a b\
----------------------------------------------------------------------------------------------------------------
Before proposed rule............         $730.58         $719.31         $879.35         $731.54         $900.53
After proposed rule.............          736.54          719.31          925.33          751.13          921.00
Change..........................            5.96            0.00           45.97           19.60           20.47
----------------------------------------------------------------------------------------------------------------
                                          HCE Compensation Level \a b\
----------------------------------------------------------------------------------------------------------------
Before proposed rule............        2,103.26        2,075.18        2,470.77        2,057.41        2,321.64
After proposed rule.............        2,125.42        2,075.18        2,514.22        2,117.56        2,349.00
Change..........................           22.16            0.00           43.45           60.15           27.36
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ The mean of the hourly wage multiplied by the mean of the hours does not necessarily equal the mean of the
  weekly earnings because the product of two averages is not necessarily equal to the average of the product.
\b\ Weekly earnings for weeks where overtime is worked. Thus for Type 3 and 4 workers weekly earnings is derived
  by multiplying the wage by usual hours worked but for Type 2 workers weekly earnings is derived by multiplying
  the wage by actual hours worked in the survey week.
* Type 1: Workers without regular OT and without occasional OT.
* Type 2: Workers without regular OT but with occasional OT. Paid overtime premium pay, so average weekly
  earnings increase, but regular rate of pay and hours fall for 50 percent of workers.
* Type 3: Workers with regular OT who become nonexempt. Paid overtime premium pay, so average weekly hours
  increase, but regular rate of pay and hours fall.
* Type 4: Workers with regular OT who remain exempt (i.e., are paid the proposed salary level).

    Weekly earnings after an increase to the proposed standard salary 
level were estimated using the new wage (i.e., the partial employment 
contract model wage) and the reduced number of overtime hours worked. 
At the proposed standard salary level, the average weekly earnings of 
all affected workers will increase from $730.58 to $736.54, a change of 
$5.96 (0.8 percent). However, these figures mask the impact on workers 
whose hours and earnings will change because Type 1 workers make up 
more than 70 percent of the pool of affected workers. If Type 1 
workers, who do not work overtime, are excluded the average increase in 
weekly earnings is $23.72.
    At the proposed standard salary level, multiplying the average 
change of $5.96 by the 4.6 million affected standard EAP workers equals 
an increase in earnings of $27.7 million per week or $1,441 million in 
the first year (Table 24). Of the weekly total, $897,000 is due to the 
minimum wage provision and $26.8 million stems from the overtime pay 
provision. For workers affected by the change in the HCE compensation 
level, average weekly earnings increase by $22.16 ($51.91 if Type 1 
workers, who do not work overtime, are excluded). When multiplied by 
36,000 affected workers, the national increase in weekly earnings will 
be $801,000 per week, or $41.7 million in the first year. Thus, Year 1 
transfer payments attributable to this proposed rule total $1,482.5 
million. If the Department assumed Type 2 workers received no 
additional pay for occasional overtime hours prior to the rulemaking 
(as discussed above), then Year 1 transfers would instead be $1,499.1 
million.

[[Page 38577]]



  Table 24--Total Change in Weekly and Annual Earnings for Affected EAP
                       Workers by Provision, 2013
------------------------------------------------------------------------
                                             Total change in earnings
                                                     (1,000s)
                Provision                -------------------------------
                                              Weekly          Annual
------------------------------------------------------------------------
Total \a\...............................         $28,509      $1,482,490
Standard salary level Total.............          27,708       1,440,825
    Minimum wage only...................             897          46,662
    Overtime pay only \b\...............          26,811       1,394,163
HCE compensation level Total............             801          41,665
    Minimum wage only...................  ..............  ..............
    Overtime pay only \b\...............             801          41,665
------------------------------------------------------------------------
\a\ Due to both the minimum wage and overtime pay provisions and
  proposed changes in both the standard salary level and the HCE
  compensation level.
\b\ Estimated by subtracting the minimum wage transfer from the total
  transfer.

4. Potential Transfers Not Quantified
    There may be additional transfers attributable to this proposed 
rulemaking; however, the magnitude of these other transfers could not 
be quantified. These transfers are discussed in this section, as well 
as in section VII.D.vii, below.
Converted to Hourly Status From Salaried Status
    Changing the EAP salary and HCE compensation level tests may impact 
whether a worker is classified as overtime ineligible or overtime 
eligible. Some evidence suggests that it is more costly for an employer 
to employ a salaried worker than an hourly worker. If true, employers 
may choose to accompany the change in exemption status with a change to 
the employee's method of pay, from salary to an hourly basis, since 
there is no longer an incentive to classify the worker as 
salaried.\124\ Several employer stakeholders noted that salaried 
workers may perceive such a change as a loss of status.
---------------------------------------------------------------------------

    \124\ There is no requirement that overtime eligible employees 
be paid on an hourly basis. Paying such employees on a salary basis 
is appropriate so long as the employee receives overtime pay for 
working more than 40 hours in the workweek. See 29 CFR 778.113.
---------------------------------------------------------------------------

    If the worker prefers to be salaried rather than hourly, then this 
change may impact the worker. The likelihood of this impact occurring 
depends on the costs to employers and benefits to employees of being 
salaried. Research has shown that salaried workers (who are not 
synonymous with exempt workers, but whose status is correlated with 
exempt status) are more likely than hourly workers to receive benefits 
such as paid vacation time and health insurance \125\ and are more 
satisfied with their benefits,\126\ and that when employer demand for 
labor decreases hourly workers tend to see their hours cut before 
salaried workers, making earnings for hourly workers less 
predictable.\127\ However, this literature generally does not control 
for differences between salaried and hourly workers such as education, 
job title, or earnings; therefore, this correlation is not necessarily 
attributable to hourly status.
---------------------------------------------------------------------------

    \125\ Lambert, S. J. (2007). Making a Difference for Hourly 
Employees. In A. Booth, & A. C. Crouter, Work-Life Policies that 
Make a Real Difference for Individuals, Families, and Communities. 
Washington, DC: Urban Institute Press.
    \126\ Balkin, D. B., & Griffeth, R. W. (1993). The Determinants 
of Employee Benefits Satisfaction. Journal of Business and 
Psychology, 7(3), 323-339.
    \127\ Lambert, S. J., & Henly, J. R. (2009). Scheduling in 
Hourly Jobs: Promising Practices for the Twenty-First Century 
Economy. The Mobility Agenda. Lambert, S. J. (2007). Making a 
Difference for Hourly Employees.
---------------------------------------------------------------------------

    Additionally, even if a worker's salaried status is not officially 
changed, a salaried worker may effectively become an hourly worker if 
managers have to monitor hours more closely, so the worker may have 
less flexibility in work schedule.\128\
---------------------------------------------------------------------------

    \128\ Swanberg, J. E., Pitt-Catsouphes, M., & Drescher-Burke, K. 
(2005). A Question of Justice: Disparities in Employees' Access to 
Flexible Schedule Arrangements. Journal of Family Issues, 26 (6), 
866-895. WorldatWork Research. (2009). Flexible Work Arrangements 
for Nonexempt Employees. WorldatWork Research.
---------------------------------------------------------------------------

Reduced earnings for some workers
    Holding regular rate of pay and work hours constant, payment of an 
overtime premium will increase weekly earnings for workers who work 
overtime. However, as discussed previously, employers may try to 
mitigate cost increases by reducing the number of overtime hours 
worked, either by transferring these hours to other workers or 
monitoring hours more closely. Depending on how hours are adjusted, a 
specific worker may earn less pay after this proposed rulemaking. For 
example, assume an exempt worker is paid for overtime hours at his 
regular rate of pay (not paid the overtime premium but still acquires a 
benefit from each additional hour worked over 40 in a week). If the 
employer does not raise the worker's salary to the new level, requiring 
the overtime premium may cause the employer to reduce the worker's 
hours to 40 per week. If the worker's regular rate of pay does not 
increase, the worker will earn less due to the lost hours of work.
vi. Deadweight Loss
    Deadweight loss (DWL) occurs when a market operates at less than 
optimal equilibrium output. This typically results from an intervention 
that sets, in the case of a labor market, wages above their equilibrium 
level. While the higher wage results in transfers from employers to 
workers, it also causes a decrease in the total number of labor hours 
that are being purchased on the market. DWL is a function of the 
difference between the wage the employers were willing to pay for the 
hours lost and the wage workers were willing to take for those hours. 
In other words, DWL represents the total loss in economic surplus 
resulting from a ``wedge'' between the employer's willingness to pay 
and the worker's willingness to accept work arising from the proposed 
change. DWL may vary in magnitude depending on market parameters, but 
is typically small when wage changes are small or when labor supply and 
labor demand are relatively price (wage) inelastic.
    The DWL resulting from this proposed rulemaking was estimated based 
on the average decrease in hours worked and increase in hourly wages 
calculated in section VII.D.v. As the cost of labor rises due to the 
requirement to pay the overtime premium, the demand for overtime hours 
decreases, which results in fewer hours of overtime worked. To 
calculate the DWL, the following values must be estimated:
     The increase in average hourly wages for affected EAP 
workers,
     the decrease in average hours per worker, and

[[Page 38578]]

     the number of affected EAP workers.
    Only 50 percent of Type 2 workers (those who work regular or 
predictable occasional overtime) and Type 3 workers are included in the 
DWL calculation because the other workers either do not work overtime 
(Type 1), continue to work the same number of overtime hours (Type 4), 
or their employers are unable to adjust their hourly wage because their 
overtime hours worked are unpredictable (the other 50 percent of Type 2 
workers). As described above, after taking into account a variety of 
potential responses by employers, the Department estimated the average 
wage change for EAP affected workers whose hours change. Workers 
impacted by the change in the standard salary level are considered 
separately from workers impacted by the change in the HCE compensation 
level.
    For workers affected by the revised standard salary level, and who 
experience a change in hours, average wages (including overtime) will 
increase by $0.68 per hour. Average hours will fall by 0.40 per week. 
These changes result in an average DWL of $0.14 per week per Type 2 
(the 50 percent who work foreseeable overtime) and Type 3 worker. An 
estimated 1.01 million workers will be eligible for the overtime 
premium on some of their hours worked after employer adjustments are 
taken into account. Multiplying the $0.14 per worker estimate by the 
number of affected workers results in a total DWL of $7.2 million in 
the first year of this proposed rulemaking attributable to the revised 
standard salary level (1.01 million workers in DWL analysis x $0.14 per 
worker per week x 52 weeks).
    For workers affected by the revised HCE compensation level and who 
experience a change in hours, the average hourly wage will increase by 
$2.14 and average hours worked will fall by 0.41 per week. This results 
in an average DWL of $0.44 per week for each of the estimated 12,000 
workers affected by the compensation level who will see their hours 
fall. Multiplying this per worker estimate by the number of affected 
workers results in a DWL of $273,000 in the first year attributable to 
the HCE component of this proposed rulemaking (12,000 workers in DWL 
analysis x $0.44 per worker x 52 weeks). Thus, total DWL attributed to 
the proposed rulemaking is estimated to be $7.4 million in Year 1, 
which is small in comparison to the size of the costs and transfers 
associated with this proposal.

          Table 25--Summary of Deadweight Loss Component Values
------------------------------------------------------------------------
                                                                HCE
                Component                    Standard      compensation
                                           salary level        level
------------------------------------------------------------------------
Average hourly wages:
    Pre.................................          $15.01          $40.31
    Post................................          $15.70          $42.45
    Change..............................           $0.68           $2.14
Average overtime hours:
    Pre.................................           10.45           13.14
    Post................................           10.05           12.73
    Change..............................           -0.40           -0.41
Affected EAP workers....................    1,010,433          12,042
DWL:
    DWL per worker per week.............           $0.14           $0.44
    Total annual DWL (millions).........           $7.15           $0.27
------------------------------------------------------------------------
Note: DWL analysis is limited to Type 2 (50%) and Type 3 workers who
  experience hour adjustments.

vii. Other Benefits, Costs and Transfers
1. Benefits, Costs and Transfers Due to Strengthening Overtime 
Protection for Other Workers
    In addition to the 4.7 million affected EAP workers who will be 
newly eligible for overtime protection (absent employer response to 
increase the salary level to retain the exemption), overtime protection 
will be strengthened for an additional 10.0 million salaried workers 
who earn between the current salary level of $455 per week and the 
proposed salary level of $921 per week. These workers, who were 
previously vulnerable to misclassification through misapplication of 
the duties test, will now be automatically overtime protected because 
their salary falls below the new salary level and therefore they will 
not be subject to the duties test. These 10.0 million workers include:
     6.3 million salaried white collar workers who are at 
particular risk of being misclassified because they currently pass the 
salary level test but do not satisfy the duties test; and
     3.7 million salaried workers in blue collar occupations 
whose overtime protection will be strengthened because their salary 
will fall below the proposed salary threshold.\129\ (Identification of 
blue collar workers is explained in section VII.B.iv).
---------------------------------------------------------------------------

    \129\ Some workers in this group may be overtime ineligible due 
to another non-EAP exemption.
---------------------------------------------------------------------------

    Although these workers are currently entitled to minimum wage and 
overtime protection, their protection is better assured with the 
proposed salary level. The salary level test is considered a bright-
line test because it is clear to employers and employees alike whether 
or not a worker passes. The duties test (which is the reason employers 
cannot claim the EAP exemption for the above workers) is more 
discretionary and therefore harder to apply. An outdated salary level 
reduces the effectiveness of this bright-line test. At the proposed 
salary level, the number of overtime-eligible white collar workers 
earning at or above the salary level will decrease by 6 million, and an 
estimated 806,562 (13.5 percent) of these workers are currently 
misclassified as exempt. Therefore, increasing the salary level is 
expected to result in less worker misclassification. Employers will be 
able to more readily determine their legal obligations and comply with 
the law, thus leading to benefits, costs and transfers that are 
qualitatively similar to the impacts discussed elsewhere in this 
analysis but the magnitudes of which have not been estimated.
2. Cost Savings: Reduction in Litigation
    Reducing the number of white collar employees for whom a duties 
analysis must be performed in order to

[[Page 38579]]

determine entitlement to overtime will also reduce litigation related 
to the EAP exemption. As previously discussed, employer uncertainty 
about which workers should be classified as EAP exempt has contributed 
to a sharp increase in FLSA lawsuits over the past decade. Much of this 
litigation has involved whether employees who satisfy the salary level 
test also meet the duties test for exemption. See, e.g, Soehnle v. Hess 
Corp., 399 F. App'x 749 (3d Cir. 2010) (gas station manager earning 
approximately $654 per week satisfied duties test for executive 
employee); Morgan v. Family Dollar Stores, Inc., 551 F.3d 1233 (11th 
Cir. 2008) (store managers earning an average weekly salary of up to 
$706 did not satisfy duties test for executive exemption).
    Setting an appropriate salary level for the standard duties test 
and maintaining the salary level with automatic updates will restore 
the test's effectiveness as a bright-line method for determining exempt 
status, and in turn decrease the litigation risk created when employers 
must apply the duties test to employees who generally are not 
performing bona fide EAP work. This will eliminate legal challenges 
regarding the duties test involving employees earning between the 
current salary level ($455) and the proposed level ($921). See, e.g., 
Little v. Belle Tire Distribs., Inc., 588 F. App'x 424 (6th Cir. 2014) 
(applicability of administrative or executive exemption to tire store 
assistant manager earning $1,100 semi-monthly); Taylor v. Autozone, 
Inc., 572 F. App'x 515 (9th Cir. 2014) (applicability of executive 
exemption to store managers earning as little as $800 per week); Diaz 
v. Team Oney, Inc., 291 F. App'x. 947 (11th Cir. 2008) (applicability 
of executive duties test to pizza restaurant assistant manager earning 
$525 per week). Setting the salary level test at the proposed level 
will alleviate the need for employers to apply the duties test in these 
types of cases, which is expected to result in decreased litigation as 
employers will be able to determine employee exemption status through 
application of the salary level test without the need to perform a 
duties analysis. See Weiss Report at 8 (The salary tests ``have amply 
proved their effectiveness in preventing the misclassification by 
employers of obviously nonexempt employees, thus tending to reduce 
litigation. They have simplified enforcement by providing a ready 
method of screening out the obviously nonexempt employees, making an 
analysis of duties in such cases unnecessary.'')
3. Benefits and Costs: Reduction in Uncertainty about Future Overtime 
Hours and Pay
    The proposed rule may have an impact on employees who are not 
currently working any overtime, but will now be entitled to minimum 
wage and overtime pay protections. These workers may face a lower risk 
of being asked to work overtime in the future, because they are now 
entitled to an overtime premium, which could reduce their uncertainty 
and improve their welfare if they do not desire to work overtime. 
Additionally, if they are asked to work overtime, they are compensated 
for the inconvenience with an overtime premium.
    Economic theory suggests that workers tend to assign monetary 
values to risk or undesirable job characteristics, as evidenced by the 
presence of compensating wage differentials for undesirable jobs, 
relative to other jobs the worker can perform in the marketplace. To 
the extent a compensating wage differential exists, compensation may 
decrease with the reduction in uncertainty.\130\ For this reason, 
overall compensation would be expected to decrease for workers whose 
uncertainty decreases. Employees who prefer the reduced uncertainty to 
the wage premium would experience a net benefit of the rule, and 
employees who prefer the wage premium to the reduced uncertainty would 
experience a net harm as a result of the rule. The Department believes 
that attempting to model the net monetary value of reduced uncertainty 
is not feasible due to its heavy reliance on data that are not readily 
available, and the potentially questionable nature of the resulting 
estimates.\131\
---------------------------------------------------------------------------

    \130\ In this case, the size of the compensating wage 
differential is a function of the likelihood of working overtime and 
the amount of overtime worked. If the probability of working 
overtime is small then the wage differential may not exist.
    \131\ For a discussion of compensating wage differentials, see 
Gronberg, T. J., & Reed, W. R. (1994). Estimating Workers' Marginal 
Willingness to Pay for Job Attributes using Duration Data. Journal 
of Human Resources, 29(3), 911-931.
---------------------------------------------------------------------------

4. Benefits and Costs: Work-Life Balance
    Due to the increase in marginal cost for overtime hours, employers 
will demand fewer hours from some of the workers affected by this 
rule.\132\ The estimated transfer payment does not take into account 
the benefit to these workers of working fewer hours in exchange for 
more (or equal) pay. Therefore, an additional benefit of this proposed 
rulemaking is the increase in time off for affected EAP workers. On 
average, affected EAP workers were estimated to work 5.2 minutes less 
per week after the proposed rulemaking. The effect is much more 
pronounced when limited to just those workers whose hours are adjusted 
(50 percent of Type 2 and all Type 3 workers); they would on average 
work 23.9 minutes less per week after the proposed rulemaking. The 
additional time off may help these workers better balance work-life 
commitments, thus potentially making them better off.
---------------------------------------------------------------------------

    \132\ The Department recognizes that not all workers would 
prefer to work fewer hours and thus some of these workers might 
experience an adverse impact. The Department has no basis for 
estimating this potential impact.
---------------------------------------------------------------------------

    Empirical evidence shows that workers in the United States 
typically work more than workers in other comparatively wealthy 
countries.\133\ Although estimates of the actual level of overwork vary 
considerably, executive, administrative, and professional occupations 
have the highest percentage of workers who would prefer to work fewer 
hours compared to other occupational categories.\134\ Therefore, the 
Department believes that the proposed rule may result in increased time 
off for a group of workers who may prefer such an outcome. However, the 
empirical evidence does not allow us to estimate how many workers would 
prefer fewer hours or how much workers value this additional time off 
so it is difficult to monetize the benefit they may receive. However, 
if we use affected workers' average wage to approximate the value they 
place on this time ($15.31), then the benefit of this additional time 
off would total $6.2 million weekly (0.40 hours x 1.02 million workers 
x $15.31). This would result in an estimated total benefit of $324.4 
million per year.
---------------------------------------------------------------------------

    \133\ For more information, see OECD series, average annual 
hours actually worked per worker, available at: https://stats.oecd.org/index.aspx?DataSetCode=ANHRS.
    \134\ Hamermesh, D.S., Kawaguchi, D., Lee, J. (2014). Does Labor 
Legislation Benefit Workers?
    Well-Being after an Hours Reduction. IZA DP No. 8077.
    Golden, L., & Gebreselassie, T. (2007). Overemployment 
mismatches: the preference for fewer work hours. Monthly Labor 
Review, 130(4), 18-37.
    Hamermesh, D.S. (2014). ``Not enough time?'' American Economist, 
59(2).
---------------------------------------------------------------------------

    This is likely an overestimate to the extent that not all workers 
would prefer to work fewer hours and thus some of these workers might 
experience an adverse impact. In addition, the estimated work loss 
represents an average over all affected workers, and some workers may 
experience a larger reduction in hours.\135\
---------------------------------------------------------------------------

    \135\ It is possible that some employers may choose to eliminate 
all overtime for affected workers and hire additional workers or 
spread the work to existing employees to replace the lost hours. The 
potential for this adjustment is uncertain, and the Department has 
found no studies that estimate the potential magnitude of this 
effect. In addition, an employer may be limited in his or her 
ability to make such adjustments; many affected employees work only 
a few hours of overtime each week; affected employees' tasks may not 
be easily divisible; and hiring new workers and/or managing 
different work flows will impose additional costs on the employer 
that will offset the savings from avoiding paying the overtime 
premium.

---------------------------------------------------------------------------

[[Page 38580]]

5. Additional Benefits and Costs not Quantified
    The largest benefit to workers from the proposed rule is the 
transfer of income from employers (as discussed in the transfer section 
of the analysis); but, to the extent that the benefits to workers 
outweigh the costs to employers, there may be a societal welfare 
increase due to this transfer. The channels through which societal 
welfare may increase and other secondary benefits may occur are 
discussed below and include increased productivity and improved worker 
outcomes, such as improved health. The discussion references the 
potential magnitude of these benefits where possible; however, due to 
data limitations and mixed evidence on the significance of such 
effects, the Department was not able to quantify the size of these 
potential benefits.

Health

    Working long hours is correlated with an increased risk of injury 
or health problems.\136\ Therefore, by reducing overtime hours, some 
affected EAP workers' health may improve. This would benefit the 
worker's welfare, their family's welfare, and society since fewer 
resources would need to be spent on health. Health has also been shown 
to be highly correlated with productivity.\137\ These beneficial 
effects, and how they compare with other potential responses by 
employers, especially regarding workers who pass the duties test and 
whose salaries are either already above the proposed threshold or would 
be adjusted to be so, have not been quantified.
---------------------------------------------------------------------------

    \136\ Keller, S. M. (2009). Effects of Extended Work Shifts and 
Shift Work on Patient Safety, Productivity, and Employee Health. 
AAOHN Journal, 57(12), 497-502.
    \137\ Loeppke, R., Taitel, M., Richling, D., Parry, T., Kessler, 
R., Hymel, P., et al. (2007). Health and Productivity as a Business 
Strategy. Journal of Occupational and Environmental Medicine, 49(7), 
712-721.
---------------------------------------------------------------------------

Increased productivity

    This proposed rule is expected to increase the marginal cost of 
some workers' labor, predominately due to the overtime pay requirement 
since almost all affected EAP workers already earn the federal minimum 
wage. However, some of the cost to employers of paying the overtime 
premium may be offset by increased worker productivity. This may occur 
through a variety of channels, including: increased marginal 
productivity as fewer hours are worked, reduction in turnover, 
efficiency wages, and worker health.
    Reduction in turnover: Research demonstrates a positive correlation 
between earnings and employee turnover: as earnings increase, employee 
turnover decreases.\138\ \139\ Reducing turnover may increase 
productivity, at least partially because new employees have less firm-
specific capital (i.e., skills and knowledge that have productive value 
in only one particular company) and thus are less productive and 
require additional supervision and training.\140\ In short, replacing 
experienced workers with new workers decreases productivity, and 
avoiding that will increase productivity. Reduced turnover should also 
reduce firms' hiring and training costs. As a result, even though 
marginal labor costs rise, they may rise by less than the amount of the 
wage change because the higher wages may be offset by lower turnover 
rates, increased productivity, and reduced hiring costs for firms.
---------------------------------------------------------------------------

    \138\ Howes, Candace, (2005). Living Wages and Retention of 
Homecare Workers in San Francisco. Industrial Relations, 44(1), 139-
163. Dube, A., Lester, T.W., & Reich, M. (2014). Minimum Wage 
Shocks, Employment Flows and Labor Market Frictions. IRLE Working 
Paper #149-13.
    \139\ Note that this literature tends to focus on changes in 
earnings for a specific sector or subset of the labor force. The 
impact on turnover when earnings increase across sectors (as would 
be the case with this regulation) may be smaller.
    \140\ Argote, L., Insko, C. A., Yovetich, N., & Romero, A. A. 
(1995). Group Learning Curves: The Effects of Turnover and Task 
Complexity on Group Performance. Journal of Applied Social 
Psychology, 25(6), 512-529.
    Shaw, J. D. (2011). Turnover Rates and Organizational 
Performance: Review, Critique, and Research Agenda. Organizational 
Psychology Review, 1(3), 187-213.
---------------------------------------------------------------------------

    It is difficult to estimate the impact of reduced turnover on 
worker productivity and firm hiring costs. The potential reduction in 
turnover is a function of several variables: the current wage, hours 
worked, turnover rate, industry, and occupation. Additionally, 
estimates of the cost of replacing a worker who quits vary 
significantly. Therefore, quantifying the potential benefit associated 
with a decrease in turnover attributed to this proposed rule is 
difficult.
    Efficiency wages: By increasing earnings this proposed rulemaking 
may increase a worker's productivity by incentivizing the worker to 
work harder. Thus the additional cost to firms may be partially offset 
by higher productivity. In particular, the estimated managerial costs 
associated with greater monitoring effort may be offset due to this 
effect. A strand of economic research, commonly referred to as 
``efficiency wages,'' considers how an increase in wages may be met 
with greater productivity.\141\ However, this literature tends to focus 
on firms voluntarily paying higher wages, and thus distinguishing 
themselves from other firms. Since this rulemaking mandates wage 
increases, extrapolating from efficiency wage theory may not provide a 
reliable guide to the likely effects of the rule.
---------------------------------------------------------------------------

    \141\ Akerlof, G. A. (1982). Labor Contracts as Partial Gift 
Exchange. The Quarterly Journal of Economics, 97(4), 543-569.
---------------------------------------------------------------------------

    Conversely, there are channels through which mandating overtime pay 
may reduce productivity. For example, some overtime hours may be spread 
to other workers. If the work requires significant project-specific 
knowledge or skills, then the new worker receiving these transferred 
hours may be less productive than the first worker, especially if there 
is a steep learning curve. Additionally, having an additional worker 
versed in the project may be beneficial to the firm if the first worker 
leaves the firm or is temporarily away (e.g., sick) or by providing 
benefits of teamwork (e.g., facilitating information exchange).
6. Transfers: Reduction in Social Assistance Expenditures
    The transfer of income resulting from this proposed rule may result 
in reduced need for social assistance (and by extension reduced social 
assistance expenditures by the government). A worker earning the 
current salary level of $455 per week earns $23,660 annually. If this 
worker resides in a family of four and is the sole earner, then the 
family will be considered impoverished. This makes the family eligible 
for many social assistance programs. Thus, transferring income to these 
workers may reduce eligibility for government social assistance 
programs and government expenditures. A social welfare improvement will 
result from the reduced resource needs for making those transfer 
payments.
    Benefits for which currently EAP exempt workers may qualify include 
Medicaid, the Supplemental Nutrition Assistance Program (SNAP), the 
Temporary Assistance for Needy Families (TANF) program, the Special 
Supplemental Nutrition Program for Women, Infants, and Children (WIC), 
and school breakfasts and lunches. Quantifying the impact of this 
proposed rulemaking on government expenditures

[[Page 38581]]

is complex and thus not estimated here. In order to conduct such an 
analysis, the Department would need estimates of the transfer per 
worker, his or her current income level, other sources of family 
income, number of family members, state of residence, and receipt of 
aid.
viii. Bounds on Transfer Payments
    Because the Department cannot predict the precise reaction of 
employers to the proposed rule, the Department also calculated bounds 
to the size of the estimated transfers from employers to workers using 
a variety of assumptions. Since transfer payments are the largest 
component of this proposed rulemaking the scenarios considered here are 
bounds around the transfer estimate. Based on the assumptions made, 
these bounds do not generate bounded estimates for costs or DWL.
    The maximum potential upper limit occurs with the assumption that 
the demand for labor is completely inelastic, and therefore neither the 
implicit regular hourly rate of pay nor hours worked adjust in response 
to the changes in the EAP standard salary level and HCE annual 
compensation level test. Employers then pay workers one and a half 
times their current implicit hourly rate of pay for all overtime hours 
currently worked (i.e., the full overtime premium). The minimum 
potential lower bound occurs when wages adjust completely and weekly 
earnings are unchanged as predicted by the employment contract model. 
The Department believes that both the maximum upper bound scenario and 
the minimum lower bound scenario are unrealistic; therefore, we 
constructed more credible bounds.
    For a more realistic upper bound on transfer payments, the 
Department assumed that all occasional overtime workers and half of 
regular overtime workers would receive the full overtime premium, as it 
was computed in the maximum upper bound methodology (i.e., such workers 
would work the same number of hours but be paid 1.5 times their 
implicit initial hourly wage for all overtime hours). Conversely, in 
the preferred model we assumed that only 50 percent of occasional 
overtime workers and no regular overtime workers would receive the full 
overtime premium. It was assumed that employers could not 
instantaneously adjust earnings for the 50 percent of affected EAP 
workers who regularly work overtime. However, for the other half of 
regular overtime workers, the Department assumed they would have their 
implicit hourly wage adjusted as predicted by the partial employment 
contract model (wage rates fall and hours are reduced but total 
earnings continue to increase, as in the preferred method). Table 26 
summarizes the assumptions described above.
    The plausible lower transfer bound also depends on whether 
employees work regular overtime or occasional overtime. For those who 
regularly work overtime hours and half of those who work occasional 
overtime, the Department assumes the employees' wages will fully adjust 
as predicted by the employment contract model, whenever possible (in 
the preferred method their wages adjust based on the partial employment 
contract model).\142\ For the other half of employees with occasional 
overtime hours, the lower bound assumes they will be paid one and one-
half times their implicit hourly wage for overtime hours worked (full 
overtime premium).
---------------------------------------------------------------------------

    \142\ The straight-time wage adjusts to a level that keeps 
weekly earnings constant when overtime hours are paid at 1.5 times 
the straight-time wage. In cases where adjusting the straight-time 
results in a wage less than the minimum wage, the straight-time wage 
is set to the minimum wage.

    Table 26--Summary of the Assumptions Used To Calculate the Lower
      Estimate, Preferred Estimate, and Upper Estimate of Transfers
------------------------------------------------------------------------
                                                        Upper transfer
     Lower transfer estimate      Preferred estimate       estimate
------------------------------------------------------------------------
                  Occasional Overtime Workers (Type 2)
------------------------------------------------------------------------
50% full employment contract      50% partial         100% full overtime
 model adj.                        employment          premium.
                                   contract model
                                   adj.
------------------------------------------------------------------------
50% full overtime premium.......  50% full overtime
                                   premium.
------------------------------------------------------------------------
                    Regular Overtime Workers (Type 3)
------------------------------------------------------------------------
100% full employment contract     100% partial        50% partial
 model adj.                        employment          employment
                                   contract model      contract model
                                   adj.                adj 50% full
                                                       overtime premium.
------------------------------------------------------------------------
Legend:
* Full overtime premium: Regular rate of pay equals the implicit hourly
  wage prior to the proposed regulation (with no adjustments); workers
  are paid 1.5 times this base wage for the same number of overtime
  hours worked prior to the regulation (assuming the worker was paid the
  minimum wage, otherwise the wage increases to the minimum wage and
  overtime hours may decrease).
* Full employment contract model adjustment: Base wages are set at the
  higher of: (1) a rate such that total earnings and hours remain the
  same before and after the proposed regulation; thus the base wage
  falls, and workers are paid 1.5 times the new base wage for overtime
  hours (the employment contract model) or (2) the minimum wage.
* Partial employment contract model adjustment: Regular rates of pay are
  partially adjusted to the wage implied by the employment contract
  model. The resulting regular rate of pay is the midpoint of: (1) a
  base wage that adjusts 40 percent of the way to the employment
  contract model wage level, assuming no overtime premium was initially
  paid and (2) a base wage that adjusts 80 percent of the way to the
  employment contract model wage level, assuming the workers initially
  received a 28 percent premium for overtime hours worked.

    The cost and transfer payment estimates associated with the bounds 
are presented in Table 27. Regulatory familiarization costs and 
adjustment costs do not vary across the scenarios. These employer costs 
are a function of the number of affected firms or affected workers, 
human resource personnel hourly wages, and time estimates. None of 
these vary based on the assumptions made above. Conversely, managerial 
costs are lower under these alternative employer response assumptions 
because fewer workers' hours are adjusted by employers and thus 
managerial costs, which depend on the number of workers whose hours 
change, will be smaller. Managerial costs vary according to employers' 
response to the proposed rule.
    Depending on how employers adjust the implicit regular hourly wage, 
the estimated transfer may range from $543.7 million to $2,851.2 
million, with the preferred estimate equal to $1,482.5 million. The DWL 
associated with the preferred estimate is $7.4 million. The

[[Page 38582]]

upper transfer estimate of DWL is smaller than the preferred estimate 
because the assumptions made for this upper bound scenario result in 
fewer hours lost. For the lower transfer estimate DWL was estimated to 
be less than $400,000; for the upper transfer estimate scenario the DWL 
was estimated to be $3.7 million.

                 Table 27--Bounds on Annual Cost and Transfer Payment Estimates, 2013 (Millions)
----------------------------------------------------------------------------------------------------------------
                                                                  Lower transfer     Preferred    Upper transfer
                          Cost/transfer                            estimate \a\      estimate        estimate
----------------------------------------------------------------------------------------------------------------
Direct employer costs...........................................
    Reg. familiarization........................................          $254.5          $254.5          $254.5
    Adjustment costs............................................           160.1           160.1           160.1
    Managerial costs............................................             1.7           178.1            82.8
Total direct employer costs.....................................           416.3           592.7           497.4
Transfers \a\...................................................           543.7         1,482.5         2,851.2
DWL.............................................................             0.4             7.4             3.7
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Due to both the minimum wage and overtime pay provisions and changes in both the standard salary level and
  the HCE compensation level.

ix. Regulatory Alternatives
    The Department proposes in this NPRM to update the standard salary 
level to the 40th percentile of weekly earnings for all full-time 
salaried workers ($921 per week). The Department considered a range of 
alternatives before deciding on this level. Seven alternatives are 
presented here. Two of these (alternatives 1 and 7) inflate the value 
of earlier salary levels to take into account inflation in the 
intervening years. Three others (alternatives 2, 3, and 5) adapt the 
2004 method or the Kantor method to set the salary level. Alternatives 
4 and 6 set the salary level to the median weekly salary for either all 
full-time hourly and salaried workers or full-time salaried workers, 
respectively. Table 28 presents the alternative salary levels 
considered and the number of workers estimated to be affected under 
these salary levels.
    Alternative 1 increases the 2004 salary level of $455 per week by 
the rate of inflation between 2004 and 2013 as measured by the CPI-U. 
This results in a salary level of $561 per week. At this salary level 
576,000 workers would be affected in Year 1, imposing direct adjustment 
and managerial costs of $36.1 million, transferring $127.9 million in 
earnings from employers to employees, and resulting in DWL of $0.5 
million.
    Alternative 2 sets the salary level using the 2004 method resulting 
in a salary level of $577 per week. At this salary level 734,000 
workers would be affected, Year 1 adjustment and managerial costs would 
equal $44.5 million, with transfers of $151.5 million, while DWL would 
equal $0.7 million.
    Alternative 3 sets the salary level using the Kantor method. This 
results in a salary level of $657 per week. At this salary level, 1.4 
million workers are affected, Year 1 adjustment and managerial costs 
are $91.8 million, Year 1 transfers are $318.6 million, and Year 1 DWL 
is $1.7 million.
    Alternative 4 sets the salary level equal to the 50th percentile, 
or median, of weekly earnings for full-time hourly and salaried 
workers. This results in a salary level of $776 per week. At this 
salary level, 2.7 million workers would be affected in Year 1, employer 
costs would total $179.8 million with transfers of $686.6 million, and 
DWL would be $3.6 million.
    Alternative 5 is based on the Kantor method but, whereas 
alternative 3 generates the salary level associated with the long 
duties test, alternative 5 generates a level more appropriate to the 
short duties test (as explained in section VII.C) and results in a 
salary level of $979 per week. At this salary level, 5.6 million 
workers would be affected in Year 1, with adjustment and managerial 
costs of $404.2 million, transfers of $1.8 billion, and DWL equal to 
$10.3 million. As previously noted, while this alternative uses the 
average difference between the Kantor long and short tests, the ratio 
of the short to long salary tests ranged between approximately 130 
percent and 180, which would result in a salary between $854 and 
$1,183.
    Alternative 6 sets the standard salary equal to the 50th 
percentile, or median, of weekly earnings for full-time salaried 
workers. This approach is similar to the proposed method but uses a 
higher weekly earnings percentile: 50th instead of the 40th. This 
results in a salary level of $1,065 per week. At this salary level, 6.9 
million workers would be affected in Year 1, employer costs would total 
$522.1 million with transfers of $2.5 billion, and DWL would be $14.8 
million.
    Alternative 7 increases the 1975 short test salary level of $250 
per week by the rate of inflation from 1975 to 2013. This results in a 
salary level of $1,083 per week. At this salary level, 7.1 million 
workers would be affected in Year 1, employer costs would total $543.0 
million with transfers of $2.7 billion, and DWL would be $15.5 million.
    The Department also examined alternatives to the proposed HCE 
compensation level. HCE alternative 1 left the current $100,000 annual 
compensation level unchanged. Therefore, no employer costs, transfers, 
or DWL are associated with this alternative.
    HCE alternative 2 sets the HCE annual compensation level at 
$150,000 per year. This compensation level would affect 52,000 workers 
in Year 1 (compared to 36,000 at the proposed compensation level), 
impose adjustment and managerial costs on employers of $5.5 million, 
transfer $71.2 million in earnings from employers to employees, and 
generate $600,000 in DWL. Because regulatory familiarization costs 
cannot realistically be differentiated into those relevant to the 
standard salary level, and those relevant to the HCE compensation 
level, the Department does not separately estimate those costs for the 
HCE alternatives.

[[Page 38583]]



  Table 28--Proposed Standard Salary and HCE Compensation Levels and Alternatives, Affected EAP Workers, Costs,
                                               and Transfers, 2013
----------------------------------------------------------------------------------------------------------------
                                                                             Year 1 impacts (Millions)
                                                   Affected EAP  -----------------------------------------------
           Alternative             Salary level       workers         Adj. &
                                                     (1,000s)       managerial       Transfers        DWL \b\
                                                                     costs \a\
----------------------------------------------------------------------------------------------------------------
                                         Standard Salary Level (Weekly)
----------------------------------------------------------------------------------------------------------------
Proposed........................            $921           4,646          $334.8        $1,440.8            $7.2
Alt. #1: Inflate 2004 levels....             561             576            36.1           127.9             0.5
Alt. #2: 2004 method............             577             734            44.5           151.5             0.7
Alt. #3: Kantor method..........             657           1,390            91.8           318.6             1.7
Alt. #4: Median full-time hourly             776           2,704           179.8           686.6             3.6
 and salaried workers...........
Alt. #5: Kantor short test......             979           5,632           404.2         1,821.3            10.3
Alt. #6: Median full-time                  1,065           6,855           522.1         2,525.8            14.8
 salaried.......................
Alt. #7: Inflate 1975 short test           1,083           7,128           543.0         2,666.1            15.5
 level..........................
----------------------------------------------------------------------------------------------------------------
                                        HCE Compensation Level (Annually)
----------------------------------------------------------------------------------------------------------------
Proposed........................        $122,148              36            $3.3           $41.7            $0.0
Alt. #1: No change..............         100,000               0  ..............  ..............  ..............
Alt. #2: 2004 percentile........         150,000              52             5.5            71.2             0.6
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Regulatory familiarization costs are excluded because they are a one-time cost that do not vary based on the
  proposed salary levels.
\b\ DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. Since
  the transfer associated with the minimum wage is negligible compared to the transfer associated with overtime
  pay, the vast majority of this cost is attributed to the overtime pay provision.

x. Projections
1. Methodology
    In addition to estimating Year 1 costs and transfers, the 
Department projected costs and transfers forward for ten years. To 
project costs and transfers, the Department used several pieces of 
data, specifically the median wage growth rate and the employment 
growth rate. These calculations are described below, after which the 
ten-year projected costs and transfers are presented.
    The projections presented in this section assume the proposed 
salary level remains constant over ten years. Thus, the number and 
percent of affected EAP workers decline over time as real earnings 
increase.\143\ The section on automatic updating of the salary level 
will present the estimated ten-year impacts based on how real earnings 
change relative to an automatically updated salary level because the 
selection of the salary level is conceptually separate from the 
decision to update (and how to update) the salary level. Thus, the 
costs and impacts of each are considered and presented separately.
---------------------------------------------------------------------------

    \143\ As described in the following paragraphs, the Department 
used historical wage growth rates to project wage growth rates.
---------------------------------------------------------------------------

    In order to identify workers whose projected salaries fall between 
the current salary level ($455 per week) and the proposed salary level 
based on the 40th earnings percentile ($921 per week), a wage growth 
rate must be applied to current earnings. The Department applied an 
annual real growth rate based on the average annual growth rate in 
median wages from 2005 to 2012.\144\ The wage growth rate is calculated 
as the geometric growth rate in median wages using the historical CPS 
MORG data for exempt workers by occupation-industry categories. The 
geometric growth rate is the constant annual growth rate that when 
compounded (applied to the first year's wage, then to the resulting 
second year's wage, etc.) yields the last historical year's wage. This 
method only depends on the value of the wage in the first available 
year and the last available year, and may be a flawed measure if either 
or both of those years were atypical; however, in this instance these 
values seem typical.
---------------------------------------------------------------------------

    \144\ In order to maximize the number of observations used in 
calculating the median wage for each occupation-industry group, 
three years of data were pooled for each of the endpoint years. 
Specifically, data from 2004, 2005, and 2006 (converted to 2005 
dollars) were used to calculate the 2005 median wage and data from 
2011, 2012, and 2013 (converted to 2012 dollars) were used to 
calculate the 2012 median wage.
---------------------------------------------------------------------------

    An alternative method would be to use the time series of median 
wage data to estimate the linear trend in the values and continue this 
to project future median wages. This method may be preferred if either 
or both of the endpoint years are outliers, since the trend will be 
less influenced by them. The linear trend may be flawed if there are 
outliers in the interim years (because these have no impact on the 
geometric mean but will influence the estimate of a linear trend). The 
Department chose to use the geometric mean because individual year 
fluctuations are difficult to predict and applying the geometric growth 
rate to each year provides a better estimate of the long-term growth in 
wages. Using this method is also consistent with the estimation of the 
employment growth rate as described below.
    The geometric wage growth rate was also calculated from the BLS' 
Occupational Employment Statistics (OES) and used as a validity 
check.\145\ Additionally, in occupation-industry categories where the 
CPS MORG data had an insufficient number of observations to reliably 
calculate median wages, the Department used the growth rate in median 
wages calculated from the OES data.\146\ Any remaining occupation-
industry combinations without estimated median growth rates were 
assigned the median of the growth rates in median wages from the CPS 
MORG data for EAP exempt workers.
---------------------------------------------------------------------------

    \145\ The difference between the OES and CPS growth measures 
averaged -0.0002 percent, but ranged from -7.2 to 5.8 percent, 
depending on the occupation-industry category. The CPS growth 
estimates were used as the primary source because the sample could 
be restricted to EAP exempt workers (the relevant population).
    \146\ To lessen small sample bias in the estimation of the 
median growth rate, this rate was only calculated using CPS MORG 
data when these data contained at least 10 observations in each time 
period.

---------------------------------------------------------------------------

[[Page 38584]]

    The Department calculated projected earnings for each worker in the 
sample by applying the annual projected wage growth rate to current 
earnings for each projected year. In each projected year, affected EAP 
workers were identified as those who are exempt in the current year 
(prior to the rule change) but have projected earnings in the projected 
year that are less than the proposed salary level.
    The employment growth rate is the geometric annual growth rate 
based on the ten-year employment projection from BLS' National 
Employment Matrix (NEM) within an occupation-industry category. This is 
the constant annual growth rate that when compounded yields the NEM 
ten-year projection. An alternative method is to spread the total 
change in the level of employment over the ten years evenly across 
years (constant change in the number employed). The Department believes 
that on average employment is more likely to grow at a constant 
percentage rate rather than by a constant level (a decreasing 
percentage rate). To account for population growth, the Department 
applied the growth rates to the sample weights of the workers. This is 
because the Department cannot introduce new observations to the CPS 
MORG data to represent the newly employed.
    Affected EAP workers may experience a reduction in hours since the 
wage they receive for overtime hours is higher after the proposed 
rulemaking. The reduction in hours is calculated as described in 
section VII.D.v. The only difference is that for projections the long-
run elasticity of labor demand is used instead of the short-run 
elasticity. The Department used a long-run elasticity of -0.4.\147\
---------------------------------------------------------------------------

    \147\ This elasticity estimate is based on the Department's 
analysis of the following paper: Lichter, A., Peichl, A. & Siegloch, 
A. (2014). The Own-Wage Elasticity of Labor Demand: A Meta-
Regression Analysis. IZA DP No. 7958.
---------------------------------------------------------------------------

2. Estimated Projections
    Projected costs and transfers both depend on the projected number 
of affected EAP workers. The Department estimated that in Year 1 4.7 
million EAP workers will be affected, with about 36,000 of these 
attributable to the revised HCE compensation level. In Year 10, if the 
salary levels are not updated, the number of affected EAP workers was 
estimated to equal 2.7 million, with fewer than 8,000 attributed to the 
HCE exemption. The projected number of affected EAP workers accounts 
for projected employment growth by increasing the number of workers 
represented by the affected EAP workers (i.e., increasing sampling 
weights). However, with no additional changes in the salary level and 
most workers experiencing positive wage growth, workers affected in 
Year 1 become less likely to still be affected in each future year. 
That is, some of these workers return to exempt status over time as 
their growing salaries eventually exceed the proposed standard salary 
level of $921 per week. The net impact is a decrease in the number of 
affected EAP workers in each subsequent year.
    The projected number of affected workers only includes workers who 
were originally determined to be exempt in 2013. However, additional 
workers may be affected in future years who were not EAP exempt in the 
base year but would have become exempt in the absence of this proposed 
rule. For example, a worker may earn less than $455 in 2013 but at 
least $455 (and less than the proposed salary level) in subsequent 
years; such a worker would not be counted as an affected worker in the 
projections above. In the absence of this proposed rule he or she would 
likely have become exempt at some point in the 10-year projections 
period; however, as a result of the proposed rule, this worker remains 
nonexempt, and is thus affected by the proposed rule.
    Therefore, the Department estimated the number of workers who were: 
Paid on a salary basis, pass the duties test, and earn less than $455 
per week in 2013, but are projected to earn at least $455 (but less 
than the proposed salary level) per week at some point in the following 
nine years. The Department found that in Year 10, an additional 398,000 
workers meet these criteria and therefore are also affected workers. 
Similarly, the Department estimated the number of workers who are paid 
on a salary basis, meet the HCE duties test, and currently earn less 
than $100,000 annually but are projected to earn more than $100,000 per 
year at some point in the following nine years. The Department 
estimated that, in Year 10, an additional 115,000 workers meet these 
criteria and therefore are also affected workers. The Department did 
not estimate costs, transfers, or DWL for these workers because it 
would be necessary to make additional assumptions such as how employers 
respond by adjusting workers' wages and hours.
    The Department quantified three types of direct employer costs in 
the ten-year projections: (1) Regulatory familiarization costs; (2) 
adjustment costs; and (3) managerial costs. Regulatory familiarization 
costs are one-time costs and only occur in Year 1. Although start-up 
firms must still become familiar with the FLSA following Year 1, the 
difference between the time necessary for familiarization with the 
current part 541 exemptions and those exemptions as modified by the 
proposed rule is essentially zero. Therefore, projected regulatory 
familiarization costs over the next nine years are zero. Similarly, 
adjustment costs are only incurred when workers' status changes from 
exempt to nonexempt, so adjustment costs are incurred predominately in 
Year 1 (some very minor adjustment costs may exist in projected years 
because some workers' earnings decrease and thus these workers may 
transition from exempt to nonexempt).
    However, managerial costs recur for all affected EAP workers whose 
hours are adjusted and were projected through Year 10. The Department 
estimated that Year 1 managerial costs would be $178.1 million (section 
VII.D.iv.4.); by Year 10 these costs would fall to $93.1 million (Table 
29). Over 97 percent of this amount ($176.0 million) in Year 1, and 
roughly 99 percent ($92.6 million) in Year 10 is attributable to the 
revised standard salary level. The projected reduction in managerial 
costs over the years is due to the reduction in the number of affected 
EAP workers over time as workers' earnings increase relative to the 
constant salary and compensation levels.
    The Department also projected two transfers associated with workers 
affected by the proposed regulation: (1) Transfers to workers from 
employers due to the minimum wage provision and (2) transfers to 
workers from employers due to the overtime pay provision. Transfers to 
workers from employers due to the minimum wage provision, estimated to 
be $46.7 million in Year 1, are projected to decline to $9.9 million in 
Year 10 as increased earnings over time move workers' regular rate of 
pay above the minimum wage.\148\ Transfers to workers from employers 
due to the overtime pay provision decrease from $1,435.8 million in 
Year 1 to $490.2 million in Year 10. Workers affected by the revised 
standard salary level account for 97 percent of overtime transfers in 
Year 1, and 99 percent in Year 10. Again, the decrease in transfers is 
primarily due to the reduction in the number of affected workers over 
time.
---------------------------------------------------------------------------

    \148\ In states with higher minimum wages, then effective state 
minimum wages were used in 2013 and 2014 and minimum wages on 
December 31, 2014 were used for projected years.
---------------------------------------------------------------------------

    Table 29 also summarizes average annualized costs and transfers 
over the ten-year projection period, using 3

[[Page 38585]]

percent and 7 percent real discount rates. The Department estimated 
that total direct employer costs have an average annualized value of 
$194.2 million per year over ten years when using a 7 percent real 
discount rate. Of this total, average annualized regulatory 
familiarization costs were estimated to be $33.9 million; the 
Department does not apportion these out between the revised standard 
salary and HCE annual compensation levels. Average annualized 
adjustment costs were estimated to be $21.5 million; roughly 99 percent 
of adjustment costs were attributed to the revised standard salary 
level. The remaining $138.9 million in average annualized direct costs 
were accounted for by managerial costs, of which 99 percent were 
associated with the revised standard salary level.
    The average annualized value of total transfers was estimated to 
equal $872.9 million. The largest component of this was the average 
annualized transfer from employers to workers due to overtime pay, 
which was $843.6 million per year, while average annualized transfers 
due to the minimum wage totaled $29.3 million per year. None of the 
transfer associated with the minimum wage was attributed to the revised 
HCE compensation level. Although composing less than one percent of 
affected workers, those receiving overtime due to the revised HCE 
compensation level account for 2.2 percent of total average annualized 
transfers ($19.5 of $872.9 million) because of their high implicit 
regular hourly rate of pay. The remaining $853.4 million in transfers 
accrue to those affected by the revised standard salary level.

                           Table 29--Projected Costs and Transfers Without Automatic Updating, Standard and HCE Salary Levels
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                           Affected EAP                        Costs                                 Transfers
              Year (Year #)                   workers    --------------------------------------------------------------------------------     DWL \a\
                                            (Millions)       Reg. Fam.      Adjustment      Managerial       Due to MW       Due to OT
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                          ..............                                         (Millions 2013$)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Year
2013 (1)                                             4.7          $254.5          $160.1          $178.1           $46.7        $1,435.8            $7.4
2014 (2)                                             4.5             0.0             1.1           169.0            44.0         1,017.1             9.8
2015 (3)                                             4.2             0.0             0.0           155.8            39.5           923.9             8.9
2016 (4)                                             4.0             0.0             0.0           146.1            33.0           843.1             8.1
2017 (5)                                             3.8             0.0             0.0           137.5            27.4           771.4             7.5
2018 (6)                                             3.6             0.0             0.0           128.5            22.6           702.0             6.8
2019 (7)                                             3.4             0.0             0.0           118.4            18.3           640.1             6.0
2020 (8)                                             3.1             0.0             0.0           108.9            14.9           582.0             5.3
2021 (9)                                             2.9             0.0             0.0           100.6            11.8           539.2             4.9
2022 (10)                                            2.7             0.0             0.1            93.1             9.9           490.2             4.3
Average Annualized
3% real rate                              ..............            29.0            18.4           135.9            27.9           815.7             7.0
7% real rate                              ..............            33.9            21.5           138.9            29.3           843.6             7.2
--------------------------------------------------------------------------------------------------------------------------------------------------------
\a\ DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions. Since the transfer associated with the minimum
  wage is negligible compared to the transfer associated with overtime pay, the vast majority of this cost is attributed to the overtime pay provision.

    The cost to society of lower employment expressed as DWL was 
estimated to be $7.4 million in Year 1. After year 2, DWL falls over 
time; in Year 10 it is projected to equal $4.3 million. DWL increases 
sharply between Year 1 and Year 2 because the Department assumes the 
market has had time to fully adjust to the revised standard salary and 
HCE annual compensation levels by Year 2. In Year 1 employers may not 
be able to fully adjust wages and hours in response to the rulemaking, 
so the Department used a short run wage elasticity of labor demand to 
reflect this constrained response; in Year 2 employers have sufficient 
time to fully adjust, and a long run wage elasticity is used. 
Therefore, the decrease in hours worked is larger in Year 2 than Year 
1, and the DWL is also larger. Finally, the Department estimated that 
average annualized DWL was $7.2 million per year; about $200,000 of DWL 
(2.7 percent) was attributed to affected HCE workers, and the remaining 
$7.0 million was attributed to workers affected by the revised standard 
salary level.
    A summary of the estimates used in calculating DWL for years 1, 2 
and 10 is presented in Table 30. The size of the DWL depends on the 
change in average hourly wages, the change in average hours, and the 
number of affected EAP workers. While the change in average hourly 
wages generally tends to increase over time in the projected years, the 
number of affected EAP workers decreases over time; because the 
relative decrease in workers is larger than the relative increase in 
wages after Year 2, there is a net decrease in annual DWL over time.

[[Page 38586]]



                         Table 30--Summary of Projected Deadweight Loss Component Values
----------------------------------------------------------------------------------------------------------------
 
----------------------------------------------------------------------------------------------------------------
                                                                      Year 1               Future years
                                                                                 -------------------------------
Component                                                         ..............          Year 2         Year 10
                                                                 -----------------------------------------------
                                                                                     Standard
----------------------------------------------------------------------------------------------------------------
Average hourly wages
    Pre.........................................................          $15.01          $15.09          $15.47
    Post \a\....................................................          $15.70          $15.59          $15.98
    Change......................................................           $0.68           $0.50           $0.51
Change in average overtime hours................................           -0.40           -0.77           -0.75
Affected EAP workers (1,000s)...................................           1,010             959             532
DWL
    Per worker per week.........................................           $0.14           $0.19           $0.19
    Nominal annual (millions)...................................            $7.2            $9.7            $5.3
    Real annual (millions of 2013$).............................            $7.2            $9.4            $4.3
----------------------------------------------------------------------------------------------------------------
                                                                  ..............                HCE
----------------------------------------------------------------------------------------------------------------
Average hourly wages
    Pre.........................................................          $40.31          $40.48          $46.19
    Post [a]....................................................          $42.45          $41.96          $47.67
    Change......................................................           $2.14           $1.48           $1.47
Change in average overtime hours................................           -0.41           -0.80           -0.62
Affected EAP workers (1,000s)...................................              12              11               3
DWL
    Per worker per week.........................................           $0.44           $0.59           $0.46
    Nominal annual (millions)...................................           $0.27           $0.34           $0.07
    Real annual (millions of 2013$).............................           $0.27           $0.34           $0.07
----------------------------------------------------------------------------------------------------------------
Note: DWL analysis is limited to workers in Types 2 and 3 who experience hour adjustments.
\a\ Despite general growth in wages, the average wage may fall slightly from Year 1 to Year 2 because the
  population has changed.

    In conclusion, because the number of affected EAP workers and 
consequently all costs and transfers diminish over time, the economic 
impact of the regulation will decrease over time as the real value of 
the salary levels fall. This occurs because real wages increase over 
time while the proposed salary levels would remain constant without 
automatic updating. However, if the salary levels are annually updated, 
the projected costs and transfers would increase over time. Cost and 
transfer projections based on the proposed standard salary level with 
annual updates are examined in section VII.E.iii.

E. Automatic Updates

i. Background
    Between periodic updates to the salary level, nominal wages 
typically increase, resulting in an increase in the number of workers 
qualifying for the EAP exemption even if there has been no change in 
their duties or real earnings. Thus, workers whom Congress intended to 
be covered by the minimum wage and overtime pay provisions of the FLSA 
lose that protection. Automatically updating the salary level would 
allow the level to keep pace with changes in either prices or earnings, 
keeping the real value of the salary level constant over time.
    The Department proposes to include in the regulations a mechanism 
for automatically updating the proposed standard salary level and 
proposed HCE annual compensation level annually either by maintaining a 
fixed percentile of earnings (40th and 90th percentile of weekly wages 
for full-time salaried workers, respectively) or by updating the salary 
and compensation levels based on changes in the CPI-U. Automatically 
updating the EAP standard salary level and HCE compensation level would 
allow these levels to continue to serve as an effective dividing line 
between potentially exempt and nonexempt workers.
    Furthermore, automatically updating the standard salary level and 
HCE compensation level would provide employers more certainty in 
knowing that the salary and compensation levels would change by a small 
amount each year, rather than the more disruptive increases caused by 
much larger changes after longer, uncertain increments of time. This 
would allow firms to better predict short- and long-term costs and 
employment needs.
ii. Automatic Updating Methods
1. Introduction
    There are many indices that could be used to adjust the salary 
levels. In general, these indices are classified into two groups: Price 
indices and earnings indices.
    Price indices are normalized averages of prices used to measure the 
change in the average level of prices in an economy over time. The 
general growth rate of prices, also known as the inflation rate, is 
calculated as the annual percentage increase in the average price 
level. A price index is intended to measure the cost of achieving a 
given level of economic well-being or utility.\149\ Because one cannot 
directly observe utility or well-being, a ``market basket'' of goods 
and services is selected to represent a given level of utility. By 
keeping the contents of this basket constant, one can approximate the 
cost of obtaining the same level of utility at different points in 
time. In order to keep utility or the cost-of-living constant, incomes 
must rise by the same amount as the price index.
---------------------------------------------------------------------------

    \149\ Nordhaus, W.D. (1998). Quality Change in Price Indexes. 
Journal of Economic Perspectives, 12(1), 59-68.
---------------------------------------------------------------------------

    An alternative to indexing the salary level to a price level is to 
update the salary level based upon an earnings measure. Price indices 
are intended to keep a consumer's utility constant by adjusting for 
changes in the cost of living due to inflation. However, while price 
indices account for changes to the price of products in the market 
basket, they may not reflect the real growth in

[[Page 38587]]

wages, growth that might result in the ability to purchase a larger 
``market basket.'' Updating the salary level by maintaining it at a 
fixed percentile of earnings would reflect real growth in wages and 
keep the percentage of workers exempt roughly constant over time, but 
may not fully account for inflation in all circumstances.
2. Updating Methods Considered
    This section details the price and earnings indices that were 
considered as methods to update the salary levels. The Department 
assessed each method's strengths, weaknesses, and current use. The 
methods considered include:
     Consumer Price Index for all urban consumers (CPI-U)
     Chained CPI (C-CPI-U)
     Earnings percentiles (fixed percentiles of the 
distribution of weekly earnings for full-time salaried workers)

The CPI-U

    The CPI-U is the most commonly used price index in the U.S. and is 
calculated monthly by BLS. The CPI-U holds quantities constant at base 
levels while allowing prices to change. The quantities are fixed to 
represent a ``basket of goods and services'' bought by the average 
consumer. However, most economists believe that the CPI-U overestimates 
the rate of inflation, although there are a broad range of views as to 
the sources and size of the overestimate. CPI-U estimates are generally 
not subject to revision.
    The CPI-U is the primary index used by the government to index 
benefit payments, program eligibility levels, and tax payments, 
including:
     Federal income tax brackets, personal exemptions, and 
standard deductions.\150\
---------------------------------------------------------------------------

    \150\ 26 U.S.C. 1(f).
---------------------------------------------------------------------------

     Both eligibility for and benefits under the Earned Income 
Tax Credit (EITC).\151\
---------------------------------------------------------------------------

    \151\ Id.
---------------------------------------------------------------------------

     Funding allocated to some government grants, such as 
funding to the Nutrition Education and Obesity Prevention Grant 
Program.\152\
---------------------------------------------------------------------------

    \152\ 7 U.S.C. 2036a(d)(1)(F).
---------------------------------------------------------------------------

     Treasury inflation-indexed debt securities' interest 
rates.\153\
---------------------------------------------------------------------------

    \153\ 31 CFR part 456, appendix D.
---------------------------------------------------------------------------

     Many government programs' income eligibility requirements, 
including school meal programs.\154\
---------------------------------------------------------------------------

    \154\ 42 U.S.C. 1758(b)(1).
---------------------------------------------------------------------------

     Federal poverty levels, which determine eligibility for 
many government social assistance programs.\155\
---------------------------------------------------------------------------

    \155\ 42 U.S.C. 9902(2).
---------------------------------------------------------------------------

The Chained CPI-U (C-CPI-U)
    The C-CPI-U is a variation of the CPI-U. The C-CPI-U is an index 
that accounts for changes in the market basket of goods from one year 
to the next. The C-CPI-U results in inflation estimates roughly 0.3 
percentage points lower than the CPI-U.\156\
---------------------------------------------------------------------------

    \156\ Congressional Budget Office. (2010). Using a Different 
Measure of Inflation for Indexing Federal Programs and the Tax Code. 
https://www.cbo.gov/publication/25036.
---------------------------------------------------------------------------

    Although the C-CPI-U is viewed by some as a more accurate measure 
of inflation than the CPI-U, it has shortcomings as an indexation 
method. ``The C-CPI-U requires data on changes in consumers' spending 
patterns. Since those data are not available for several years the BLS 
releases preliminary estimates of the C-CPI-U and revises them over the 
following two years.'' \157\ Thus any measure using the C-CPI-U would 
have to be either (1) indexed to a preliminary estimate of the C-CPI-U 
that is subject to estimation error and revision or (2) indexed to 
changes in prices from a few years prior.
---------------------------------------------------------------------------

    \157\ See https://www.bls.gov/cpi/cpisupqa.htm.
---------------------------------------------------------------------------

    Earnings percentiles (fixed percentiles of the distribution of 
weekly earnings for full-time salaried workers)
    Updating the salary levels based upon the growth rate of earnings 
at a specified percentile of the earnings distribution is consistent 
with the Department's historical practice of using salary level as a 
key criterion for the exemption. The growth rate of earnings reflecting 
labor market conditions is an appropriate measure of the relative 
status, responsibility, and independence that characterize exempt 
workers.
    While earnings and prices generally mirror one another over time, 
they do not change in tandem. A price index maintains a constant level 
of utility or economic well-being; an earnings index reflects real 
gains in the standard of living. Accordingly, if earnings grow more 
quickly than prices an earnings index will increase the salary levels 
by more than a price index. Conversely, if prices grow more quickly 
than earnings a price index will increase the salary levels more than 
an earnings index.
3. Comparison of Indices
    The Department proposes to automatically update the standard salary 
level and the HCE annual compensation level annually either by 
maintaining them at a fixed percentile of earnings (the 40th and 90th 
percentiles of weekly wages for full-time salaried workers, 
respectively) or by updating the levels based on changes in the CPI-U. 
Updating salary and compensation levels based on earnings would keep 
the share of workers who are exempt fairly constant over time, while 
updating based on prices will keep the earnings power of the levels 
constant over time.
    The Department is seeking detailed comments on both methods of 
updating the standard salary and HCE compensation levels. The CPI-U is 
based on a tremendous amount of data that represents average prices 
paid by a majority of Americans and is by far the best-known and most 
widely-used index. While earnings percentiles are less familiar, BLS 
publishes the deciles of weekly earnings for full-time salaried workers 
on a quarterly basis. In recent years the CPI-U has grown at a rate 
very closely aligned with the 40th percentile of earnings for full-time 
salaried workers; between 2003 and 2013 the average annual growth rates 
for the 40th percentile and CPI-U have been: 2.6 percent and 2.4 
percent respectively. The Department therefore expects that both 
methods would produce similar standard salary levels in future years. 
Growth in CPI-U in recent years has been smaller than growth at the 
90th percentile of earnings, however, so the HCE total annual 
compensation levels generated by these two methods may vary in the 
future.
iii. Estimated Impacts of Automatically Updating the EAP Salary and HCE 
Compensation Levels
    In section VII.D.x. the Department projected ten years of costs and 
transfers due to a one-time increase in the standard salary and HCE 
compensation levels. Updating these salary levels annually will 
increase the number of affected workers because more workers will earn 
below the higher indexed salary levels than the fixed salary levels. 
Consequently, the projected costs and transfers of the proposed rule 
will increase with indexation.
    In this section, the Department describes and quantifies the annual 
costs and transfer payments associated with automatically updating the 
salary levels under both methods (fixed percentile and CPI-U). To 
predict the salary and compensation levels in 2014 through 2022 using 
the fixed percentile method, the Department estimated the salary levels 
using data from 2003 through 2013, calculated the geometric average 
annual growth rate, and applied it to the future years. For example, 
between 2003 and 2013 the 40th percentile of earnings for full-time 
salaried workers increased by an average of 2.6 percent annually; 
therefore, the projected salary level for Year 2 is $945 ($921 x 
1.026). For the

[[Page 38588]]

CPI-U method, the Department used the predicted annual CPI-U values for 
2014 through 2022 from the Congressional Budget Office.\158\ For 
example, CPI-U for 2014 is predicted to be 1.5 percent; therefore, the 
projected salary level for Year 2 is $935 ($921 x 1.015). In other 
years, predicted CPI-U ranges from 1.9 percent to 2.4 percent.
---------------------------------------------------------------------------

    \158\ Congressional Budget Office. (2014). The Budget and 
Economic Outlook: 2014 to 2024. Pub. No. 4869. Table G-2.
---------------------------------------------------------------------------

    As the required salary levels are updated in Year 2 through Year 10 
of the analysis, more workers will potentially be affected with 
automatic updating than without. With automatic updating of the salary 
levels, the number of affected EAP workers is projected to increase 
from 4.7 million to between 5.1 and 5.6 million over 10 years, 
depending on the updating methodology used. Conversely, in the absence 
of automatic updating, the number of affected EAP workers is projected 
to decline from 4.7 to 2.7 million (Table 31). The relatively constant 
number of affected workers over the years with updating validates the 
choice of indexing methods. Starting in Year 1 and running through Year 
10 the population of affected workers as a percent of potentially 
affected workers (defined using the current salary level) increases 
modestly from 21.9 to 23.4 percent using the fixed percentile method, 
but declines modestly to 21.2 percent using the CPI-U method.
    The three costs to employers previously considered are (1) 
regulatory familiarization costs, (2) adjustment costs, and (3) 
managerial costs. Regulatory familiarization costs only occur in Year 1 
and thus do not vary with automatic updating. Adjustment costs and 
managerial costs are a function of the number of affected EAP workers 
and so will be higher with automatic updating. Adjustment costs will 
occur in projected years when workers are newly affected (which--while 
relatively rare--will be more common with automatic updating than 
without). Management costs recur each year for all affected EAP workers 
whose hours are adjusted. Therefore, managerial costs fall 
significantly over time without updating (since the number of affected 
EAP workers decreases over time) but increase modestly over time with 
annual updating (where the number of affected EAP workers increases 
over time because of the higher salary level). Similarly, transfers and 
DWL will both be higher with automatic updating than without because 
the number of affected workers will increase, rather than decrease, 
over time.
    Table 31 presents the projected estimated costs, transfer payments, 
and DWL with and without automatic updating. Total direct costs were 
projected to decrease from $592.7 million in Year 1 to $225.3 million 
in Year 10 with fixed percentile updating and to $198.6 million in Year 
10 with CPI-U updating. In the absence of automatic updating, costs 
were projected to decrease to $93.1 million in Year 10. Transfers from 
employers to employees were projected to decrease from $1,482.5 million 
to $1,339.6 million using the fixed percentile method, and to $1,191.4 
million using the CPI-U method. Without updating, transfers were 
projected to decrease to $500.1 million in Year 10. DWL increases over 
time with automatic updating, but decreases over time without it. With 
updating, DWL was estimated to increase from $7.4 million to $11.2 
million (fixed percentile method) or to $9.7 million (CPI-U method), 
but decline from $7.4 million to $4.3 million without updating.

                      Table 31--Projected Costs and Transfers; Standard and HCE Salary Levels, With and Without Automatic Updating
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                         Year
                Automatic updating method \a\                -------------------------------------------------------------------------------------------
                                                                   1            2            3           . . .          8            9            10
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                               Affected Workers (Millions)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Without.....................................................          4.7          4.5          4.2        . . .           3.1          2.9          2.7
Percentile..................................................          4.7          4.8          4.9        . . .           5.4          5.5          5.6
CPI-U.......................................................          4.7          4.7          4.7        . . .           4.9          5.0          5.1
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                      Total Direct Employer Costs (Millions 2013$)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Without.....................................................       $592.7       $170.0       $155.8        . . .        $108.9       $100.6        $93.1
Percentile..................................................        592.7        188.8        191.9        . . .         214.8        220.1        225.3
CPI-U.......................................................        592.7        181.1        178.6        . . .         191.6        195.2        198.6
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                            Total Transfers (Millions 2013$)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Without.....................................................     $1,482.5     $1,061.2       $963.4        . . .        $596.9       $551.0       $500.1
Percentile..................................................      1,482.5      1,160.2      1,162.4        . . .       1,315.2      1,320.6      1,339.6
CPI-U.......................................................      1,482.5      1,126.4      1,104.3        . . .       1,150.6      1,192.7      1,191.4
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                  DWL (Millions 2013$)
--------------------------------------------------------------------------------------------------------------------------------------------------------
Without.....................................................         $7.4         $9.8         $8.9        . . .          $5.3         $4.9         $4.3
Percentile..................................................          7.4         10.8         10.9        . . .          11.0         11.1         11.2
CPI-U.......................................................          7.4         10.3         10.1        . . .           9.7          9.7          9.7
--------------------------------------------------------------------------------------------------------------------------------------------------------
Note: For the purposes of projecting costs, transfers, and DWL, Year 1 corresponds to 2013 and Year 10 corresponds to 2022.
\a\ The percentile method sets the standard salary level at the 40th percentile of weekly earnings for full-time salaried workers and the HCE
  compensation level at the 90th percentile. The CPI-U method adjusts both salary levels based on the annual percent change in the CPI-U.


[[Page 38589]]

    In Years 1 through 10, using a 7 percent real discount rate, total 
annualized adjustment and managerial costs were estimated to average 
between $205.7 and $221.4 million per year with automatic updating 
(using CPI-U or fixed percentile, respectively) and $160.3 million 
without updating (Table 32). Therefore, the incremental average 
annualized direct employer costs of automatic updating is between $45.4 
and $61.1 million per year. Average annualized total transfers were 
estimated to be between $1,178.0 and $1,271.4 million with automatic 
updating (using CPI-U or fixed percentile, respectively) and $872.9 
million without updating, resulting in incremental transfers of between 
$305.2 and $398.5 million per year. Projected average annualized DWL 
totals between $9.5 and $10.5 million per year with automatic updating 
(using CPI-U or fixed percentile, respectively) and $7.2 million per 
year without updating. Thus, automatic updating increases DWL by 
between $2.3 and $3.3 million per year on average. Benefits were not 
monetized for either Year 1 or Years 2 through 10; therefore this 
section does not repeat the previous discussion on potential benefits.

Table 32--Summary of Ten-Year Average Annualized Regulatory Costs and Transfers, Standard and HCE Salary Levels,
                                       With and Without Automatic Updating
----------------------------------------------------------------------------------------------------------------
                                                          Average annualized values (Millions 2013$) \a\
                                                ----------------------------------------------------------------
                 Cost/transfer                                    Fixed percentile                CPI-U
                                                   Without   ---------------------------------------------------
                                                   updating      Values     Difference     Values     Difference
----------------------------------------------------------------------------------------------------------------
                                        Regulatory Familiarization Costs
----------------------------------------------------------------------------------------------------------------
Regulatory familiarization \b\.................        $33.9        $33.9         $0.0        $33.9         $0.0
----------------------------------------------------------------------------------------------------------------
                                              Standard Salary Level
----------------------------------------------------------------------------------------------------------------
Adj. & managerial costs........................       $158.7       $218.6        $59.9       $203.3        $44.6
Transfers......................................        853.4      1,232.4        379.1      1,144.2        290.8
DWL............................................          7.0         10.0          3.0          9.2          2.2
----------------------------------------------------------------------------------------------------------------
                                             HCE Compensation Level
----------------------------------------------------------------------------------------------------------------
Adj. & managerial costs........................         $1.7         $2.9         $1.2         $2.4         $0.8
Transfers......................................         19.5         39.0         19.5         33.8         14.3
DWL............................................          0.2          0.5          0.3          0.3          0.2
----------------------------------------------------------------------------------------------------------------
                                                      Total
----------------------------------------------------------------------------------------------------------------
Adj. & managerial costs........................       $160.3       $221.4        $61.1       $205.7        $45.4
Transfers......................................        872.9      1,271.4        398.5      1,178.0        305.2
DWL............................................          7.2         10.5          3.3          9.5          2.3
----------------------------------------------------------------------------------------------------------------
\a\ Over ten years, using a discount rate of 7 percent.
\b\ Regulatory familiarization costs are a one-time cost that do not vary based on the proposed salary levels or
  automatic updating.

    The above table demonstrates that the two updating methods yield 
similar costs and transfers estimates. However, this does not imply 
these indices will necessarily result in similar salary levels over 
time. The Department compared the standard salary levels that would 
have resulted from 1998 to 2013 if (1) the standard salary level was 
set each year to the 40th percentile of weekly earnings for full-time 
salaried workers, and (2) the standard salary level was set using the 
growth in the CPI-U (and setting the level in 2013 to match the 40th 
percentile earnings level, i.e., $921 per week) (Figure 5). While not 
identical, the data show that during this sixteen year period these two 
methods produced similar results.

[[Page 38590]]

[GRAPHIC] [TIFF OMITTED] TP06JY15.004

F. Duties Test

    The Department has not proposed specific revisions to the standard 
duties tests; however, as mentioned in section III., we received 
significant input regarding this issue from both employer and employee 
representatives during the Department's stakeholder listening sessions. 
If changes were made to the standard duties tests, the Department would 
need to consider whether any of the probabilities of exemption for 
specific occupations used in the analysis would need to be revised 
since the new duties test would potentially result in workers in some 
occupations being more or less likely to meet the duties tests.
    The Department has begun to consider whether O*NET can be used to 
identify any occupations for which the Department may need to adjust 
its assumptions of the likelihood of exemption should the Department 
revise the duties test. The O*NET database contains information on 
hundreds of standardized and occupation-specific descriptors. The 
database, which is available to the public, is continually updated by 
surveying a broad range of workers from each occupation. The database 
of occupational requirements and worker attributes describes 
occupations in terms of the skills and knowledge required, how the work 
is performed, and typical work settings.\159\
---------------------------------------------------------------------------

    \159\ See https://www.onetcenter.org/overview.html.
---------------------------------------------------------------------------

    For each occupation, O*NET includes a list of tasks performed, and 
rates the tasks' frequency, importance, relevance, and whether it is a 
core or supplemental task. O*NET also includes data on work activities, 
including the importance, relevancy, and frequency of specified tasks 
performed in each occupation. This information could inform the 
Department in determining whether the task is indicative of exempt 
duties.
    The Department believes it could use O*NET data to construct a 
model to identify occupations for which the probability of exemption 
would be impacted by any changes to the duties tests. The Department 
also could look to O*NET data to determine changes to the probability 
codes for those identified occupations. Therefore, if there are any 
changes to the duties test, the Department would likely update its 
estimate of the impact of the rule based on its analysis of the O*NET 
data for any occupations for which the probability codes were modified.
    The Department invites detailed comments on this proposed 
methodology and alternative data sources for determining the impact of 
any changes to the standard duties tests.

Appendix A: Methodology for Estimating Exemption Status

    The number of workers exempt under the FLSA's part 541 regulations 
is unknown. It is neither reported by employers to any central agency 
nor asked in either an employee or establishment survey. The Department 
estimated the number of exempt workers using the following methodology. 
This methodology is based largely on the approach used during the 2004 
revisions.\160\ This appendix expands on the methodology description in 
this NPRM. The methodology explained there is not repeated here unless 
additional details are provided.
---------------------------------------------------------------------------

    \160\ 69 FR 22196-22209 (Apr. 23, 2004).
---------------------------------------------------------------------------

A.1 The Duties Tests Probability Codes

    The CPS MORG data do not include information about job duties. To 
determine whether a worker meets the duties test the Department again 
employs the methodology it used in the 2004 Final Rule. Each occupation 
is assigned a probability representing the odds that a worker in that 
occupation would pass the duties test. For the EAP duties test, the 
five probability intervals are:
     Category 0: Occupations not likely to include any workers 
eligible for the EAP exemptions.
     Category 1: Occupations with probabilities between 90 and 
100 percent.
     Category 2: Occupations with probabilities between 50 and 
90 percent.
     Category 3: Occupations with probabilities between 10 and 
50 percent.
     Category 4: Occupations with probabilities between 0 and 
10 percent.\161\
---------------------------------------------------------------------------

    \161\ Table A2 lists the probability codes by occupation used to 
estimate exemption status.
---------------------------------------------------------------------------

    The occupations identified in this classification system represent 
an earlier occupational classification scheme (the 1990 Census Codes). 
Therefore, an occupational crosswalk was used to map the previous 
occupational codes to

[[Page 38591]]

the 2002 Census Occupational Codes which are used in the CPS MORG 2002 
through 2010 data.162 163 When the new occupational category 
was comprised of more than one previous occupation, the Department 
assigned a probability category using the weighted average of the 
previous occupations' probabilities, rounded to the closest category 
code.
---------------------------------------------------------------------------

    \162\ To match 1990 Census Codes to the corresponding 2000 
Census Codes see: https://www.census.gov/people/io/methodology/. To 
translate the 2000 Census Codes into the 2002 Census Codes each code 
is multiplied by 10.
    \163\ Beginning January 2011, the MORG data use the 2010 Census 
Codes. The Department translates these codes into the equivalent 
2002 Census Codes to create continuity. The crosswalk is available 
at: https://www.census.gov/people/io/methodology/.
---------------------------------------------------------------------------

    Next, the Department must determine which workers to classify as 
exempt.\164\ For example, the probability codes indicate that out of 
every ten public relation managers between five and nine are exempt; 
however, the Department does not know which five to nine workers are 
exempt. Exemption status could be randomly assigned but this would bias 
the earnings of exempt workers downward, since higher paid workers are 
more likely to perform the required duties. Therefore, the probability 
of being classified as exempt should increase with earnings. First, the 
Department assigned the upper bound of the probability range in each 
exemption category to workers with top-coded weekly earnings. For all 
other white collar salaried workers earning at least $455 per week in 
each exemption category, the Department estimated the probability of 
exemption for each worker in the data based on both occupation and 
earnings using a gamma distribution.165 166 For the gamma 
distribution, the shape parameter alpha was set to the squared quotient 
of the sample mean divided by the sample standard deviation, and the 
scale parameter beta was set to the sample variance divided by the 
sample mean. These parameter calculations are based on the method 
described in the 2004 rulemaking, except for the use of the standard 
deviation instead of the standard error.\167\ Table A1 shows that the 
expected number of workers exempt using a gamma distribution method is 
similar to the expected number exempt when assigning the midpoint of 
each probability code range to all workers in that probability code. 
After determining the probabilities of exemption for each worker in the 
data (dependent on both occupation and earnings), the Department 
randomly assigns exemption status to each worker, conditional on the 
worker's probability of exemption.
---------------------------------------------------------------------------

    \164\ These probabilities are applied to the population of 
workers who are either (1) in occupational categories associated 
with named occupations or (2) white collar, earn $455 or more per 
week, and are salaried.
    \165\ The gamma distribution was chosen because during the 2004 
revision it fit the data the best of the non-linear distributions 
considered, which included normal, lognormal, and gamma. 69 FR 
22204-08.
    \166\ A gamma distribution is a general type of statistical 
distribution that is based on two parameters, in this case alpha and 
beta.
    \167\ Since the standard error is much smaller than the sample 
standard deviation, using the standard error to calculate the shape 
and location parameters resulted in probabilities that vary less 
with earnings.

       Table A1--Comparison of Part 541--Exempt Worker Estimates a
------------------------------------------------------------------------
                                         Midpoint            Gamma
     Probability code category         probability        distribution
                                         estimate        model estimate
------------------------------------------------------------------------
High probability of exemption (1).         21,947,066         22,014,576
Probably exempt (2)...............          4,557,146          4,573,895
Probably not exempt (3)...........          1,617,632          1,605,096
Low or no probability of exemption            281,382            297,336
 (4)..............................
    Total.........................         28,403,227         28,490,903
------------------------------------------------------------------------
\a\ Numbers shown are the expected value of the number of workers exempt
  in each of the four probability code categories.

    The 2004 Final Rule assigned probabilities for whether workers in 
each occupation would pass the HCE abbreviated duties test if they 
earned $100,000 or more in total annual compensation; these 
probabilities are:
     Category 0: Occupations not likely to include any workers 
eligible for the HCE exemption.
     Category 1: Occupations with a probability of 100 percent.
     Category 2: Occupations with probabilities between 94 and 
96 percent.
     Category 3: Occupations with probabilities between 58.4 
and 60 percent.
     Category 4: Occupations with a probability 15 percent.
    Like under the standard test, there is a positive relationship 
between earnings and exemption status; however, unlike the standard 
test, the relationship for the HCE analysis can be represented well 
with a linear function. Once individual probabilities are determined, 
workers are randomly assigned to exemption status.
A.2 Other Exemptions
    There are many other exemptions to the minimum wage and overtime 
pay provisions of the FLSA. Accordingly, in the 2004 Final Rule, the 
Department excluded workers in agriculture and certain transportation 
occupations from the analysis. The Department now is, in addition, 
estimating those workers who fall under one of the other exemptions in 
section 13(a) of the FLSA, because such workers are exempt from both 
minimum wage and overtime pay under the relevant section and would 
remain exempt regardless of any changes to the EAP exemption. In fact, 
many of the workers estimated below as falling within one of the 
section 13(a) exemptions will already have been excluded from the 
analysis because they are paid on an hourly basis or are in a blue 
collar occupation. The methodology for identifying the workers who fall 
under the section 13(a) exemptions is explained here and is based 
generally on the methodology the Department used in 1998 when it issued 
its last report under section 4(d) of the FLSA. Section 4(d) previously 
required the Department to submit a report to Congress every two years 
regarding coverage under the FLSA.
A.2.1 Section 13(a)(1) Outside Sales Workers
    Outside sales workers are a subset of the section 13(a)(1) 
exemptions, but since they are not affected by the salary regulations 
they are not discussed in detail in the preamble. Outside sales workers 
are included in occupational category ``door-to-door sales workers, 
news and street vendors, and related workers'' (Census code 4950). This 
category is composed of workers who both would and would not qualify 
for the outside sales worker exemption; for

[[Page 38592]]

example, street vendors would not qualify. Therefore, the percentage of 
these workers that qualify for the exemption was estimated. The 
Department believes that, under the 1990 Census Codes system, outside 
sales workers were more or less uniquely identified with occupational 
category ``street & door-to-door sales workers'' (277). Therefore, the 
Department exempts the share of workers in category 4950 who under the 
old classification system would have been classified as code 277 (43 
percent).
A.2.2 Agricultural Workers
    Similar to the 2004 analysis, the Department excluded agricultural 
workers from the universe of affected employees. Agricultural workers 
were identified by occupational-industry combination. However, in the 
2004 Final Rule all workers in agricultural industries were excluded; 
here only workers also in select occupations were excluded since not 
all workers in agricultural industries qualify for the agricultural 
overtime pay exemptions. This method better approximates the true 
number of exempt agricultural workers and provides a more conservative 
estimate of the number of affected workers. Industry categories 
include: ``crop production'' (0170), ``animal production'' (0180), and 
``support activities for agriculture and forestry'' (0290). 
Occupational categories include all blue collar occupations (identified 
with the probability codes), ``farm, ranch, and other agricultural 
managers'' (0200), ``general and operations managers'' (0020), and 
``first-line supervisors/managers of farming, fishing, and forestry 
workers'' (6000).
A.2.3 Other Section 13(a) Exemptions
    The following methodology relies mainly on CPS MORG data but also 
incorporates alternative data sources when necessary.

Section 13(a)(3): Seasonal amusement and recreational establishment

    Any employee of an amusement or recreational establishment may be 
exempt from minimum wage and overtime pay if the establishment meets 
either of the following tests: (a) It operates for seven months or less 
during any calendar year, or (b) its revenue for the six lowest months 
of the year is less than one-third of the other six months of such 
year. Amusement and recreational establishments are defined as 
``establishments frequented by the public for its amusement or 
recreation,'' and ``typical examples of such are the concessionaires at 
amusement parks and beaches.'' \168\ In the CPS MORG data the 
Department identifies general amusement and recreation in the following 
industry categories:
---------------------------------------------------------------------------

    \168\ 29 CFR 779.385.

     ``independent artists, performing arts, spectator sports, 
and related industries'' (8560),
     ``museums, art galleries, historical sites, and similar 
institutions'' (8570),
     ``bowling centers'' (8580),
     ``other amusement, gambling, and recreation industries'' 
(8590), and
     ``recreational vehicle parks and camps, and rooming and 
boarding houses'' (8670).\169\
---------------------------------------------------------------------------

    \169\ The Department does not believe that all employees in this 
industry category would qualify for this exemption. However, we had 
no way to segregate in the data employees who would and would not 
qualify for exemption.
---------------------------------------------------------------------------

    The CPS MORG data does not provide information on employers' 
operating information or revenue. Using Business Employment Dynamics 
(BED) data, the Department estimated the share of leisure and 
hospitality employees working for establishments that are closed for at 
least one quarter a year.\170\ Although not technically the same as the 
FLSA definition of ``seasonal,'' this is the best available 
approximation of ``seasonal'' employees. The Department estimated that 
3 percent of amusement and recreational workers will be exempt.
---------------------------------------------------------------------------

    \170\ Seasonal employment was calculated by taking the 
difference in employment between establishment openings (all 
establishments that are either opening for the first time or 
reopening) and establishment births (establishments that are opening 
for the first time)--resulting in employment in only establishments 
reopening. Similarly, seasonal employment was estimated by taking 
the difference in employment between establishment closings and 
establishment deaths. These two estimates were then averaged. The 
analysis is limited to the leisure and hospitality industry. Since 
the exemption is limited to workers in ``establishments frequented 
by the public for its amusement or recreation'' the Department must 
assume the rate of employment in seasonal establishments, relative 
to all establishments, is equivalent across these amusement or 
recreation establishments and all leisure and hospitality 
establishments.
---------------------------------------------------------------------------

    The 1998 section 4(d) report estimated the number of exempt workers 
by applying an estimate determined in 1987 by a detailed report from 
the Employment Standards Administration. The Department chose not to 
use this estimate because it is outdated.
    Section 13(a)(3) also exempts employees of seasonal religious or 
non-profit educational centers, but many of these workers have already 
been excluded from the analysis either as religious workers (not 
covered by the FLSA) or as teachers (professional exemption) and so are 
not estimated.

Section 13(a)(5): Fishermen

    Any employee, such as a fisherman, employed in the catching, 
harvesting, or farming of fish or other aquatic life forms, is exempt 
from minimum wage and overtime pay. Fishermen are identified in 
occupational categories ``fishers and related fishing workers'' (6100) 
and ``ship and boat captains and operators'' (9310) and the industry 
category ``fishing, hunting, and trapping'' (0280). Workers identified 
in both these occupational and industry categories are considered 
exempt.

Section 13(a)(8): Small, local newspapers

    This exemption from minimum wage and overtime pay applies to any 
employee employed by a newspaper with circulation of less than 4,000 
and circulated mainly within the county where published. Newspaper 
employees are identified in the following occupational categories:
     ``news analysts, reporters and correspondents'' (2810),
     ``editors'' (2830),
     ``technical writers'' (2840),
     ``writers and authors'' (2850), and
     ``miscellaneous media and communication workers'' (2860).
    The exemption is limited to the industry category ``newspaper 
publishers'' (6470). To limit the exemption to small, local papers, the 
Department limits the exemption to employees in rural areas. Although 
employment in a rural area is not synonymous with employment at a small 
newspaper, this is the best approach currently available. 
Alternatively, the Department could use data from Dun and Bradstreet 
(D&B) as was done in the 1998 section 4(d) report. This data would 
provide information on which establishments are in rural areas; from 
this the Department could estimate the share of employment in rural 
areas. This approach would be much more time intensive but would not 
necessarily provide a better result.

Section 13(a)(10): Switchboard operators

    An independently owned public telephone company that has not more 
than 750 stations may claim the minimum wage and overtime pay exemption 
for its switchboard operators. ``Switchboard operators, including 
answering service'', are exempt under occupation code 5010 and industry 
classifications ``wired telecommunications carriers'' (6680) and 
``other telecommunications carriers'' (6690). Using the 2007 Economic 
Census, the Department

[[Page 38593]]

estimated that 0.84 percent of employees in the relevant 
telecommunication sub-industries are employed by firms with fewer than 
ten employees (the estimated level of employment necessary to service 
seven hundred and fifty stations).
    According to the 1998 section 4(d) report, fewer than 10,000 
workers were exempt in 1987 and so the Department did not develop a 
methodology for estimating the number exempt.

Section 13(a)(12): Seamen on foreign vessels

    Any employee employed as a seaman on a vessel other than an 
American vessel is exempt from minimum wage and overtime pay. Seamen 
are identified by occupational categories:
     ``sailors and marine oilers'' (9300),
     ``ship and boat captains and operators'' (9310), and
     ``ship engineers'' (9330).
    The CPS MORG data does not identify whether the vessel is foreign 
or domestic. The best approach the Department has devised is to assume 
that the number of workers in the occupation ``deep sea foreign 
transportation of freight'' (SIC 441) in 2000 is roughly equivalent to 
the number of workers on foreign vessels. The 2000 Occupational 
Employment Statistics estimates there were 14,210 workers in this 
occupation and thus that number of seamen are assigned exempt status on 
a random basis.\171\
---------------------------------------------------------------------------

    \171\ Revisions to the SIC classification system since 2000 have 
eliminated this category; thus, more recent data are not available.

---------------------------------------------------------------------------
Section 13(a)(15): Companions

    Domestic service workers employed to provide ``companionship 
services'' for an elderly person or a person with an illness, injury, 
or disability are not required to be paid the minimum wage or overtime 
pay. Companions are classified under occupational categories:
     ``nursing, psychiatric, and home health aides'' (3600) and
     ``personal and home care aides'' (4610).

And industry categories:

     ``home health care services'' (8170),
     ``individual and family services'' (8370), and
     ``private households'' (9290).

All the workers who fall within these occupational and industry 
categories were previously excluded from the analysis because they are 
paid on an hourly basis and/or are in an occupation where workers have 
no likelihood of qualifying for the section 13(a)(1) exemption.

Section 13(a)(16): Criminal investigators

    The criminal investigator must be employed by the federal 
government and paid ``availability pay.'' \172\ Criminal investigators 
are identified in occupational categories:
---------------------------------------------------------------------------

    \172\ Availability pay is compensation for hours when the agent 
must be available to perform work over and above the standard 40 
hours per week. See https://www.opm.gov/oca/pay/HTML/AP.HTM.
---------------------------------------------------------------------------

     ``detectives and criminal investigators'' (3820),
     ``fish and game wardens'' (3830), and
     ``private detectives and investigators'' (3910).
    This exemption was not mentioned in the 1998 section 4(d) report. 
The Department exempts all workers in the occupations identified above 
and employed by the federal government.

Section 13(a)(17): Computer workers

    Computer workers who meet the duties test are exempt under two 
sections of the FLSA. Salaried computer workers who earn a weekly 
salary of not less than $455 are exempt under section 13(a)(1) and 
computer workers who are paid hourly are exempt under section 13(a)(17) 
if they earn at least $27.63 an hour.
    Occupations that may be considered exempt include: ``computer and 
information systems managers'' (110), ``computer scientists and systems 
analysts'' (1000), ``computer programmers'' (1010), ``computer software 
engineers'' (1020), ``computer support specialists'' (1040), ``database 
administrators'' (1060), ``network and computer systems 
administrators'' (1100), ``network systems and data communications 
analysts'' (1110), ``computer operators'' (5800), and ``computer 
control programmers and operators'' (7900).
    To identify computer workers exempt under section 13(a)(17), we 
restrict the population to workers who are paid on an hourly basis and 
who earn at least $27.63 per hour. To determine which of these workers 
pass the computer duties test, we use the probabilities of exemption 
assigned to these occupations by the Department and assume a linear 
relationship between earnings and exemption status.
A.2.4 Section 13(b) Exemptions
Section 13(b)(1): Motor carrier employees

    This exemption eliminated overtime pay for ``any employee with 
respect to whom the Secretary of Transportation has power to establish 
qualifications and maximum hours of service pursuant to the provisions 
of Section 31502 of Title 49[.]'' In essence, these are motor carrier 
workers,\173\ identified by industry category ``truck transportation'' 
(6170).
---------------------------------------------------------------------------

    \173\ 49 U.S.C. 31502. The text of the law is available at: 
https://www.gpo.gov/fdsys/pkg/USCODE-2011-title49/html/USCODE-2011-title49-subtitleVI-partB-chap315-sec31502.htm.
---------------------------------------------------------------------------

    To be exempt, these workers must engage in ``safety affecting 
activities''. Examples of exempt occupations include: ``driver, 
driver's helper, loader, or mechanic''.\174\ The relevant occupational 
categories are:
---------------------------------------------------------------------------

    \174\ Fact Sheet #19: The Motor Carrier Exemption under the Fair 
Labor Standards Act (FLSA).
---------------------------------------------------------------------------

     ``electronic equipment installers and repairers, motor 
vehicles'' (7110),
     ``automotive service technicians and mechanics'' (7200),
     ``bus and truck mechanics and diesel engine specialists'' 
(7210),
     ``heavy vehicle and mobile equipment service technicians 
and mechanics'' (7220), and
     ``driver/sales workers and truck drivers'' (9130).\175\
---------------------------------------------------------------------------

    \175\ The 2004 methodology used 1990 Census codes 505, 507, and 
804 which crosswalk to these occupations. However, occupations 605, 
613, and 914 (included in the 1990 Census code 804 crosswalk) were 
excluded because under the new classification system they were 
deemed irrelevant.

---------------------------------------------------------------------------
Section 13(b)(2): Rail carrier employees

    Section 13(b)(2) exempts ``any employee of an employer engaged in 
the operation of a rail carrier subject to part A of subtitle IV of 
Title 49.'' \176\ This includes industrial category ``rail 
transportation'' (6080). The 1998 methodology did not include 
occupational requirements but the 2004 methodology did, so this 
restriction was included. Occupations are limited to:
---------------------------------------------------------------------------

    \176\ 49 U.S.C. 10101-11908. Text of the law is available at: 
https://www.gpo.gov/fdsys/pkg/USCODE-2013-title49/pdf/USCODE-2013-title49-subtitleIV-partA.pdf.
---------------------------------------------------------------------------

     ``locomotive engineers and operators'' (9200),
     ``railroad brake, signal, and switch operators'' (9230),
     ``railroad conductors and yardmasters'' (9240), and
     ``subway, streetcar, and other rail transportation 
workers'' (9260).

Section 13(b)(3): Air carrier employees

    This section exempts employees subject to the ``provisions of title 
II of the Railway Labor Act.'' \177\ In essence, this exempts air 
carrier employees, identified by industry category ``air 
transportation'' (6070). The 1998 methodology did not include 
occupational requirements but the 2004 methodology did, so this 
restriction was included. Occupations are limited to ``aircraft pilots 
and flight engineers''

[[Page 38594]]

(9030) and ``aircraft mechanics and service technicians'' (7140).
---------------------------------------------------------------------------

    \177\ 45 U.S.C. 181 et seq. Available at: https://www.gpo.gov/fdsys/pkg/USCODE-2013-title45/html/USCODE-2013-title45-chap8-subchapII.htm.

---------------------------------------------------------------------------
Section 13(b)(6): Seamen

    Occupational categories include ``sailors and marine oilers'' 
(9300), ``ship and boat captains and operators'' (9310), and ``ship 
engineers'' (9330).\178\ The exemption is limited to the ``water 
transportation'' industry (6090).
---------------------------------------------------------------------------

    \178\ The 2004 methodology used 1990 Census codes 828, 829, and 
833 which crosswalk to these occupations. However, occupation 952 
(dredge, excavating, and loading machine operators) was excluded 
because under the new classification system they were deemed 
irrelevant.

---------------------------------------------------------------------------
Section 13(b)(10): Salesmen, partsmen, or mechanics

    The Department limited this exemption to workers employed in a 
``nonmanufacturing establishment primarily engaged in the business of 
selling such vehicles or implements to ultimate purchasers.'' Industry 
classifications include: ``automobile dealers'' (4670) and ``other 
motor vehicle dealers'' (4680). In the 2004 Final Rule, the industry 
was limited to 1990 Census code 612 which became Census code 
``automobile dealers'' (4670). Category 4680 (``other motor vehicle 
dealers'') is also included here in keeping with the 1998 section 4(d) 
report methodology.
    The 1998 methodology did not include an occupational restriction; 
however, the 2004 methodology limited the exemption to automobiles, 
trucks, or farm implement sales workers and mechanics.

Automobiles, trucks, or farm implement sales workers include:

     ``parts salespersons'' (4750), and
     ``retail salespersons'' (4760).\179\
---------------------------------------------------------------------------

    \179\ The 2004 methodology used codes 263 and 269 which 
crosswalk to these codes plus a few others which have been deemed 
irrelevant and excluded (4700, 4740, and 4850).

---------------------------------------------------------------------------
Mechanics include:

     ``electronic equipment installers and repairers, motor 
vehicles'' (7110),
     ``automotive body and related repairers'' (7150),
     ``automotive glass installers and repairers'' (7160),
     ``automotive service technicians and mechanics'' (7200),
     ``bus and truck mechanics and diesel engine specialists'' 
(7210),
     ``heavy vehicle and mobile equipment service technicians 
and mechanics'' (7220),
     ``small engine mechanics'' (7240), and
     ``miscellaneous vehicle and mobile equipment mechanics, 
installers, and repairers'' (7260).\180\
---------------------------------------------------------------------------

    \180\ The 2004 methodology used codes 505, 506, 507, and 514 
which generally crosswalk to these codes. A few additional codes 
were added which were deemed relevant (7240 and 7260).

                Table A2--Probability Codes by Occupation
------------------------------------------------------------------------
                                                             Probability
    2002 Census code                  Occupation                 code
------------------------------------------------------------------------
10                        Chief executives.................            1
20                        General and operations managers..            1
40                        Advertising and promotions                   1
                           managers.
50                        Marketing and sales managers.....            1
60                        Public relations managers........            2
100                       Administrative services managers.            1
110                       Computer and information systems             1
                           managers.
120                       Financial managers...............            1
130                       Human resources managers.........            1
140                       Industrial production managers...            1
150                       Purchasing managers..............            1
160                       Transportation, storage, and                 1
                           distribution managers.
200                       Farm, ranch, and other                       3
                           agricultural managers.
210                       Farmers and ranchers.............            0
220                       Construction managers............            1
230                       Education administrators.........            1
300                       Engineering managers.............            1
310                       Food service managers............            3
320                       Funeral directors................            2
330                       Gaming managers..................            2
340                       Lodging managers.................            3
350                       Medical and health services                  1
                           managers.
360                       Natural sciences managers........            1
400                       Postmasters and mail                         0
                           superintendents.
410                       Property, real estate, and                   3
                           community association managers.
420                       Social and community service                 1
                           managers.
430                       Managers, all other..............            1
500                       Agents and business managers of              2
                           artists, performers, and
                           athletes.
510                       Purchasing agents and buyers,                2
                           farm products.
520                       Wholesale and retail buyers,                 2
                           except farm products.
530                       Purchasing agents, except                    2
                           wholesale, retail, and farm
                           products.
540                       Claims adjusters, appraisers,                2
                           examiners, and investigators.
560                       Compliance officers, except                  3
                           agriculture, construction,
                           health and safety, and
                           transportation.
600                       Cost estimators..................            1
620                       Human resources, training, and               2
                           labor relations specialists.
700                       Logisticians.....................            1
710                       Management analysts..............            2
720                       Meeting and convention planners..            2
730                       Other business operations                    2
                           specialists.
800                       Accountants and auditors.........            1
810                       Appraisers and assessors of real             3
                           estate.
820                       Budget analysts..................            2

[[Page 38595]]

 
830                       Credit analysts..................            2
840                       Financial analysts...............            2
850                       Personal financial advisors......            2
860                       Insurance underwriters...........            1
900                       Financial examiners..............            3
910                       Loan counselors and officers.....            2
930                       Tax examiners, collectors, and               1
                           revenue agents.
940                       Tax preparers....................            2
950                       Financial specialists, all other.            2
1000                      Computer scientists and systems              1
                           analysts.
1010                      Computer programmers.............            2
1020                      Computer software engineers......            1
1040                      Computer support specialists.....            1
1060                      Database administrators..........            1
1100                      Network and computer systems                 1
                           administrators.
1110                      Network systems and data                     1
                           communications analysts.
1200                      Actuaries........................            1
1210                      Mathematicians...................            1
1220                      Operations research analysts.....            1
1230                      Statisticians....................            1
1240                      Miscellaneous mathematical                   1
                           science occupations.
1300                      Architects, except naval.........            1
1310                      Surveyors, cartographers, and                3
                           photogrammetrists.
1320                      Aerospace engineers..............            1
1330                      Agricultural engineers...........            1
1340                      Biomedical engineers.............            1
1350                      Chemical engineers...............            1
1360                      Civil engineers..................            1
1400                      Computer hardware engineers......            1
1410                      Electrical and electronic                    1
                           engineers.
1420                      Environmental engineers..........            1
1430                      Industrial engineers, including              1
                           health and safety.
1440                      Marine engineers and naval                   1
                           architects.
1450                      Materials engineers..............            1
1460                      Mechanical engineers.............            1
1500                      Mining and geological engineers,             1
                           including mining safety
                           engineers.
1510                      Nuclear engineers................            1
1520                      Petroleum engineers..............            1
1530                      Engineers, all other.............            1
1540                      Drafters.........................            4
1550                      Engineering technicians, except              4
                           drafters.
1560                      Surveying and mapping technicians            4
1600                      Agricultural and food scientists.            1
1610                      Biological scientists............            1
1640                      Conservation scientists and                  1
                           foresters.
1650                      Medical scientists...............            1
1700                      Astronomers and physicists.......            1
1710                      Atmospheric and space scientists.            1
1720                      Chemists and materials scientists            1
1740                      Environmental scientists and                 1
                           geoscientists.
1760                      Physical scientists, all other...            3
1800                      Economists.......................            2
1810                      Market and survey researchers....            2
1820                      Psychologists....................            1
1830                      Sociologists.....................            2
1840                      Urban and regional planners......            3
1860                      Miscellaneous social scientists              2
                           and related workers.
1900                      Agricultural and food science                4
                           technicians.
1910                      Biological technicians...........            4
1920                      Chemical technicians.............            4
1930                      Geological and petroleum                     4
                           technicians.
1940                      Nuclear technicians..............            4
1960                      Other life, physical, and social             4
                           science technicians.
2000                      Counselors.......................            2
2010                      Social workers...................            3
2020                      Miscellaneous community and                  3
                           social service specialists.
2040                      Clergy...........................            0
2050                      Directors, religious activities              0
                           and education.
2060                      Religious workers, all other.....            0
2100                      Lawyers..........................            1

[[Page 38596]]

 
2110                      Judges, magistrates, and other               1
                           judicial workers.
2140                      Paralegals and legal assistants..            4
2150                      Miscellaneous legal support                  3
                           workers.
2200                      Postsecondary teachers...........            1
2300                      Preschool and kindergarten                   2
                           teachers.
2310                      Elementary and middle school                 1
                           teachers.
2320                      Secondary school teachers........            1
2330                      Special education teachers.......            1
2340                      Other teachers and instructors...            1
2400                      Archivists, curators, and museum             1
                           technicians.
2430                      Librarians.......................            1
2440                      Library Technicians..............            4
2540                      Teacher assistants...............            4
2550                      Other education, training, and               1
                           library workers.
2600                      Artists and related workers......            2
2630                      Designers........................            1
2700                      Actors...........................            1
2710                      Producers and directors..........            1
2720                      Athletes, coaches, umpires, and              2
                           related workers.
2740                      Dancers and choreographers.......            1
2750                      Musicians, singers, and related              1
                           workers.
2760                      Entertainers and performers,                 1
                           sports and related workers, all
                           other.
2800                      Announcers.......................            2
2810                      News analysts, reporters and                 3
                           correspondents.
2820                      Public relations specialists.....            3
2830                      Editors..........................            3
2840                      Technical writers................            3
2850                      Writers and authors..............            2
2860                      Miscellaneous media and                      2
                           communication workers.
2900                      Broadcast and sound engineering              4
                           technicians and radio operators.
2910                      Photographers....................            1
2920                      Television, video, and motion                2
                           picture camera operators and
                           editors.
2960                      Media and communication equipment            4
                           workers, all other.
3000                      Chiropractors....................            1
3010                      Dentists.........................            1
3030                      Dietitians and nutritionists.....            3
3040                      Optometrists.....................            1
3050                      Pharmacists......................            1
3060                      Physicians and surgeons..........            1
3110                      Physician assistants.............            2
3120                      Podiatrists......................            1
3130                      Registered nurses................            1
3140                      Audiologists.....................            2
3150                      Occupational therapists..........            3
3160                      Physical therapists..............            2
3200                      Radiation therapists.............            3
3210                      Recreational therapists..........            2
3220                      Respiratory therapists...........            3
3230                      Speech-language pathologists.....            2
3240                      Therapists, all other............            2
3250                      Veterinarians....................            1
3260                      Health diagnosing and treating               1
                           practitioners, all other.
3300                      Clinical laboratory technologists            3
                           and technicians.
3310                      Dental hygienists................            3
3320                      Diagnostic related technologists             3
                           and technicians.
3400                      Emergency medical technicians and            3
                           paramedics.
3410                      Health diagnosing and treating               4
                           practitioner support technicians.
3500                      Licensed practical and licensed              4
                           vocational nurses.
3510                      Medical records and health                   4
                           information technicians.
3520                      Opticians, dispensing............            0
3530                      Miscellaneous health                         2
                           technologists and technicians.
3540                      Other healthcare practitioners               3
                           and technical occupations.
3600                      Nursing, psychiatric, and home               0
                           health aides.
3610                      Occupational therapist assistants            0
                           and aides.
3620                      Physical therapist assistants and            0
                           aides.
3630                      Massage therapists...............            0
3640                      Dental assistants................            0
3650                      Medical assistants and other                 4
                           healthcare support occupations.
3700                      First-line supervisors/managers              2
                           of correctional officers.
3710                      First-line supervisors/managers              3
                           of police and detectives.

[[Page 38597]]

 
3720                      First-line supervisors/managers              3
                           of fire fighting and prevention
                           workers.
3730                      Supervisors, protective service              3
                           workers, all other.
3740                      Fire fighters....................            0
3750                      Fire inspectors..................            0
3800                      Bailiffs, correctional officers,             0
                           and jailers.
3820                      Detectives and criminal                      0
                           investigators.
3830                      Fish and game wardens............            0
3840                      Parking enforcement workers......            0
3850                      Police and sheriff's patrol                  0
                           officers.
3860                      Transit and railroad police......            0
3900                      Animal control workers...........            0
3910                      Private detectives and                       4
                           investigators.
3920                      Security guards and gaming                   0
                           surveillance officers.
3940                      Crossing guards..................            0
3950                      Lifeguards and other protective              0
                           service workers.
4000                      Chefs and head cooks.............            0
4010                      First-line supervisors/managers              3
                           of food preparation and serving
                           workers.
4020                      Cooks............................            0
4030                      Food preparation workers.........            0
4040                      Bartenders.......................            0
4050                      Combined food preparation and                0
                           serving workers, including fast
                           food.
4060                      Counter attendants, cafeteria,               0
                           food concession, and coffee shop.
4110                      Waiters and waitresses...........            0
4120                      Food servers, nonrestaurant......            0
4130                      Dining room and cafeteria                    0
                           attendants and bartender helpers.
4140                      Dishwashers......................            0
4150                      Hosts and hostesses, restaurant,             4
                           lounge, and coffee shop.
4160                      Food preparation and serving                 0
                           related workers, all other.
4200                      First-line supervisors/managers              4
                           of housekeeping and janitorial
                           workers.
4210                      First-line supervisors/managers              3
                           of landscaping, lawn service,
                           and groundskeeping workers.
4220                      Janitors and building cleaners...            0
4230                      Maids and housekeeping cleaners..            0
4240                      Pest control workers.............            0
4250                      Grounds maintenance workers......            0
4300                      First-line supervisors/managers              1
                           of gaming workers.
4320                      First-line supervisors/managers              4
                           of personal service workers.
4340                      Animal trainers..................            4
4350                      Nonfarm animal caretakers........            0
4400                      Gaming services workers..........            0
4410                      Motion picture projectionists....            0
4420                      Ushers, lobby attendants, and                0
                           ticket takers.
4430                      Miscellaneous entertainment                  0
                           attendants and related workers.
4460                      Funeral service workers..........            0
4500                      Barbers..........................            0
4510                      Hairdressers, hairstylists, and              0
                           cosmetologists.
4520                      Miscellaneous personal appearance            0
                           workers.
4530                      Baggage porters, bellhops, and               0
                           concierges.
4540                      Tour and travel guides...........            0
4550                      Transportation attendants........            0
4600                      Child care workers...............            0
4610                      Personal and home care aides.....            0
4620                      Recreation and fitness workers...            2
4640                      Residential advisors.............            0
4650                      Personal care and service                    0
                           workers, all other.
4700                      First-line supervisors/managers              2
                           of retail sales workers.
4710                      First-line supervisors/managers              2
                           of non-retail sales workers.
4720                      Cashiers.........................            4
4740                      Counter and rental clerks........            4
4750                      Parts salespersons...............            4
4760                      Retail salespersons..............            4
4800                      Advertising sales agents.........            2
4810                      Insurance sales agents...........            2
4820                      Securities, commodities, and                 2
                           financial services sales agents.
4830                      Travel agents....................            4
4840                      Sales representatives, services,             3
                           all other.
4850                      Sales representatives, wholesale             3
                           and manufacturing.
4900                      Models, demonstrators, and                   4
                           product promoters.
4920                      Real estate brokers and sales                3
                           agents.
4930                      Sales engineers..................            3
4940                      Telemarketers....................            4

[[Page 38598]]

 
4950                      Door-to-door sales workers, news             4
                           and street vendors, and related
                           workers.
4960                      Sales and related workers, all               3
                           other.
5000                      First-line supervisors/managers              1
                           of office and administrative
                           support workers.
5010                      Switchboard operators, including             4
                           answering service.
5020                      Telephone operators..............            4
5030                      Communications equipment                     4
                           operators, all other.
5100                      Bill and account collectors......            4
5110                      Billing and posting clerks and               4
                           machine operators.
5120                      Bookkeeping, accounting, and                 4
                           auditing clerks.
5130                      Gaming cage workers..............            4
5140                      Payroll and timekeeping clerks...            4
5150                      Procurement clerks...............            4
5160                      Tellers..........................            4
5200                      Brokerage clerks.................            4
5210                      Correspondence clerks............            4
5220                      Court, municipal, and license                4
                           clerks.
5230                      Credit authorizers, checkers, and            3
                           clerks.
5240                      Customer service representatives.            3
5250                      Eligibility interviewers,                    3
                           government programs.
5260                      File Clerks......................            4
5300                      Hotel, motel, and resort desk                4
                           clerks.
5310                      Interviewers, except eligibility             4
                           and loan.
5320                      Library assistants, clerical.....            4
5330                      Loan interviewers and clerks.....            3
5340                      New accounts clerks..............            4
5350                      Order clerks.....................            4
5360                      Human resources assistants,                  4
                           except payroll and timekeeping.
5400                      Receptionists and information                4
                           clerks.
5410                      Reservation and transportation               4
                           ticket agents and travel clerks.
5420                      Information and record clerks,               4
                           all other.
5500                      Cargo and freight agents.........            4
5510                      Couriers and messengers..........            4
5520                      Dispatchers......................            4
5530                      Meter readers, utilities.........            4
5540                      Postal service clerks............            4
5550                      Postal service mail carriers.....            4
5560                      Postal service mail sorters,                 4
                           processors, and processing
                           machine operators.
5600                      Production, planning, and                    4
                           expediting clerks.
5610                      Shipping, receiving, and traffic             4
                           clerks.
5620                      Stock clerks and order fillers...            0
5630                      Weighers, measurers, checkers,               4
                           and samplers, recordkeeping.
5700                      Secretaries and administrative               4
                           assistants.
5800                      Computer operators...............            4
5810                      Data entry keyers................            4
5820                      Word processors and typists......            4
5830                      Desktop publishers...............            4
5840                      Insurance claims and policy                  3
                           processing clerks.
5850                      Mail clerks and mail machine                 4
                           operators, except postal service.
5860                      Office clerks, general...........            4
5900                      Office machine operators, except             4
                           computer.
5910                      Proofreaders and copy markers....            4
5920                      Statistical assistants...........            4
5930                      Office and administrative support            4
                           workers, all other.
6000                      First-line supervisors/managers              4
                           of farming, fishing, and
                           forestry workers.
6010                      Agricultural inspectors..........            3
6020                      Animal breeders..................            3
6040                      Graders and sorters, agricultural            0
                           products.
6050                      Miscellaneous agricultural                   0
                           workers.
6100                      Fishers and related fishing                  0
                           workers.
6110                      Hunters and trappers.............            0
6120                      Forest and conservation workers..            0
6130                      Logging workers..................            0
6200                      First-line supervisors/managers              4
                           of construction trades and
                           extraction workers.
6210                      Boilermakers.....................            0
6220                      Brickmasons, blockmasons, and                0
                           stonemasons.
6230                      Carpenters.......................            0
6240                      Carpet, floor, and tile                      0
                           installers and finishers.
6250                      Cement masons, concrete                      0
                           finishers, and terrazzo workers.
6260                      Construction laborers............            0
6300                      Paving, surfacing, and tamping               0
                           equipment operators.

[[Page 38599]]

 
6310                      Pile-driver operators............            0
6320                      Operating engineers and other                0
                           construction equipment operators.
6330                      Drywall installers, ceiling tile             0
                           installers, and tapers.
6350                      Electricians.....................            0
6360                      Glaziers.........................            0
6400                      Insulation workers...............            0
6420                      Painters, construction and                   0
                           maintenance.
6430                      Paperhangers.....................            0
6440                      Pipelayers, plumbers,                        0
                           pipefitters, and steamfitters.
6460                      Plasterers and stucco masons.....            0
6500                      Reinforcing iron and rebar                   0
                           workers.
6510                      Roofers..........................            0
6520                      Sheet metal workers..............            0
6530                      Structural iron and steel workers            0
6600                      Helpers, construction trades.....            0
6660                      Construction and building                    3
                           inspectors.
6700                      Elevator installers and repairers            0
6710                      Fence erectors...................            0
6720                      Hazardous materials removal                  0
                           workers.
6730                      Highway maintenance workers......            0
6740                      Rail-track laying and maintenance            0
                           equipment operators.
6750                      Septic tank servicers and sewer              0
                           pipe cleaners.
6760                      Miscellaneous construction and               0
                           related workers.
6800                      Derrick, rotary drill, and                   0
                           service unit operators, oil,
                           gas, and mining.
6820                      Earth drillers, except oil and               0
                           gas.
6830                      Explosives workers, ordnance                 0
                           handling experts, and blasters.
6840                      Mining machine operators.........            0
6910                      Roof bolters, mining.............            0
6920                      Roustabouts, oil and gas.........            0
6930                      Helpers--extraction workers......            0
6940                      Other extraction workers.........            0
7000                      First-line supervisors/managers              3
                           of mechanics, installers, and
                           repairers.
7010                      Computer, automated teller, and              0
                           office machine repairers.
7020                      Radio and telecommunications                 0
                           equipment installers and
                           repairers.
7030                      Avionics technicians.............            0
7040                      Electric motor, power tool, and              0
                           related repairers.
7050                      Electrical and electronics                   0
                           installers and repairers,
                           transportation equipment.
7100                      Electrical and electronics                   0
                           repairers, industrial and
                           utility.
7110                      Electronic equipment installers              0
                           and repairers, motor vehicles.
7120                      Electronic home entertainment                0
                           equipment installers and
                           repairers.
7130                      Security and fire alarm systems              0
                           installers.
7140                      Aircraft mechanics and service               0
                           technicians.
7150                      Automotive body and related                  0
                           repairers.
7160                      Automotive glass installers and              0
                           repairers.
7200                      Automotive service technicians               0
                           and mechanics.
7210                      Bus and truck mechanics and                  0
                           diesel engine specialists.
7220                      Heavy vehicle and mobile                     0
                           equipment service technicians
                           and mechanics.
7240                      Small engine mechanics...........            0
7260                      Miscellaneous vehicle and mobile             0
                           equipment mechanics, installers,
                           and repairers.
7300                      Control and valve installers and             0
                           repairers.
7310                      Heating, air conditioning, and               0
                           refrigeration mechanics and
                           installers.
7320                      Home appliance repairers.........            0
7330                      Industrial and refractory                    0
                           machinery mechanics.
7340                      Maintenance and repair workers,              0
                           general.
7350                      Maintenance workers, machinery...            0
7360                      Millwrights......................            0
7410                      Electrical power-line installers             0
                           and repairers.
7420                      Telecommunications line                      0
                           installers and repairers.
7430                      Precision instrument and                     0
                           equipment repairers.
7510                      Coin, vending, and amusement                 0
                           machine servicers and repairers.
7520                      Commercial divers................            4
7540                      Locksmiths and safe repairers....            0
7550                      Manufactured building and mobile             0
                           home installers.
7560                      Riggers..........................            0
7600                      Signal and track switch repairers            0
7610                      Helpers--installation,                       0
                           maintenance, and repair workers.
7620                      Other installation, maintenance,             0
                           and repair workers.
7700                      First-line supervisors/managers              3
                           of production and operating
                           workers.
7710                      Aircraft structure, surfaces,                0
                           rigging, and systems assemblers.
7720                      Electrical, electronics, and                 0
                           electromechanical assemblers.

[[Page 38600]]

 
7730                      Engine and other machine                     0
                           assemblers.
7740                      Structural metal fabricators and             0
                           fitters.
7750                      Miscellaneous assemblers and                 0
                           fabricators.
7800                      Bakers...........................            0
7810                      Butchers and other meat, poultry,            0
                           and fish processing workers.
7830                      Food and tobacco roasting,                   0
                           baking, and drying machine
                           operators and tenders.
7840                      Food batchmakers.................            0
7850                      Food cooking machine operators               0
                           and tenders.
7900                      Computer control programmers and             4
                           operators.
7920                      Extruding and drawing machine                0
                           setters, operators, and tenders,
                           metal and plastic.
7930                      Forging machine setters,                     0
                           operators, and tenders, metal
                           and plastic.
7940                      Rolling machine setters,                     0
                           operators, and tenders, metal
                           and plastic.
7950                      Cutting, punching, and press                 0
                           machine setters, operators, and
                           tenders, metal and plastic.
7960                      Drilling and boring machine tool             0
                           setters, operators, and tenders,
                           metal and plastic.
8000                      Grinding, lapping, polishing, and            0
                           buffing machine tool setters,
                           operators, and tenders, metal
                           and plastic.
8010                      Lathe and turning machine tool               0
                           setters, operators, and tenders,
                           metal and plastic.
8020                      Milling and planing machine                  0
                           setters, operators, and tenders,
                           metal and plastic.
8030                      Machinists.......................            0
8040                      Metal furnace and kiln operators             0
                           and tenders.
8060                      Model makers and patternmakers,              0
                           metal and plastic.
8100                      Molders and molding machine                  0
                           setters, operators, and tenders,
                           metal and plastic.
8120                      Multiple machine tool setters,               0
                           operators, and tenders, metal
                           and plastic.
8130                      Tool and die makers..............            0
8140                      Welding, soldering, and brazing              0
                           workers.
8150                      Heat treating equipment setters,             0
                           operators, and tenders, metal
                           and plastic.
8160                      Lay-out workers, metal and                   0
                           plastic.
8200                      Plating and coating machine                  0
                           setters, operators, and tenders,
                           metal and plastic.
8210                      Tool grinders, filers, and                   0
                           sharpeners.
8220                      Metalworkers and plastic workers,            0
                           all other.
8230                      Bookbinders and bindery workers..            0
8240                      Job printers.....................            0
8250                      Prepress technicians and workers.            0
8260                      Printing machine operators.......            0
8300                      Laundry and dry-cleaning workers.            0
8310                      Pressers, textile, garment, and              0
                           related materials.
8320                      Sewing machine operators.........            0
8330                      Shoe and leather workers and                 0
                           repairers.
8340                      Shoe machine operators and                   0
                           tenders.
8350                      Tailors, dressmakers, and sewers.            0
8360                      Textile bleaching and dyeing                 0
                           machine operators and tenders.
8400                      Textile cutting machine setters,             0
                           operators, and tenders.
8410                      Textile knitting and weaving                 0
                           machine setters, operators, and
                           tenders.
8420                      Textile winding, twisting, and               0
                           drawing out machine setters,
                           operators, and tenders.
8430                      Extruding and forming machine                0
                           setters, operators, and tenders,
                           synthetic and glass fibers.
8440                      Fabric and apparel patternmakers.            0
8450                      Upholsterers.....................            0
8460                      Textile, apparel, and furnishings            0
                           workers, all other.
8500                      Cabinetmakers and bench                      0
                           carpenters.
8510                      Furniture finishers..............            0
8520                      Model makers and patternmakers,              0
                           wood.
8530                      Sawing machine setters,                      0
                           operators, and tenders, wood.
8540                      Woodworking machine setters,                 0
                           operators, and tenders, except
                           sawing.
8550                      Woodworkers, all other...........            0
8600                      Power plant operators,                       0
                           distributors, and dispatchers.
8610                      Stationary engineers and boiler              0
                           operators.
8620                      Water and liquid waste treatment             0
                           plant and system operators.
8630                      Miscellaneous plant and system               0
                           operators.
8640                      Chemical processing machine                  0
                           setters, operators, and tenders.
8650                      Crushing, grinding, polishing,               0
                           mixing, and blending workers.
8710                      Cutting workers..................            0
8720                      Extruding, forming, pressing, and            0
                           compacting machine setters,
                           operators, and tenders.
8730                      Furnace, kiln, oven, drier, and              0
                           kettle operators and tenders.
8740                      Inspectors, testers, sorters,                0
                           samplers, and weighers.
8750                      Jewelers and precious stone and              0
                           metal workers.
8760                      Medical, dental, and ophthalmic              0
                           laboratory technicians.
8800                      Packaging and filling machine                0
                           operators and tenders.
8810                      Painting workers.................            0
8830                      Photographic process workers and             0
                           processing machine operators.
8840                      Semiconductor processors.........            0
8850                      Cementing and gluing machine                 0
                           operators and tenders.

[[Page 38601]]

 
8860                      Cleaning, washing, and metal                 0
                           pickling equipment operators and
                           tenders.
8900                      Cooling and freezing equipment               0
                           operators and tenders.
8910                      Etchers and engravers............            0
8920                      Molders, shapers, and casters,               0
                           except metal and plastic.
8930                      Paper goods machine setters,                 0
                           operators, and tenders.
8940                      Tire builders....................            0
8950                      Helpers--production workers......            0
8960                      Production workers, all other....            0
9000                      Supervisors, transportation and              3
                           material moving workers.
9030                      Aircraft pilots and flight                   4
                           engineers.
9040                      Air traffic controllers and                  3
                           airfield operations specialists.
9110                      Ambulance drivers and attendants,            0
                           except emergency medical
                           technicians.
9120                      Bus drivers......................            0
9130                      Driver/sales workers and truck               0
                           drivers.
9140                      Taxi drivers and chauffeurs......            0
9150                      Motor vehicle operators, all                 0
                           other.
9200                      Locomotive engineers and                     0
                           operators.
9230                      Railroad brake, signal, and                  0
                           switch operators.
9240                      Railroad conductors and                      0
                           yardmasters.
9260                      Subway, streetcar, and other rail            0
                           transportation workers.
9300                      Sailors and marine oilers........            0
9310                      Ship and boat captains and                   0
                           operators.
9330                      Ship engineers...................            4
9340                      Bridge and lock tenders..........            0
9350                      Parking lot attendants...........            0
9360                      Service station attendants.......            0
9410                      Transportation inspectors........            0
9420                      Other transportation workers.....            0
9500                      Conveyor operators and tenders...            0
9510                      Crane and tower operators........            0
9520                      Dredge, excavating, and loading              0
                           machine operators.
9560                      Hoist and winch operators........            0
9600                      Industrial truck and tractor                 0
                           operators.
9610                      Cleaners of vehicles and                     0
                           equipment.
9620                      Laborers and freight, stock, and             0
                           material movers, hand.
9630                      Machine feeders and offbearers...            0
9640                      Packers and packagers, hand......            0
9650                      Pumping station operators........            0
9720                      Refuse and recyclable material               0
                           collectors.
9730                      Shuttle car operators............            0
9740                      Tank car, truck, and ship loaders            0
9750                      Material moving workers, all                 0
                           other.
------------------------------------------------------------------------

Appendix B. Additional Tables

   Table B1--EAP Exempt Workers Potentially Affected by This Proposed
                      Rulemaking, by Industry, 2013
------------------------------------------------------------------------
                                       Potentially       As percent of
                                       affected EAP       potentially
             Industry                    workers          affected EAP
                                        (millions)     workers (percent)
------------------------------------------------------------------------
Total \a\.........................               21.4              100.0
Agriculture.......................                0.0                0.1
Forestry, logging, fishing,                       0.0                0.0
 hunting, and trapping............
Mining............................                0.2                0.8
Construction......................                0.8                3.6
Nonmetallic mineral product                       0.1                0.3
 manufacturing....................
Primary metals and fabricated                     0.2                1.0
 metal products...................
Machinery manufacturing...........                0.3                1.4
Computer and electronic product                   0.6                2.9
 manufacturing....................
Electrical equipment, appliance                   0.1                0.5
 manufacturing....................
Transportation equipment                          0.6                2.6
 manufacturing....................
Wood products.....................                0.0                0.2
Furniture and fixtures                            0.1                0.2
 manufacturing....................
Miscellaneous and not specified                   0.3                1.5
 manufacturing....................
Food manufacturing................                0.2                0.8
Beverage and tobacco products.....                0.1                0.3

[[Page 38602]]

 
Textile, apparel, and leather                     0.1                0.3
 manufacturing....................
Paper and printing................                0.1                0.6
Petroleum and coal products                       0.1                0.3
 manufacturing....................
Chemical manufacturing............                0.4                2.0
Plastics and rubber products......                0.1                0.3
Wholesale trade...................                0.8                3.9
Retail trade......................                1.6                7.5
Transportation and warehousing....                0.5                2.4
Utilities.........................                0.3                1.3
Publishing industries (except                     0.2                0.9
 internet)........................
Motion picture and sound recording                0.0                0.2
 industries.......................
Broadcasting (except internet)....                0.2                0.8
Internet publishing and                           0.0                0.2
 broadcasting.....................
Telecommunications................                0.4                1.6
Internet service providers and                    0.0                0.2
 data processing services.........
Other information services........                0.1                0.3
Finance...........................                1.9                9.0
Insurance.........................                1.0                4.7
Real estate.......................                0.3                1.4
Rental and leasing services.......                0.1                0.3
Professional and technical                        3.6               16.8
 services.........................
Management of companies and                       0.1                0.3
 enterprises......................
Administrative and support                        0.5                2.3
 services.........................
Waste management and remediation                  0.0                0.2
 services.........................
Educational services..............                0.8                3.9
Hospitals.........................                1.0                4.7
Health care services, except                      1.2                5.5
 hospitals........................
Social assistance.................                0.4                1.8
Arts, entertainment, and                          0.4                1.7
 recreation.......................
Accommodation.....................                0.1                0.5
Food services and drinking places.                0.3                1.2
Repair and maintenance............                0.1                0.5
Personal and laundry services.....                0.1                0.3
Membership associations and                       0.4                1.8
 organizations....................
Private households................                0.0                0.0
Public administration.............                0.8                3.9
------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Columns may not sum to total due to rounding.

VIII. Initial Regulatory Flexibility Analysis (IRFA)

    The Regulatory Flexibility Act of 1980 (RFA) as amended by the 
Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA), 
hereafter jointly referred to as the RFA, requires agencies to prepare 
regulatory flexibility analyses and make them available for public 
comment, when proposing regulations that will have a significant 
economic impact on a substantial number of small entities. See 5 U.S.C. 
603. If the rule is not expected to have a significant economic impact 
on a substantial number of small entities, the RFA allows an agency to 
certify such, in lieu of preparing an analysis. See 5 U.S.C. 605.
    The Department specifically invites comment on the impacts of the 
proposed rule on small businesses, including whether alternatives exist 
that will reduce burden on small entities while still meeting the 
objectives of the FLSA. The Chief Counsel for Advocacy of the Small 
Business Administration (SBA) was notified of a draft of this rule upon 
submission of the rule to OMB under E.O. 12866.

A. Reasons Why Action by the Agency Is Being Considered

    The EAP exemption salary level test that is the focus of this 
proposed rulemaking has been updated seven times since the FLSA was 
originally enacted in 1938. These updates were necessary in order for 
the required salary level to keep pace with increases in earnings in 
the economy so that it could continue to serve as an effective bright-
line test that separates workers who Congress intended to remain 
entitled to minimum wage and overtime protection and those who may 
qualify as bona fide EAP exempt workers.
    The standard salary level and HCE total compensation levels have 
not been updated since 2004 and, as described in detail in section 
VII.A.ii., the standard salary level has declined considerably in real 
terms relative to both its 2004 and 1975 values. As a result, the 
exemption removes workers from overtime protection who were not 
intended to be within the exemption. Similarly, the HCE annual 
compensation requirement is out of date; more than twice as many 
workers earn at least $100,000 annually compared to when it was adopted 
in 2004. Therefore, the Department believes that rulemaking is 
necessary in order to restore the effectiveness of these levels.

B. Statement of Objectives and Legal Basis for the Proposed Rule

    Section 13(a)(1) creates a minimum wage and overtime pay exemption 
for

[[Page 38603]]

bona fide executive, administrative, professional, outside sales 
employees, and teachers and academic administrative personnel, as those 
terms are defined and delimited by the Secretary of Labor. The 
regulations in part 541 contain specific criteria that define each 
category of exemption. The regulations also define those computer 
employees who are exempt under section 13(a)(1) and section 13(a)(17). 
To qualify for exemption, employees must meet certain tests regarding 
their job duties and generally must be paid on a salary basis at not 
less than $455 per week.
    The Department's primary objective in this rulemaking is to ensure 
that the revised salary levels will continue to provide a useful and 
effective test for exemption. The salary levels were designed to 
operate as a ready guide to assist employers in deciding which 
employees were more likely to meet the duties tests for the exemptions. 
If left unchanged, however, the effectiveness of the salary level test 
as a means of determining exempt status diminishes as nonexempt 
employee wages increase over time.
    The Department last updated the salary levels in the 2004 Final 
Rule, setting the standard test threshold at $455 per week for EAP 
employees. The 2004 Final Rule also created a new ``highly 
compensated'' test for exemption. Under the HCE exemption, employees 
who are paid total annual compensation of at least $100,000 (which must 
include at least $455 per week paid on a salary or fee basis) are 
exempt from the FLSA's overtime requirements if they customarily and 
regularly perform at least one of the duties or responsibilities of an 
exempt EAP employee identified in the standard tests for exemption. 
Sec.  541.601.
    Employees who meet the requirements of part 541 are excluded from 
the Act's minimum wage and overtime pay protections. As a result, 
employees may work any number of hours in the workweek and not be 
subject to the FLSA's overtime pay requirements. Some State laws have 
stricter exemption standards than those described above. The FLSA does 
not preempt any such stricter State standards. If a State law 
establishes a higher standard than the provisions of the FLSA, the 
higher standard applies in that specific state. See 29 U.S.C. 218.
    In order to restore the ability of the standard salary level and 
the HCE compensation requirement to serve as appropriate bright-line 
tests between overtime-protected employees and employees who may be EAP 
exempt, the Department proposes to increase the minimum salary level 
test from $455 to the 40th percentile of the weekly wages of all full-
time salaried employees ($921 per week), and the level for the HCE test 
from $100,000 to the annual equivalent of the 90th percentile of weekly 
earnings for full-time salaried employees ($122,148 in annual 
earnings). The Department reached the proposed salary levels after 
considering available data on actual salary levels currently being paid 
in the economy. In order to ensure that these levels continue to 
function appropriately in the future, the Department also proposes to 
automatically update them annually either by maintaining the respective 
earnings percentile or updating the levels based on changes in the CPI-
U.

C. Description of the Number of Small Entities To Which the Proposed 
Rule Will Apply

i. Definition of Small Entity
    The RFA defines a ``small entity'' as a (1) small not-for-profit 
organization, (2) small governmental jurisdiction, or (3) small 
business. The Department used the entity size standards defined by SBA 
to classify entities as small for the purpose of this analysis. SBA 
establishes separate standards for individual 6-digit NAICS industry 
codes, and standard cutoffs are typically based on either the average 
annual number of employees, or the average annual receipts. For 
example, the SBA has two widely used size standards: 500 employees for 
manufacturing, and $7 million in annual receipts for nonmanufacturing 
services. However, some exceptions do exist, the most notable being 
that depository institutions (including credit unions, commercial 
banks, and non-commercial banks) are classified by total assets. Small 
governmental jurisdictions are another noteworthy exception; they are 
defined as the governments of cities, counties, towns, townships, 
villages, school districts, or special districts with population of 
less than 50,000 people.\181\
---------------------------------------------------------------------------

    \181\ See https://www.sba.gov/advocacy/regulatory-flexibility-act 
for details.
---------------------------------------------------------------------------

ii. Data Sources and Methods
    The Department obtained data from several different sources to 
determine employment in small entities for each industry. Categorical 
tabulations from the Statistics of U.S. Businesses (SUSB, 2007 and 
2011) were used for most industries. Industries that used data from 
alternative sources include Credit Unions (National Credit Union 
Association, 2010), Commercial and Non-Commercial Banks (Federal 
Depository Insurance Corporation, 2013), and Public Administration, 
where employees were classified based on employment estimates from the 
Census of Governments (2012), and local population estimates from the 
Census of Population and Housing (2012). The Department used the latest 
available data in each case, so data years differ between sources.\182\
---------------------------------------------------------------------------

    \182\ Latest available year of data for each source in 
parentheses. SUSB employment data are for 2011 (although since the 
time of writing 2012 data have become available) and receipts data 
are for 2007.
---------------------------------------------------------------------------

    For each industry, the total number of employees is organized in 
categories based on different characteristics of the employing entity. 
The categories are defined using employment, annual revenue, and 
assets. The Department combined these categories with the corresponding 
SBA standards to estimate the proportion of workers in each industry 
who are employed by a small entity.
    The general methodological approach was to classify all employees 
in categories below the SBA cutoff as in ``small entity'' employment. 
If a cutoff fell in the middle of a defined category, a uniform 
distribution of employees across that bracket was assumed in order to 
determine what proportion should be classified as in small entity 
employment. The Department assumed that the small entity distribution 
across revenue categories for Other Depository Institutions, which was 
not separately represented in FDIC asset data, was similar to that of 
Credit Unions.
iii. Number of Small Entities Impacted by Proposed Rule
    It is difficult to estimate precisely the number of small entities 
that will be impacted by the proposed rule. The employee, payroll, and 
receipts data in SUSB are tabulated by ``enterprise size,'' where the 
definition of ``enterprise'' is equivalent to ``entity'' for the 
purposes of the current discussion. However, this data does not 
directly report the number of enterprises, but instead provides data on 
``establishments'' (individual plants, regardless of ownership), and 
``firms'' (a collection of all plants with a single owner within a 
given state and industry). Therefore, an enterprise may consist of 
multiple firms, depending on the number of states and industries it 
operates in. Using the SUSB number of small firms as a proxy may thus 
overestimate the number of small entities nationally. However, this 
effect is unlikely to be large, because most small entities would 
probably operate on smaller scales (i.e., will either consist

[[Page 38604]]

of a single establishment, or operate within a single state and 
industry).
    The estimated probability that an EAP exempt worker is employed by 
a small entity is set equal to the calculated proportion of workers 
employed in the corresponding industry. For example, if an industry has 
50 percent of workers employed in small entities, then on average one 
out of every two EAP exempt workers in this industry is expected to be 
small-entity employed. The Department applied these probabilities to 
the population of EAP exempt workers in order to find the number of 
workers (total and affected by the rule) employed by small entities, 
their payroll under the current and the proposed salary levels, and the 
number of small entities employing affected workers. The Department 
also tabulated the total number of affected entities and employees by 
industry group.
    With these limitations, the Department estimates that the proposed 
rule will affect 4.7 million workers in an estimated 290,800 
establishments (Table 33).\183\ Among affected workers, 1.8 million 
were estimated to be employed by small entities, working in 211,000 
small establishments (Table 34). While nearly 40 percent of affected 
EAP workers are employed in small entities, this composes a very small 
percentage of overall small entity employment in the economy; affected 
workers account for 3.5 percent of small establishment employment on 
average, with at most 7.0 percent of workers affected in any industry. 
The industries with the most affected small entity employees are:
---------------------------------------------------------------------------

    \183\ To estimate the number of establishments the ratio of 
affected workers to total workers was applied to the total number of 
establishments. For example, 4.7 million of the total 132 million 
workers are affected, or 3.5 percent; 3.5 percent of the total 7.4 
million establishments is 290,000 establishments with affected 
workers.
---------------------------------------------------------------------------

     Education and health services with 336,800 affected 
workers (3.5 percent of employees) in 26,800 establishments;
     Professional and business services with 319,200 affected 
workers (5.0 percent of employees) in 56,100 establishments; and
     Wholesale and retail trade with 241,700 affected workers 
(3.7 percent of employees) in 38,000 establishments.
    The financial activities industry has the largest percent of 
affected small entity employees; 7 percent are affected.

         Table 33--Affected Entities Under Proposed Standard Salary and HCE Compensation Level Increases
----------------------------------------------------------------------------------------------------------------
                                                Establishments (1,000s)      Workers (1,000s) \a\       Annual
                   Industry                   ------------------------------------------------------   payroll
                                                  Total     Affected \b\     Total       Affected     (billions)
----------------------------------------------------------------------------------------------------------------
Total........................................        7,427         290.8      132,084       4,682.4       $5,881
Agriculture, forestry, fishing, & hunting....           21           0.1        1,150           7.1           35
Mining.......................................           28           0.6          931          20.4           61
Construction.................................          658          15.1        6,804         155.7          314
Manufacturing................................          296           8.1       14,844         406.1          759
Wholesale & retail trade.....................        1,475          52.1       18,733         662.1          657
Transportation & utilities...................          229           5.2        6,911         156.7          334
Information..................................          134           8.3        2,969         183.0          164
Financial activities.........................          809          61.4        9,009         683.3          499
Professional & business services.............        1,281          69.2       13,573         733.0          734
Education & health services..................          910          28.1       32,120         992.4        1,427
Leisure & hospitality........................          772          16.3       12,166         256.7          303
Other services...............................          722          23.2        5,699         183.2          193
Public administration \c\....................           90           3.0        7,175         242.7          399
----------------------------------------------------------------------------------------------------------------
Note: Establishment data from the Survey of U.S. Businesses 2011; worker data from CPS MORG using pooled data
  for 2011-2013.
\a\ Excludes the self-employed and unpaid workers. Affected workers are those who would become overtime eligible
  under the proposed increased salary levels if weekly earnings did not change.
\b\ The number of affected establishments depends on assumptions made by the Department. The numbers presented
  here assume the share of establishments that are affected is equal to the share of workers who are affected
  within an industry.
\c\ Establishment number represents the total number of governments, including state and local. Data from
  Government Organization Summary Report: 2012.


     Table 34--Affected Small Entities and Workers Under Proposed Standard Salary and HCE Compensation Level
                                                    Increases
----------------------------------------------------------------------------------------------------------------
                                                      Small entity           Small entity workers       Annual
                                                establishments (1,000s)          (1,000s) \a\           small
                   Industry                   ------------------------------------------------------    entity
                                                                                                       payroll
                                                  Total     Affected \b\     Total     Affected \c\   (billions)
----------------------------------------------------------------------------------------------------------------
Total \d\....................................        6,045         210.6       50,355       1,754.0       $2,110
Agriculture, forestry, fishing, & hunting....           20           0.1          624           3.9           18
Mining.......................................           23           0.6          351           9.8           23
Construction.................................          640          14.3        4,373          97.8          201
Manufacturing................................          265           7.2        6,372         172.6          309
Wholesale & retail trade.....................        1,038          38.0        6,600         241.7          251
Transportation & utilities...................          178           4.1        1,711          39.7           76
Information..................................           73           4.6          768          48.6           40
Financial activities.........................          550          38.7        2,812         198.2          147
Professional & business services.............        1,121          56.1        6,374         319.2          339
Education & health services..................          763          26.8        9,573         336.8          382
Leisure & hospitality........................          632          13.0        6,380         131.6          155
Other services...............................          668          23.4        3,724         130.2          134
Public administration \e\....................           73           2.5          692          23.9           34
----------------------------------------------------------------------------------------------------------------
Note: Establishment data from the Survey of U.S. Businesses 2011; worker data from CPS MORG using pooled data
  for 2011-2013.

[[Page 38605]]

 
\a\ Excludes the self-employed and unpaid workers. Affected workers are those who would become overtime eligible
  under the proposed increased salary levels if weekly earnings did not change.
\b\ The number of affected establishments depends on assumptions made by the Department. The numbers presented
  here assume the share of workers in small entities who are affected is also the share of small entity
  establishments that are affected.
\c\ These numbers are also equal to the number of small entity establishments under the assumption that each
  affected establishment has one affected worker.
\d\ The components do not necessarily equal the totals due to when averages are taken.
\e\ Establishment number represents the total number of governments, including state and local.

D. Projected Reporting, Recordkeeping, and Other Compliance 
Requirements of the Proposed Rule

    The FLSA sets minimum wage, overtime pay, and recordkeeping 
requirements for employment subject to its provisions. Unless exempt, 
covered employees must be paid at least the minimum wage and not less 
than one and one-half times their regular rates of pay for overtime 
hours worked.
    Every covered employer must keep certain records for each nonexempt 
worker. The regulations at part 516 require employers to maintain 
records for employees subject to the minimum wage and overtime pay 
provisions of the FLSA. Thus, the recordkeeping requirements are not 
new requirements; however, employers would need to keep some additional 
records for additional affected employees if the NPRM were to be made 
final without change. As indicated in this analysis, the NPRM would 
expand minimum wage and overtime pay coverage to 4.6 million affected 
EAP workers (including HCE workers and excluding Type 4 workers who 
remain exempt). This would result in an increase in employer burden and 
was estimated in the PRA portion (section VI.) of this NPRM. Note that 
the burdens reported for the PRA section of this NPRM include the 
entire information collection and not merely the additional burden 
estimated as a result of this NPRM.
i. Costs to Small Entities
    As detailed in section VII.D.iv., three direct costs to employers 
are quantified in this analysis: (1) Regulatory familiarization costs; 
(2) adjustment costs; and (3) managerial costs. Regulatory 
familiarization costs are the costs incurred to read and become 
familiar with the requirements of the rule. Adjustment costs are the 
costs accrued to determine workers' new exemption statuses, notify 
employees of policy changes, and update payroll systems. Managerial 
costs associated with this proposed rulemaking occur because hours of 
workers who are newly entitled to overtime may be more closely 
scheduled and monitored to minimize or avoid paying the overtime 
premium. Regardless of business size, the Department estimates that 
each establishment will spend one hour of time for regulatory 
familiarization; one hour per each affected worker in adjustment costs; 
and five minutes per week scheduling and monitoring each affected 
worker expected to be reclassified as overtime eligible as a result of 
this proposed rule.
    For small entities, the Department projected annual regulatory 
familiarization, adjustment, and managerial costs, and payments to 
employees in terms of extra wages paid. The Department believes that 
the minimum and maximum per-establishment costs are the most accurate 
possible estimates for the range of impact of the proposed rule on 
individual employers.
    As a direct result of this proposed rule, the Department expects 
total direct employer costs (regulatory familiarization, adjustment, 
and managerial costs) of $134.5 to 186.6 million will be incurred by 
small entities in the first year after the promulgation of the proposed 
rule (Table 35). The three industries with the most affected small 
entity employees (educational and health services, professional and 
business services, and wholesale and retail trade) account for more 
than 50 percent of direct costs.
    Average weekly earnings for affected EAP workers in small entities 
are expected to increase by $6.16 per week per affected worker due to 
both the standard salary level and HCE total annual compensation level 
proposed increases. This results in costs to employer of $561.5 million 
in wage increases to employees, which compose 0.1 to 0.8 percent of 
aggregate affected entity payroll (Table 36).
    The Department evaluated the impacts to small entities employing 
affected workers using a range to represent minimum and maximum costs 
incurred by an average establishment. To define the average 
establishment, the Department divided the total number of employees and 
payroll among small establishments by the total number of small 
establishments on an industry-specific basis. The minimum level of 
impacts is defined by assuming only one worker employed by the average 
establishment is affected by the revised salary level. The maximum 
level is defined by assuming 100 percent of workers employed by the 
average establishment are affected by the revised salary level.\184\ On 
average, depending on the number of affected workers it employs, an 
affected establishment is expected to incur $100 to $600 in direct 
costs and $320 to $2,700 in additional payroll to employees in the 
first year after the promulgation of the proposed rule. On average, 
these combined first year costs and transfers account for approximately 
0.11 to 0.95 percent of average establishment payroll (depending on how 
affected small establishments are defined).
---------------------------------------------------------------------------

    \184\ Larger than average small establishments in each industry 
might employ a larger number of affected employees, and such 
establishments might incur larger costs and transfers than the 
``average'' establishment used as a benchmark in this analysis. 
However, although such establishments' costs and transfers will 
increase in proportion to the number of affected workers, these 
establishments' payroll will also increase in approximate proportion 
to the number of workers they employ. Since such establishments can 
never have more than 100 percent of their employees affected by the 
proposed rule, the rule's impact as measured by costs and transfers 
as a percentage of establishment payroll will be roughly the same 
magnitude as an average establishment with 100 percent of employees 
affected. Thus, the scalability of the average establishment impacts 
adequately captures impacts to establishments both larger and 
smaller than average.

[[Page 38606]]



      Table 35--Costs to Small Entities Under Proposed Standard Salary and HCE Compensation Level Increases
----------------------------------------------------------------------------------------------------------------
                                                        Cost to small entities in year 1 \a\
                                   -----------------------------------------------------------------------------
                                        Total (millions)            Per affected            Percent of annual
             Industry              --------------------------  establishment (1,000s)            payroll
                                                             ---------------------------------------------------
                                      Min \b\      Max \b\      Min \b\      Max \b\      Min \b\      Max \b\
----------------------------------------------------------------------------------------------------------------
Total.............................       $186.6       $134.5         $0.1         $0.6        $0.03       $0.18%
Agriculture, forestry, fishing,             0.4          0.3          0.1          2.2         0.01         0.25
 and hunting......................
Mining............................          1.0          0.7          0.1          1.2         0.01         0.12
Construction......................         10.4          7.6          0.1          0.5         0.03         0.17
Manufacturing.....................         18.4         12.7          0.1          1.8         0.01         0.15
Wholesale and retail trade........         25.7         18.8          0.1          0.5         0.04         0.20
Transportation and utilities......          4.2          3.0          0.1          0.7         0.03         0.17
Information.......................          5.2          3.7          0.1          0.8         0.02         0.14
Financial activities..............         21.1         15.6          0.1          0.4         0.04         0.15
Professional and business services         34.0         25.0          0.1          0.4         0.04         0.15
Educational and health services...         35.8         25.2          0.1          0.9         0.02         0.19
Leisure and hospitality...........         14.0         10.0          0.1          0.8         0.04         0.31
Other services....................         13.9         10.2          0.1          0.4         0.05         0.22
Public administration.............          2.5          1.8          0.1          0.7         0.02         0.15
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Direct costs include regulatory familiarization, adjustment, and managerial costs.
\b\ The range of costs per establishment depends on the number of affected establishments. The minimum assumes
  that each affected establishment has one affected worker (therefore, the number of affected establishments is
  equal to the number of affected workers). The maximum assumes the share of workers in small entities who are
  affected is also the share of small entity establishments that are affected.


   Table 36--Transfers for Small Entities under Proposed Standard Salary and HCE Compensation Level Increases
----------------------------------------------------------------------------------------------------------------
                                                            Transfers for small entities in year 1 \a\
                                                ----------------------------------------------------------------
                                                                    Per affected            Percent of annual
                    Industry                        Total      establishment (1,000s)            payroll
                                                  (millions) ---------------------------------------------------
                                                                Min \b\      Max \b\      Min \b\      Max \b\
----------------------------------------------------------------------------------------------------------------
Total..........................................       $561.5        $0.32         $2.7        $0.09       $0.76%
Agriculture, forestry, fishing, and hunting....          0.5         0.12          3.8         0.01         0.42
Mining.........................................          3.1         0.31          4.9         0.03         0.49
Construction...................................         54.4         0.56          3.8         0.18         1.21
Manufacturing..................................         53.5         0.31          7.4         0.03         0.64
Wholesale and retail trade.....................        101.4         0.42          2.7         0.17         1.10
Transportation and utilities...................         10.2         0.26          2.5         0.06         0.58
Information....................................         19.9         0.41          4.3         0.07         0.78
Financial activities...........................         53.1         0.27          1.4         0.10         0.51
Professional and business services.............         84.2         0.26          1.5         0.09         0.50
Educational and health services................         75.1         0.22          2.8         0.04         0.56
Leisure and hospitality........................         70.0         0.53          5.4         0.22         2.19
Other services.................................         31.4         0.24          1.3         0.12         0.67
Public administration..........................          4.7         0.20          1.9         0.04         0.39
----------------------------------------------------------------------------------------------------------------
Note: Pooled data for 2011-2013.
\a\ Aggregate change in total annual payroll experienced by small entities under the proposed salary levels
  after labor market adjustments. This amount represents the total amount of (wage) transfers from employers to
  employees.
\b\ The range of transfers per establishment depends on the number of affected establishments (the denominator).
  The minimum assumes that each affected establishment has one affected worker (therefore, the number of
  affected establishments is equal to the number of affected workers). The maximum assumes the share of workers
  in small entities who are affected is also the share of small entity establishments that are affected.

ii. Differing Compliance and Reporting Requirements for Small Entities
    This NPRM provides no differing compliance requirements and 
reporting requirements for small entities. The Department has strived 
to minimize respondent recordkeeping burden by requiring no specific 
form or order of records under the FLSA and its corresponding 
regulations. Moreover, employers would normally maintain the records 
under usual or customary business practices.
iii. Least Burdensome Option or Explanation Required
    The Department believes it has chosen the most effective option 
that updates and clarifies the rule and which results in the least 
burden. Among the options considered by the Department, the least 
restrictive option was taking no regulatory action and the most 
restrictive was updating the 1975 short test salary level for inflation 
based upon the CPI-U (which would result in a standard salary level of 
$1,083 per week). Taking no regulatory action does not address the 
Department's concerns discussed above under Need for Regulation. The 
Department found the most restrictive option to be overly burdensome on 
business in general and

[[Page 38607]]

specifically small business, and high in light of the fact that there 
no longer is a long duties test with an associated lower salary level 
that employers may use to establish that employees are exempt.
    Pursuant to section 603(c) of the RFA, the following alternatives 
are to be addressed:
    i. Differing compliance or reporting requirements that take into 
account the resources available to small entities. The FLSA creates a 
level playing field for businesses by setting a floor below which 
employers may not pay their employees. To establish differing 
compliance or reporting requirements for small businesses would 
undermine this important purpose of the FLSA and appears to not be 
necessary given the small annualized cost of the rule. The Year 1 cost 
of the proposed rule for the average employer that qualifies as small 
was estimated to range from a minimum of $400 to a maximum of $3,300. 
The Department makes available a variety of resources to employers for 
understanding their obligations and achieving compliance. Therefore the 
Department has not proposed differing compliance or reporting 
requirements for small businesses.
    ii. The clarification, consolidation, or simplification of 
compliance and reporting requirements for small entities. The proposed 
rule imposes no new reporting requirements. The Department makes 
available a variety of resources to employers for understanding their 
obligations and achieving compliance.
    iii. The use of performance rather than design standards. Under the 
proposed rule, employers may achieve compliance through a variety of 
means. Employers may elect to continue to claim the EAP exemption for 
affected employees by adjusting salary levels, hiring additional 
workers or spreading overtime hours to other employees, or compensating 
employees for overtime hours worked. The Department makes available a 
variety of resources to employers for understanding their obligations 
and achieving compliance.
    iv. An exemption from coverage of the rule, or any part thereof, 
for such small entities. Creating an exemption from coverage of this 
rule for businesses with as many as 500 employees, those defined as 
small businesses under SBA's size standards, is inconsistent with 
Congressional intent in the enactment of the FLSA, which applies to all 
employers that satisfy the enterprise coverage threshold or employ 
individually covered employees. See 29 U.S.C. 203(s). Moreover, 
creating a regulatory exemption for small businesses would be beyond 
the scope of the Department's statutory authority to define and delimit 
the meaning of the term ``employed in a bona fide executive, 
administrative, or professional capacity.'' 29 U.S.C. 213(a)(1).

E. Identification, to the Extent Practicable, of All Relevant Federal 
Rules That May Duplicate, Overlap, or Conflict With the Proposed Rule

    The Department is not aware of any federal rules that duplicate, 
overlap, or conflict with this NPRM.

IX. Unfunded Mandates Reform Act Analysis

    The Unfunded Mandates Reform Act of 1995 (UMRA), 2 U.S.C. 1501, 
requires agencies to prepare a written statement for rules for which a 
general notice of proposed rulemaking was published and that include 
any federal mandate that may result in increased expenditures by state, 
local, and tribal governments, in the aggregate, or by the private 
sector, of $156 million ($100 million in 1995 dollars adjusted for 
inflation) or more in any one year. This statement must: (1) Identify 
the authorizing legislation; (2) present the estimated costs and 
benefits of the rule and, to the extent that such estimates are 
feasible and relevant, its estimated effects on the national economy; 
(3) summarize and evaluate state, local, and tribal government input; 
and (4) identify reasonable alternatives and select, or explain the 
non-selection, of the least costly, most cost-effective, or least 
burdensome alternative.

A. Authorizing Legislation

    This proposed rule is issued pursuant to section 13(a)(1) of the 
Fair Labor Standards Act, 29 U.S.C. 213(a)(1). The section exempts from 
the FLSA's minimum wage and overtime pay requirements ``any employee 
employed in a bona fide executive, administrative, or professional 
capacity (including any employee employed in the capacity of academic 
administrative personnel or teacher in elementary or secondary 
schools), or in the capacity of outside salesman (as such terms are 
defined and delimited from time to time by regulations of the 
Secretary, subject to the provisions of [the Administrative Procedure 
Act]. . .).'' 29 U.S.C. 213(a)(1). The requirements of the exemption 
provided by this section of the Act are contained in part 541 of the 
Department's regulations. Section 3(e) of the FLSA, 29 U.S.C. 203(e), 
defines ``employee'' to include most individuals employed by a state, 
political subdivision of a state, or interstate governmental agency. 
Section 3(x) of the FLSA, 29 U.S.C. 203(x), also defines public 
agencies to include the government of a state or political subdivision 
thereof, or any interstate governmental agency.

B. Assessment of Costs and Benefits

    For purposes of the UMRA, this rule includes a Federal mandate that 
is expected to result in increased expenditures by the private sector 
of more than $156 million in at least one year, but the rule will not 
result in increased expenditures by state, local and tribal 
governments, in the aggregate, of $156 million or more in any one year.
    Costs to state and local governments: Based on the RIA, the 
Department determined that the proposed rule will result in Year 1 
costs for state and local governments totaling $111.5 million; of which 
$28.3 million are direct employer costs and $83.2 million are transfers 
(Table 37). Additionally, the proposed rule will lead to $0.3 million 
in DWL. In subsequent years, the Department estimated that state and 
local governments may experience payroll increases of as much as $79.1 
million in any one year when the salary level is automatically updated 
(with automatic updating using the fixed percentile method).
    Costs to the private sector: The Department determined that the 
proposed rule will result in Year 1 costs to the private sector of 
approximately $2.0 billion, of which $563.8 million are direct employer 
costs and $1,396.2 million are transfers. Additionally, the proposed 
rule will result in $7.0 million in DWL. In subsequent years, the 
Department estimated that the private sector may experience a payroll 
increase of as much as $1,219.1 million per year (with automatic 
updating using the fixed percentile method).

[[Page 38608]]



      Table 37--Summary of Year 1 Affected EAP Workers, Regulatory Costs, and Transfers by Type of Employer
----------------------------------------------------------------------------------------------------------------
                                                             Total              Private         Government \a\
----------------------------------------------------------------------------------------------------------------
                                          Affected EAP Workers (1,000s)
----------------------------------------------------------------------------------------------------------------
Number..............................................             4,682               4,163                 507
----------------------------------------------------------------------------------------------------------------
                                        Direct Employer Costs (Millions)
----------------------------------------------------------------------------------------------------------------
Regulatory familiarization..........................              $254.5              $251.4                $3.1
Adjustment..........................................               160.1               142.3                17.3
Managerial..........................................               178.1               170.0                 7.9
Total direct costs..................................               592.7               563.8                28.3
----------------------------------------------------------------------------------------------------------------
                                              Transfers (Millions)
----------------------------------------------------------------------------------------------------------------
From employers to workers...........................            $1,482.5            $1,396.2               $83.2
----------------------------------------------------------------------------------------------------------------
                                  Direct Employer Costs & Transfers (Millions)
----------------------------------------------------------------------------------------------------------------
From employers......................................            $2,075.2            $1,960.0              $111.5
----------------------------------------------------------------------------------------------------------------
                                                 DWL (Millions)
----------------------------------------------------------------------------------------------------------------
DWL \b\.............................................                $7.4                $7.0                $0.3
----------------------------------------------------------------------------------------------------------------
\a\ Includes only state, local, and tribal governments.
\b\ DWL was estimated based on the aggregate impact of both the minimum wage and overtime pay provisions.

    The largest benefit to workers is the transfer of income from 
employers; but, to the extent that the benefits to workers outweigh the 
costs to employers, there may be a societal welfare increase due to 
this transfer. The channels through which societal welfare may 
increase, and other secondary benefits may occur, include: Decreased 
litigation costs due to fewer workers subject to the duties test, the 
multiplier effect of the transfer, increased productivity, reduced 
dependence on social assistance, and a potential increase in time off 
and its associated benefits to the social welfare of workers. 
Additionally, because of the increased salary level, overtime 
protection will be strengthened for 6.3 million salaried white collar 
workers and 3.7 million salaried blue collar workers who do not meet 
the duties requirements for the EAP exemption, but who earn between the 
current minimum salary level of $455 per week and the proposed salary 
level because their right to minimum wage and overtime protection will 
be clear rather than depend upon an analysis of their duties.
    UMRA requires agencies to estimate the effect of a regulation on 
the national economy if, at its discretion, such estimates are 
reasonably feasible and the effect is relevant and material. 5 U.S.C. 
1532(a)(4). However, OMB guidance on this requirement notes that such 
macro-economic effects tend to be measurable in nationwide econometric 
models only if the economic impact of the regulation reaches 0.25 
percent to 0.5 percent of GDP, or in the range of $41.9 billion to 
$83.8 billion (using 2013 GDP). A regulation with smaller aggregate 
effect is not likely to have a measurable impact in macro-economic 
terms unless it is highly focused on a particular geographic region or 
economic sector, which is not the case with this proposed rule.
    The Department's RIA estimates that the total first-year costs 
(direct employer costs, transfers from employers to workers, and 
deadweight loss) of the proposed rule will be approximately $2.0 
billion for private employers and $111.8 million for state and local 
governments. Given OMB's guidance, the Department has determined that a 
full macro-economic analysis is not likely to show any measurable 
impact on the economy. Therefore, these costs are compared to payroll 
costs and revenue to demonstrate the feasibility of adapting to these 
new rules.
    Total first-year private sector costs compose less than 0.04 
percent of private sector payrolls nationwide (2013 payroll costs were 
estimated to be $5.4 trillion).\185\ Total private sector first-year 
costs compose less than 0.006 percent of national private sector 
revenues (2013 revenues were estimated to be $32.9 trillion).\186\ The 
Department concludes that impacts of this magnitude are affordable and 
will not result in significant disruptions to typical firms in any of 
the major industry categories.
---------------------------------------------------------------------------

    \185\ Private sector payroll costs in 2007 were $4.8 trillion 
using the 2007 Economic Census of the United States. This was 
inflated to 2013 dollars using the CPI-U. Table EC0700A1: All 
sectors: Geographic Area Series: Economy-Wide Key Statistics: 2007.
    \186\ Private sector revenues in 2007 were $29.3 trillion using 
the 2007 Economic Census of the United States. This was inflated to 
2013 dollars using the CPI-U. Table EC0700A1: All sectors: 
Geographic Area Series: Economy-Wide Key Statistics: 2007.
---------------------------------------------------------------------------

    Total first-year state and local government costs compose 
approximately 0.01 percent of state and local government payrolls (2013 
payroll costs were estimated to be $864 billion).\187\ First-year state 
and local government costs compose 0.003 percent of state and local 
government revenues (2013 revenues were estimated to be $3.5 
trillion).\188\ Impacts of this magnitude will not result in 
significant disruptions to typical state and local governments. The 
$111.5 million in state and local government costs constitutes an 
average of approximately $1,240 for each of the approximately 90,100 
state and local entities. The Department considers impacts of this 
magnitude to be quite small both in absolute terms and in relation to 
payrolls and revenue.
---------------------------------------------------------------------------

    \187\ State and local payroll costs in 2012 were reported in the 
Census of Governments data as $852 billion. This was inflated to 
2013 dollars using the CPI-U. 2012 Census of Governments: Employment 
Summary Report. Available at: https://www2.census.gov/govs/apes/2012_summary_report.pdf.
    \188\ State and local revenues in 2011 were reported by the 
Census as $3.4 trillion. This was inflated to 2013 dollars using the 
CPI-U. State and Local Government Finances Summary: 2011. Available 
at: https://www2.census.gov/govs/local/summary_report.pdf.

---------------------------------------------------------------------------

[[Page 38609]]

C. Summary of State, Local, and Tribal Government Input

    As part of the Department's outreach program prior to the issuance 
of this NPRM, the Department conducted stakeholder listening sessions 
with representatives of state and local governments and tribal 
governments. In these sessions the Department asked stakeholders to 
address, among other issues, three questions: (1) What is the 
appropriate salary level for exemption; (2) what, if any, changes 
should be made to the duties tests; and (3) how can the regulations be 
simplified. The input received from state, local, and tribal government 
representatives was similar to that provided by representatives of 
private businesses and is summarized in section III. of this preamble. 
The discussions in the listening sessions have informed the development 
of this NPRM. The Department specifically seeks comments from state, 
local, and tribal governments concerning the ability of these entities 
to absorb the costs related to the proposed revisions.

D. Least Burdensome Option or Explanation Required

    The Department's consideration of various options has been 
described throughout the preamble. The Department believes that it has 
chosen the least burdensome but still cost-effective mechanism to 
update the salary level and index future levels that is also consistent 
with the Department's statutory obligation. Although some alternative 
options considered would have set the standard salary level at a rate 
lower than the proposed salary level, which might impose lower direct 
payroll costs on employers, that outcome may not necessarily be the 
most cost-effective or least burdensome alternative for employers. A 
lower salary level--or a degraded stagnant level over time--could 
result in a less effective bright-line test for separating exempt 
workers from those nonexempt workers intended to be within the Act's 
protection. A low salary level will also increase the role of the 
duties test in determining whether an employee is exempt, which would 
increase the likelihood of misclassification and, in turn, increase the 
risk that employees who should receive overtime and minimum wage 
protections under the FLSA are denied those protections.
    Selecting a standard salary level inevitably impacts both the risk 
and cost of misclassification of overtime-eligible employees earning 
above the salary level as well as the risk and cost of providing 
overtime protection to employees performing bona fide EAP duties who 
are paid below the salary level. An unduly low level risks increasing 
employer liability from unintentionally misclassifying workers as 
exempt; but an unduly high standard salary level increases labor costs 
to employers precluded from claiming the exemption for employees 
performing bona fide EAP duties. Thus the ultimate cost of the 
regulation is increased if the standard salary level is set either too 
low or too high. The Department has determined that setting the 
standard salary level at the 40th percentile of earnings for full-time 
salaried workers and automatically updating this level annually either 
by maintaining that earnings percentile or using the CPI-U best 
balances the risks and costs of misclassification of exempt status.

X. Executive Order 13132, Federalism

    The Department has (1) reviewed this proposed rule in accordance 
with Executive Order 13132 regarding federalism and (2) determined that 
it does not have federalism implications. The proposed rule would not 
have substantial direct effects on the States, on the relationship 
between the national government and the States, or on the distribution 
of power and responsibilities among the various levels of government.

XI. Executive Order 13175, Indian Tribal Governments

    This proposed rule would not have substantial direct effects on one 
or more Indian tribes, on the relationship between the Federal 
Government and Indian tribes, or on the distribution of power and 
responsibilities between the Federal Government and Indian tribes.

XII. Effects on Families

    The undersigned hereby certifies that the proposed rule would not 
adversely affect the well-being of families, as discussed under section 
654 of the Treasury and General Government Appropriations Act, 1999.

XIII. Executive Order 13045, Protection of Children

    This proposed rule would have no environmental health risk or 
safety risk that may disproportionately affect children.

XIV. Environmental Impact Assessment

    A review of this proposed rule in accordance with the requirements 
of the National Environmental Policy Act of 1969 (NEPA), 42 U.S.C. 4321 
et seq.; the regulations of the Council on Environmental Quality, 40 
CFR part 1500 et seq.; and the Departmental NEPA procedures, 29 CFR 
part 11, indicates that the rule would not have a significant impact on 
the quality of the human environment. There is, thus, no corresponding 
environmental assessment or an environmental impact statement.

XV. Executive Order 13211, Energy Supply

    This proposed rule is not subject to Executive Order 13211. It will 
not have a significant adverse effect on the supply, distribution, or 
use of energy.

XVI. Executive Order 12630, Constitutionally Protected Property Rights

    This proposed rule is not subject to Executive Order 12630 because 
it does not involve implementation of a policy that has takings 
implications or that could impose limitations on private property use.

XVII. Executive Order 12988, Civil Justice Reform Analysis

    This proposed rule was drafted and reviewed in accordance with 
Executive Order 12988 and will not unduly burden the Federal court 
system. The proposed rule was: (1) Reviewed to eliminate drafting 
errors and ambiguities; (2) written to minimize litigation; and (3) 
written to provide a clear legal standard for affected conduct and to 
promote burden reduction.

List of Subjects in 29 CFR Part 541

    Labor, Minimum wages, Overtime pay, Salaries, Teachers, Wages.

    Signed at Washington, DC this 18th day of June, 2015.
David Weil,
Administrator, Wage and Hour Division.

    For the reasons set out in the preamble, the Department of Labor 
proposes to amend title 29 of the Code of Federal Regulations, part 541 
as follows:

PART 541--DEFINING AND DELIMITING THE EXEMPTIONS FOR EXECUTIVE, 
ADMINISTRATIVE, PROFESSIONAL, COMPUTER AND OUTSIDE SALES EMPLOYEES

0
1. The authority citation for part 541 is revised to read as follows:

    Authority: 29 U.S.C. 213; Pub. L. 101-583, 104 Stat. 2871; 
Reorganization Plan No. 6 of 1950 (3 CFR, 1945-53 Comp., p. 1004); 
Secretary's Order 01-2014 (Dec. 10, 2014), 79 FR 77527 (Dec. 24, 
2014).

0
2. Revise paragraph (a)(1) of Sec.  541.100 to read as follows:

[[Page 38610]]

Sec.  541.100  General rule for executive employees.

    (a) * * *
    (1) Compensated on a salary basis as of [EFFECTIVE DATE OF FINAL 
RULE] at a rate per week of not less than $921 (or $774 per week, if 
employed in American Samoa by employers other than the Federal 
government), exclusive of board, lodging or other facilities. As of 
[DATE TBD] on each subsequent year, compensated on a salary basis at a 
rate per week of not less than the updated salary rate published 
annually by the Secretary in the Federal Register at least 60 days 
earlier (with the rate for American Samoa to be calculated at 84 
percent of the updated salary rate, provided that when the highest 
industry minimum wage for American Samoa equals the minimum wage under 
29 U.S.C. 206(a)(1), exempt employees employed in all industries in 
American Samoa shall be paid the full salary rate), exclusive of board, 
lodging or other facilities;
* * * * *
0
3. Revise paragraph (a)(1) of Sec.  541.200 to read as follows:


Sec.  541.200  General rule for administrative employees.

    (a) * * *
    (1) Compensated on a salary or fee basis as of [EFFECTIVE DATE OF 
FINAL RULE] at a rate per week of not less than $921 (or $774 per week, 
if employed in American Samoa by employers other than the Federal 
government), exclusive of board, lodging or other facilities. As of 
[DATE TBD] on each subsequent year, compensated on a salary or fee 
basis at a rate per week of not less than the updated salary rate 
published annually by the Secretary in the Federal Register at least 60 
days earlier (with the rate for American Samoa to be calculated at 84 
percent of the updated salary rate, provided that when the highest 
industry minimum wage for American Samoa equals the minimum wage under 
29 U.S.C. 206(a)(1), exempt employees employed in all industries in 
American Samoa shall be paid the full salary rate), exclusive of board, 
lodging or other facilities;
* * * * *
0
4. Revise paragraph (a)(1) of Sec.  541.204 to read as follows:


Sec.  541.204  Educational establishments.

    (a) * * *
    (1) Compensated on a salary or fee basis as of [EFFECTIVE DATE OF 
FINAL RULE] at a rate per week of not less than $921 (or $774 per week, 
if employed in American Samoa by employers other than the Federal 
government), exclusive of board, lodging or other facilities; or on a 
salary basis which is at least equal to the entrance salary for 
teachers in the educational establishment by which employed. As of 
[DATE TBD] on each subsequent year, compensated on a salary or fee 
basis at a rate per week of not less than the updated salary rate 
published annually by the Secretary in the Federal Register at least 60 
days earlier (with the rate for American Samoa to be calculated at 84 
percent of the updated salary rate, provided that when the highest 
industry minimum wage for American Samoa equals the minimum wage under 
29 U.S.C. 206(a)(1), exempt employees employed in all industries in 
American Samoa shall be paid the full salary rate), exclusive of board, 
lodging or other facilities; or on a salary basis which is at least 
equal to the entrance salary for teachers in the educational 
establishment by which employed; and
* * * * *
0
5. Revise paragraph (a)(1) of Sec.  541.300 to read as follows:


Sec.  541.300  General rule for professional employees.

    (a) * * *
    (1) Compensated on a salary or fee basis as of [EFFECTIVE DATE OF 
FINAL RULE] at a rate per week of not less than $921 (or $774 per week, 
if employed in American Samoa by employers other than the Federal 
government), exclusive of board, lodging or other facilities. As of 
[DATE TBD] on each subsequent year, compensated on a salary or fee 
basis at a rate per week of not less than the updated salary rate 
published annually by the Secretary in the Federal Register at least 60 
days earlier (with the rate for American Samoa to be calculated at 84 
percent of the updated salary rate, provided that when the highest 
industry minimum wage for American Samoa equals the minimum wage under 
29 U.S.C. 206(a)(1), exempt employees employed in all industries in 
American Samoa shall be paid the full salary rate), exclusive of board, 
lodging or other facilities; and
* * * * *
0
6. Remove the first sentence of Sec.  541.400(b) introductory text and 
add three sentences in its place to read as follows:


Sec.  541.400  General rule for computer employees.

* * * * *
    (b) The section 13(a)(1) exemption applies to any computer employee 
who, as of [EFFECTIVE DATE OF FINAL RULE] is compensated on a salary or 
fee basis at a rate per week of not less than $921 (or $774 per week, 
if employed in American Samoa by employers other than the Federal 
government), exclusive of board, lodging or other facilities. As of 
[DATE TBD] on each subsequent year, the section 13(a)(1) exemption 
applies to any computer employee who is compensated on a salary or fee 
basis at a rate per week of not less than the updated salary rate 
published annually by the Secretary in the Federal Register at least 60 
days earlier (with the rate for American Samoa to be calculated at 84 
percent of the updated salary rate, provided that when the highest 
industry minimum wage for American Samoa equals the minimum wage under 
29 U.S.C. 206(a)(1), exempt employees employed in all industries in 
American Samoa shall be paid the full salary rate), exclusive of board, 
lodging or other facilities. The section 13(a)(17) exemption applies to 
any computer employee compensated on an hourly basis at a rate of not 
less than $27.63 an hour. * * *
* * * * *
0
7. Amend Sec.  541.600 by:
0
a. Removing the first sentence of paragraph (a) and adding two 
sentences in its place; and
0
b. Removing the first sentence of paragraph (b) and adding two 
sentences in its place.
    The additions read as follows:


Sec.  541.600  Amount of salary required.

    (a) To qualify as an exempt executive, administrative or 
professional employee under section 13(a)(1) of the Act, an employee 
must be compensated on a salary basis as of [EFFECTIVE DATE OF FINAL 
RULE] at a rate per week of not less than $921 (or $774 per week, if 
employed in American Samoa by employers other than the Federal 
government), exclusive of board, lodging or other facilities. As of 
[DATE TBD] on each subsequent year, such employee must be compensated 
on a salary basis at a rate per week of not less than the updated 
salary rate published annually by the Secretary in the Federal Register 
at least 60 days earlier (with the rate for American Samoa to be 
calculated at 84 percent of the updated salary rate, provided that when 
the highest industry minimum wage for American Samoa equals the minimum 
wage under 29 U.S.C. 206(a)(1), exempt employees employed in all 
industries in American Samoa shall be paid the full salary rate), 
exclusive of board, lodging or other facilities. * * *
    (b) The required amount of compensation per week may be translated 
into equivalent amounts for periods longer than one week. The

[[Page 38611]]

requirement will be met if the employee is compensated biweekly on a 
salary basis of $[DOUBLE THE 40th PERCENTILE AMOUNT], semimonthly on a 
salary basis of $[THE 40th PERCENTILE AMOUNT, MULTIPLIED BY 52 AND 
DIVIDED BY 24], or monthly on a salary basis of $[THE 40th PERCENTILE 
AMOUNT MULTIPLIED BY 52 AND DIVIDED BY 12]. * * *
* * * * *
0
8. Amend Sec.  541.601 by:
0
a. Revising paragraph (a);
0
b. Removing the first sentence of paragraph (b)(1) and adding two 
sentences in its place; and
0
c. Revising paragraph (b)(2).
    The revisions read as follows:


Sec.  541.601  Highly compensated employees.

    (a) An employee with total annual compensation of at least $122,148 
as of [EFFECTIVE DATE OF FINAL RULE] is deemed exempt under section 
13(a)(1) of the Act if the employee customarily and regularly performs 
any one or more of the exempt duties or responsibilities of an 
executive, administrative or professional employee identified in 
subparts B, C, or D of this part. As of [DATE TBD] on each subsequent 
year, an employee with total annual compensation of at least the 
updated compensation rate published annually by the Secretary in the 
Federal Register at least 60 days earlier is deemed exempt under 
section 13(a)(1) of the Act if the employee customarily and regularly 
performs any one or more of the exempt duties or responsibilities of an 
executive, administrative or professional employee identified in 
subparts B, C, or D of this part.
    (b)(1) ``Total annual compensation'' must include at least a weekly 
amount that is, as of [EFFECTIVE DATE OF FINAL RULE] $921 paid on a 
salary or fee basis. As of [DATE TBD] of each year, ``total annual 
compensation'' must include a weekly amount that is not less than the 
updated salary rate published annually by the Secretary in the Federal 
Register at least 60 days earlier , paid on a salary or fee basis. * * 
*
    (2) If an employee's total annual compensation does not total at 
least the minimum amount established in paragraph (a) of this section 
by the last pay period of the 52-week period, the employer may, during 
the last pay period or within one month after the end of the 52-week 
period, make one final payment sufficient to achieve the required 
level. For example, if the current annual salary level for a highly 
compensated employee is $122,148, an employee may earn $100,000 in base 
salary, and the employer may anticipate based upon past sales that the 
employee also will earn $25,000 in commissions. However, due to poor 
sales in the final quarter of the year, the employee actually only 
earns $10,000 in commissions. In this situation, the employer may 
within one month after the end of the year make a payment of at least 
$12,148 to the employee. Any such final payment made after the end of 
the 52-week period may count only toward the prior year's total annual 
compensation and not toward the total annual compensation in the year 
it was paid. If the employer fails to make such a payment, the employee 
does not qualify as a highly compensated employee, but may still 
qualify as exempt under subparts B, C, or D of this part.
* * * * *
0
9. Revise Sec.  541.604 to read as follows:


Sec.  541.604  Minimum guarantee plus extras.

    (a) An employer may provide an exempt employee with additional 
compensation without losing the exemption or violating the salary basis 
requirement, if the employment arrangement also includes a guarantee of 
at least the minimum weekly-required amount paid on a salary basis. 
Thus, for example, if the current weekly salary level is $921, an 
exempt employee guaranteed at least $921 each week paid on a salary 
basis may also receive additional compensation of a one percent 
commission on sales. An exempt employee also may receive a percentage 
of the sales or profits of the employer if the employment arrangement 
also includes a guarantee of at least $921 each week paid on a salary 
basis. Similarly, the exemption is not lost if an exempt employee who 
is guaranteed at least $921 each week paid on a salary basis also 
receives additional compensation based on hours worked for work beyond 
the normal workweek. Such additional compensation may be paid on any 
basis (e.g., flat sum, bonus payment, straight-time hourly amount, time 
and one-half or any other basis), and may include paid time off.
    (b) An exempt employee's earnings may be computed on an hourly, a 
daily or a shift basis, without losing the exemption or violating the 
salary basis requirement, if the employment arrangement also includes a 
guarantee of at least the minimum weekly required amount paid on a 
salary basis regardless of the number of hours, days or shifts worked, 
and a reasonable relationship exists between the guaranteed amount and 
the amount actually earned. The reasonable relationship test will be 
met if the weekly guarantee is roughly equivalent to the employee's 
usual earnings at the assigned hourly, daily or shift rate for the 
employee's normal scheduled workweek. Thus, for example, if the weekly 
salary level is $921, an exempt employee guaranteed compensation of at 
least $1,000 for any week in which the employee performs any work, and 
who normally works four or five shifts each week, may be paid $300 per 
shift without violating the salary basis requirement. The reasonable 
relationship requirement applies only if the employee's pay is computed 
on an hourly, daily or shift basis. It does not apply, for example, to 
an exempt store manager paid a guaranteed salary per week that exceeds 
the current salary level who also receives a commission of one-half 
percent of all sales in the store or five percent of the store's 
profits, which in some weeks may total as much as, or even more than, 
the guaranteed salary.
0
10. Revise paragraph (b) of Sec.  541.605 to read as follows:


Sec.  541.605  Fee basis.

* * * * *
    (b) To determine whether the fee payment meets the minimum amount 
of salary required for exemption under these regulations, the amount 
paid to the employee will be tested by determining the time worked on 
the job and whether the fee payment is at a rate that would amount to 
at least the minimum required salary per week if the employee worked 40 
hours. Thus, if the salary level were $921, an artist paid $500 for a 
picture that took 20 hours to complete meets the minimum salary 
requirement for exemption since earnings at this rate would yield the 
artist $1000 if 40 hours were worked.
0
11. Revise Sec.  541.709 to read as follows:


Sec.  541.709  Motion picture producing industry.

    The requirement that the employee be paid ``on a salary basis'' 
does not apply to an employee in the motion picture producing industry 
who is compensated, as of [EFFECTIVE DATE OF FINAL RULE], at a base 
rate of at least $1,404 per week (exclusive of board, lodging, or other 
facilities); and as of [DATE TBD] on each subsequent year, is 
compensated at a base rate of at least $[MOST RECENTLY EFFECTIVE MOTION 
PICTURE INDUSTRY BASE RATE INCREASED AT THE SAME RATIO AS THE STANDARD 
SALARY LEVEL IS INCREASED] (exclusive of board, lodging, or other 
facilities). Thus, an employee in this industry who is otherwise exempt 
under subparts B, C, or D of this part, and who is employed at a base 
rate of at least the applicable current minimum amount a week is

[[Page 38612]]

exempt if paid a proportionate amount (based on a week of not more than 
6 days) for any week in which the employee does not work a full 
workweek for any reason. Moreover, an otherwise exempt employee in this 
industry qualifies for exemption if the employee is employed at a daily 
rate under the following circumstances:
    (a) The employee is in a job category for which a weekly base rate 
is not provided and the daily base rate would yield at least the 
minimum weekly amount if 6 days were worked; or
    (b) The employee is in a job category having the minimum weekly 
base rate and the daily base rate is at least one-sixth of such weekly 
base rate.

[FR Doc. 2015-15464 Filed 7-2-15; 8:45 am]
 BILLING CODE P
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.