Mortality Tables for Determining Present Value Under Defined Benefit Pension Plans, 95911-95929 [2016-30906]

Agencies

• Internal Revenue Service
• DEPARTMENT OF THE TREASURY

[Federal Register Volume 81, Number 250 (Thursday, December 29, 2016)]
[Proposed Rules]
[Pages 95911-95929]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 2016-30906]


========================================================================
Proposed Rules
                                                Federal Register
________________________________________________________________________

This section of the FEDERAL REGISTER contains notices to the public of 
the proposed issuance of rules and regulations. The purpose of these 
notices is to give interested persons an opportunity to participate in 
the rule making prior to the adoption of the final rules.

========================================================================


Federal Register / Vol. 81, No. 250 / Thursday, December 29, 2016 / 
Proposed Rules

[[Page 95911]]



DEPARTMENT OF THE TREASURY

Internal Revenue Service

26 CFR Part 1

[REG-112324-15]
RIN 1545-BM71


Mortality Tables for Determining Present Value Under Defined 
Benefit Pension Plans

AGENCY: Internal Revenue Service (IRS), Treasury.

ACTION: Notice of proposed rulemaking and notice of public hearing.

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

SUMMARY: This document contains proposed regulations prescribing 
mortality tables to be used by most defined benefit pension plans. The 
tables specify the probability of survival year-by-year for an 
individual based on age, gender, and other factors. This information is 
used (together with other actuarial assumptions) to calculate the 
present value of a stream of expected future benefit payments for 
purposes of determining the minimum funding requirements for the plan. 
These mortality tables are also relevant to determining the minimum 
required amount of a lump-sum distribution from such a plan. In 
addition, this document contains proposed regulations to update the 
requirements that a plan sponsor must meet in order to obtain IRS 
approval to use mortality tables specific to the plan for minimum 
funding purposes (instead of the generally applicable mortality 
tables). These regulations affect participants in, beneficiaries of, 
employers maintaining, and administrators of certain retirement plans.

DATES: Comments and outlines of topics to be discussed at the public 
hearing scheduled for April 13, 2017 must be received by March 29, 
2017.

ADDRESSES: Send submissions to: CC:PA:LPD:PR (REG-112324-15), Room 
5203, Internal Revenue Service, P.O. Box 7604, Ben Franklin Station, 
Washington, DC 20044. Submissions may be hand-delivered Monday through 
Friday between the hours of 8 a.m. and 4 p.m. to CC:PA:LPD:PR (REG-
112324-15), Courier's Desk, Internal Revenue Service, 1111 Constitution 
Avenue NW., Washington, DC 20224, or sent electronically, via the 
Federal eRulemaking Portal at www.regulations.gov (IRS REG-112324-15). 
The public hearing will be held in the IRS Auditorium, Internal Revenue 
Building, 1111 Constitution Avenue NW., Washington, DC 20224.

FOR FURTHER INFORMATION CONTACT: Concerning the regulations, Thomas 
Morgan at (202) 317-6700; concerning the construction of the base 
mortality tables and the static mortality tables for 2018, Michael 
Spaid at (206) 946-3480; concerning submission of comments, the 
hearing, and/or to be placed on the building access list to attend the 
hearing, Regina Johnson at (202) 317-6901 (not toll-free numbers).

SUPPLEMENTARY INFORMATION: 

Background

A. Generally applicable mortality tables

    Section 412 of the Internal Revenue Code (Code) prescribes minimum 
funding requirements for defined benefit pension plans. Section 430, 
which was added to the Code by the Pension Protection Act of 1996, 
Public Law 109-280 (120 Stat. 780 (2006)), specifies the minimum 
funding requirements that apply generally to defined benefit plans that 
are not multiemployer plans.\1\ Section 430(a) defines the minimum 
required contribution by reference to the plan's funding target for the 
plan year. Under section 430(d)(1), a plan's funding target for a plan 
year generally is the present value of all benefits accrued or earned 
under the plan as of the first day of that plan year.
---------------------------------------------------------------------------

    \1\ Section 302 of the Employee Retirement Income Security Act 
of 1974, Public Law 93-406, as amended (ERISA) sets forth funding 
rules that are parallel to those in section 412 of the Code, and 
section 303 of ERISA sets forth additional funding rules for defined 
benefit plans (other than multiemployer plans) that are parallel to 
those in section 430 of the Code. Under section 101 of 
Reorganization Plan No. 4 of 1978 (43 FR 47713) and section 302 of 
ERISA, the Secretary of the Treasury has interpretive jurisdiction 
over the subject matter addressed in these proposed regulations for 
purposes of ERISA, as well as the Code. Thus, these proposed 
Treasury regulations issued under section 430 of the Code apply as 
well for purposes of section 303 of ERISA.
---------------------------------------------------------------------------

    Section 430(h)(3) contains rules regarding the mortality tables to 
be used under section 430. Under section 430(h)(3)(A), except as 
provided in section 430(h)(3)(C) or (D), the Secretary is to prescribe 
by regulation mortality tables to be used in determining any present 
value or making any computation under section 430. Those mortality 
tables are to be based on the actual mortality experience of pension 
plan participants and projected trends in that experience. In 
prescribing those mortality tables, the Secretary is required to take 
into account results of available independent studies of mortality of 
individuals covered by pension plans.\2\ Under section 430(h)(3)(B), 
the Secretary is required to revise any mortality table in effect under 
section 430(h)(3)(A) at least every 10 years to reflect actual 
mortality experience of pension plan participants and projected trends 
in that experience.
---------------------------------------------------------------------------

    \2\ The standards prescribed for developing these mortality 
tables are the same as the standards that are prescribed for 
developing mortality tables for multiemployer plans under section 
431(c)(6)(D)(iv)(II) (which are used determine current liability in 
order to determine the minimum full funding limitation under section 
431(c)(6)(B)). These standards also apply for purposes of 
determining current liability in order to determine the minimum full 
funding limitation under section 433(c)(2)(C) for a CSEC plan (as 
defined in section 414(y)).
---------------------------------------------------------------------------

    Section 430(h)(3)(D) provides for the use of separate mortality 
tables with respect to certain individuals who are entitled to benefits 
on account of disability. These separate mortality tables are permitted 
to be used with respect to disabled individuals in lieu of the 
generally applicable mortality tables provided pursuant to section 
430(h)(3)(A) or the substitute mortality tables under section 
430(h)(3)(C). The Secretary is to establish separate tables for 
individuals with disabilities occurring in plan years beginning before 
January 1, 1995, and in later plan years, with the mortality tables for 
individuals with disabilities occurring in those later plan years 
applying only to individuals who are disabled within the meaning of 
Title II of the Social Security Act.
    Section 417(e)(3) generally provides that the present value of 
certain benefits under a qualified pension plan (including single-sum 
distributions) must not be less than the present value of the accrued 
benefit using applicable interest rates and the applicable mortality 
table. Section 417(e)(3)(B)

[[Page 95912]]

defines the term ``applicable mortality table'' as the mortality table 
specified for the plan year for minimum funding purposes under section 
430(h)(3)(A) (without regard to the rules for substitute mortality 
tables under section 430(h)(3)(C) or mortality tables for disabled 
individuals under section 430(h)(3)(D)), modified as appropriate by the 
Secretary. The modifications made by the Secretary to the section 
430(h)(3)(A) mortality table to determine the section 417(e)(3)(B) 
applicable mortality table are not addressed in these proposed 
regulations. Revenue Ruling 2007-67, 2007-2 CB 1047, describes the 
modifications that are currently applied to determine the section 
417(e)(3)(B) applicable mortality table.
    Final regulations under section 430(h)(3) were published in the 
Federal Register on July 31, 2008 in TD 9419, 73 FR 44632. The final 
regulations issued in 2008 include rules regarding generally applicable 
mortality tables, which are set forth in Sec.  1.430(h)(3)-1 (the 2008 
general mortality table regulations). The final regulations issued in 
2008 also include rules regarding substitute mortality tables, which 
are set forth in Sec.  1.430(h)(3)-2 (the 2008 substitute mortality 
table regulations).
    The 2008 general mortality table regulations prescribe a base 
mortality table and a set of mortality improvement rates, which may be 
reflected through the use of either generational mortality tables or 
static mortality tables. Generational mortality tables are a series of 
mortality tables, one for each year of birth, each of which fully 
reflects projected trends in mortality rates. The static mortality 
tables (which are updated annually) use a single mortality table for 
all years of birth to approximate the present value that would be 
determined using the generational morality tables. Section 1.430(h)(3)-
1 includes static mortality tables for valuation dates occurring in 
2008 and provides that static mortality tables for valuation dates 
occurring in later years are to be published in the Internal Revenue 
Bulletin.
    The mortality tables included in Sec.  1.430(h)(3)-1 are based on 
the mortality tables included in the RP-2000 Mortality Tables Report 
(referred to in this preamble as the RP-2000 mortality tables) released 
by the Society of Actuaries in July 2000 (updated in May 2001) and a 
set of mortality improvement projection factors (the Scale AA 
Projection Factors) that was also included in the RP-2000 Mortality 
Tables Report.
    Section 1.431(c)(6)-1 provides that the same mortality assumptions 
that apply for purposes of section 430(h)(3)(A) and Sec.  1.430(h)(3)-
1(a)(2) are used to determine a multiemployer plan's current liability 
for purposes of applying the full-funding rules of section 431(c)(6). 
For this purpose, a multiemployer plan is permitted to apply either the 
annually-adjusted static mortality tables or the generational mortality 
tables.
    Static mortality tables for valuation dates occurring during 2009-
2013 were published in Notice 2008-85, 2008-42 IRB 905. Updated static 
mortality tables for valuation dates occurring during 2014 and 2015 
were published in Notice 2013-49, 2013-32 IRB 127. Updated static 
mortality tables for valuation dates occurring in 2016 were published 
in Notice 2015-53, 2015-33 IRB 190. Updated static mortality tables for 
valuation dates occurring in 2017 were published in Notice 2016-50, 
2016-38 IRB 371.
    Notice 2013-49 requested comments on whether it continues to be 
necessary to provide multiple alternative versions of the mortality 
tables in order to accommodate limitations in some actuarial software. 
Notice 2013-49 also requested comments on whether a separate disability 
mortality table is still warranted with respect to participants who 
became disabled before 1995. Finally, Notice 2013-49 noted that the 
Treasury Department (Treasury) and the IRS were aware that the Society 
of Actuaries was conducting a mortality study of pension plan 
participants and specifically requested comments on whether other 
studies of actual mortality experience of pension plan participants and 
projected trends of that experience are available that should be 
considered for use in developing mortality tables for future use under 
section 430(h)(3).
    In October 2014, the Retirement Plans Experience Committee (RPEC) 
of the Society of Actuaries issued a new mortality study of 
participants in private pension plans, referred to as the RP-2014 
Mortality Tables Report (which sets forth mortality tables that are 
referred to as the RP-2014 mortality tables). The RP-2014 Mortality 
Tables Report, as revised November 2014, is available at www.soa.org/Research/Experience-Study/pension/research-2014-rp.aspx. At the same 
time, RPEC issued a companion study of mortality improvement, referred 
to as the Mortality Improvement Scale MP-2014 Report (which sets forth 
mortality improvement rates that are referred to as Scale MP-2014 
Rates). As described in the Mortality Improvement Scale MP-2014 Report, 
(available at www.soa.org/Research/Experience-Study/pension/research-2014-mp.aspx), the Scale MP-2014 rates were based on mortality 
improvement experience for the general population through 2009.
    In October 2015, RPEC released an update to the Scale MP-2014 
Rates. The updated rates, referred to as Scale MP-2015 Rates, were 
released as part of the Mortality Improvement Scale MP-2015 Report 
(which is available at https://www.soa.org/Research/Experience-Study/Pension/research-2015-mp.aspx). The Scale MP-2015 Rates were created 
using historical data for mortality improvement for the general 
population through 2011, and the same model and parameters that were 
used to produce Scale MP-2014 Rates. In conjunction with the release of 
the updated rates, RPEC indicated the intent to reflect the latest data 
available by providing future annual updates to the model as soon as 
practicable following the public release of updated data upon which the 
model is constructed.
    In October 2016, RPEC released a further update to the Scale MP-
2014 Rates, which are referred to as the Scale MP-2016 Rates. The Scale 
MP-2016 Rates take into account data for mortality improvement for the 
general population for years 2012 and 2013, along with an estimate of 
mortality rates for 2014. As described in the Mortality Improvement 
Scale MP-2016 Report (which is available at www.soa.org/Research/Experience-Study/Pension/research-2016-mp.aspx), in developing the 
Scale MP-2016 rates, RPEC changed some of the parameters from those 
that were used in developing the Scale MP-2014 Rates.

B. Plan-Specific Substitute Mortality Tables

    Section 430(h)(3)(C) prescribes rules for a plan sponsor's use of 
substitute mortality tables reflecting the specific mortality 
experience of a plan's population instead of using the generally 
applicable mortality tables. Under section 430(h)(3)(C), the plan 
sponsor may request the Secretary's approval to use plan-specific 
substitute mortality tables that meet requirements specified in the 
statute. If approved, these substitute mortality tables are used to 
determine present values and make computations under section 430 during 
the period of consecutive plan years (not to exceed 10) specified in 
the request. In order for a plan sponsor to use a substitute mortality 
table for a plan, the statute requires that: (1) The plan has a 
sufficient number of plan participants and has been maintained for a 
sufficient period of time in order

[[Page 95913]]

to have credible mortality information necessary to create a substitute 
mortality table; and (2) the tables reflect the actual mortality 
experience of the plan's participants and projected trends in general 
mortality experience of participants in pension plans. Except as 
provided by the Secretary, a plan sponsor must not use substitute 
mortality tables for any plan unless substitute mortality tables are 
established and used for each plan maintained by the plan sponsor and 
its controlled group.
    Regulations issued in 2008 set forth rules regarding the use of 
substitute mortality tables. Under Sec.  1.430(h)(3)-2(b), in order to 
use substitute mortality tables with respect to a plan, a plan sponsor 
must submit a written request to the Commissioner that demonstrates 
that those substitute mortality tables comply with applicable 
requirements. A request to use substitute mortality tables must specify 
the first plan year, and the term of years (not more than 10), for 
which the tables are requested to be used. In general, substitute 
mortality tables may not be used for a plan year unless the plan 
sponsor submits the request at least 7 months prior to the first day of 
the first plan year for which the substitute mortality tables are to 
apply.
    The Commissioner has a 180-day period to review a request for the 
use of substitute mortality tables. If the Commissioner does not issue 
a denial within this 180-day period, the request is deemed to have been 
approved unless the Commissioner and the plan sponsor have agreed to 
extend that period. The Commissioner may request additional information 
with respect to a submission. Failure to provide that information on a 
timely basis is grounds for denial of the plan sponsor's request. In 
addition, the Commissioner will deny a request if the request fails to 
meet the requirements to use substitute mortality tables or if the 
Commissioner determines that a substitute mortality table does not 
sufficiently reflect the mortality experience of the applicable plan 
population.
    Under Sec.  1.430(h)(3)-2(c)(1)(i), substitute mortality tables 
must reflect the actual mortality experience of the pension plan for 
which the tables are to be used. Separate mortality tables must be 
established for each gender under the plan, and a substitute mortality 
table may be established for a gender only if the plan has credible 
mortality experience with respect to that gender. If the mortality 
experience for one gender is credible but the mortality experience for 
the other gender is not credible, the substitute mortality tables are 
used for the gender that has credible mortality experience, and the 
mortality tables under Sec.  1.430(h)(3)-1 are used for the gender that 
does not have credible mortality experience.
    Section 1.430(h)(3)-2(c)(1)(ii) provides that, for purposes of 
determining whether substitute mortality tables may be used, there is 
credible mortality experience for a gender within a plan if and only 
if, over the period covered by the experience study, there are at least 
1,000 deaths within that gender.\3\ Pursuant to Sec.  1.430(h)(3)-
2(c)(2)(ii), the minimum length of the experience study period is 2 
years and the maximum length of the experience study period is 5 years 
(and can be lengthened in published guidance). Furthermore, under that 
provision, the last day of the final year reflected in the experience 
data must be less than three years before the first day of the first 
plan year for which the substitute mortality tables are to apply.
---------------------------------------------------------------------------

    \3\ The 1,000-death threshold for credible mortality experience 
under the regulations was intended to provide a high degree of 
confidence that the plan's past mortality experience will be 
predictive of its future mortality, and is consistent with relevant 
actuarial literature (see, for example, Thomas N. Herzog, 
Introduction to Credibility Theory (1999); Stuart A. Klugman, et. 
al., Loss Models: From Data to Decisions (2004)).
---------------------------------------------------------------------------

    Under Sec.  1.430(h)(3)-2(c)(2), development of a substitute 
mortality table under the regulations requires creation of a base table 
and identification of a base year, which are then used to determine a 
substitute mortality table. The base table must be developed from a 
study of the mortality experience of the plan using amounts-weighted 
data.
    Under Sec.  1.430(h)(3)-2(c)(3), a plan's substitute mortality 
tables must be generational mortality tables. Substitute mortality 
tables are determined using the base mortality tables developed from 
the experience study and the Scale AA Projection Factors, which are 
also used for the generally applicable mortality tables.
    Under Sec.  1.430(h)(3)-2(c)(4), separate substitute mortality 
tables are permitted (but not required) to be established for separate 
populations within a gender, such as annuitants and nonannuitants or 
hourly and salaried individuals. Under that provision, separate 
substitute mortality tables generally are permitted to be used for a 
separate population within a gender under a plan only if all 
individuals of that gender in the plan are divided into separate 
populations, each separate population has credible mortality experience 
(determined in the same manner as determining whether a gender has 
credible mortality experience), and the separate substitute mortality 
table for each separate population is developed using mortality 
experience data for that population.
    Section 1.430(h)(3)-2(d)(3) prescribes rules for aggregating plans 
for purposes of using substitute mortality tables. Under Sec.  
1.430(h)(3)-2(d)(3), in order to use a set of substitute mortality 
tables for two or more plans, the rules set forth in the regulations 
must be applied by treating those plans as a single plan. In such a 
case, the substitute mortality tables must be used for all such plans 
and must be based on data collected with respect to all such plans.
    Section 1.430(h)(3)-2(d)(4) provides for the early termination of 
the use of substitute mortality tables in certain specified 
circumstances, including pursuant to a replacement of the mortality 
tables specified in Sec.  1.430(h)(3)-1. The early termination pursuant 
to such a replacement must be effective as of a date specified in 
guidance published in the Internal Revenue Bulletin.
    Rev. Proc. 2008-62, 2008-2 CB 935, sets forth the procedure by 
which a plan sponsor of a defined benefit plan may request and obtain 
approval for the use of plan-specific substitute mortality tables in 
accordance with section 430(h)(3)(C). The revenue procedure specifies 
the information that must be provided in order to request the use of 
substitute mortality tables and specifies two alternative acceptable 
methods of construction for base substitute mortality tables. Under 
section 11 of Rev. Proc. 2008-62, a base table for a population can be 
created from the unadjusted base table for the population through the 
application of a graduation method generally used by the actuarial 
profession in the United States (for example, the Whittaker-Henderson 
Type B graduation method or the Karup-King graduation method). Section 
12 of Rev. Proc. 2008-62 provides for an alternative method of 
constructing a base table through the application of a fixed percentage 
to the mortality rates of a standard mortality table, projected to the 
base year. This alternative method can be used only if the IRS 
determines that the resulting base table sufficiently reflects the 
mortality experience of the applicable plan population. In general, the 
standard mortality table that is used under this alternative method is 
a projection of the base mortality table that applies for the 
population under Sec.  1.430(h)(3)-1; however, the IRS will consider 
requests for the approval of base tables constructed through the 
application of a fixed percentage to the

[[Page 95914]]

mortality rates of other published generally accepted mortality tables.
    Section 503 of the Bipartisan Budget Act of 2015, Public Law 114-
74, 129 Stat. 584, which was enacted November 2, 2015, provides for 
changes to the rules on the use of substitute mortality tables. Under 
that section, the determination of whether a plan has credible 
information that can be used to develop a substitute mortality table 
must be made in accordance with established actuarial credibility 
theory, which (1) is materially different from the rules for using 
substitute mortality tables (including Revenue Procedure 2007-37) \4\ 
that are in effect on November 2, 2015; and (2) permits the use of 
mortality tables that reflect adjustments to the generally applicable 
mortality tables, if those adjustments are based on the actual 
experience of the pension plan maintained by the plan sponsor. This 
provision applies to plan years beginning after December 31, 2015.
---------------------------------------------------------------------------

    \4\ Rev. Proc. 2007-37, 2007-1 CB 1433, was not in effect on 
November 2, 2015. It was issued in 2007 in conjunction with proposed 
regulations regarding substitute mortality tables (REG-143601-06, 72 
FR 29456), and was replaced by Rev. Proc. 2008-62 when those 
regulations were finalized in 2008.
---------------------------------------------------------------------------

Explanation of Provisions
    These proposed regulations set forth the methodology Treasury and 
the IRS would use to update the generally applicable mortality tables 
that are used to determine present value or make any computation under 
section 430. Pursuant to section 417(e)(3)(B), a modified version of 
these updated tables would be used for purposes of determining the 
amount of a single-sum distribution (or another accelerated form of 
distribution).\5\ This methodology for developing updated tables under 
section 430(h)(3)(A) is being proposed pursuant to the requirement 
under section 430(h)(3)(B) to revise the mortality tables used under 
section 430 to reflect the actual mortality experience of pension plan 
participants and projected trends in that experience. As under the 2008 
general mortality table regulations, the methodology involves the 
separate determination of base tables and the projection of mortality 
improvement.
---------------------------------------------------------------------------

    \5\ After these regulations are finalized, the section 
417(e)(3)(B) applicable mortality table will be specified in 
guidance published in the Internal Revenue Bulletin. See Sec.  
601.601(d)(2)(ii)(b) of this chapter.
---------------------------------------------------------------------------

    These proposed regulations also set forth rules for the use of 
substitute mortality tables. The rules on substitute mortality tables 
are being proposed pursuant to section 503 of the Bipartisan Budget Act 
of 2015, which requires that the determination of whether the plan has 
credible information be made in accordance with established actuarial 
credibility theory. Pursuant to that requirement, Treasury and the IRS 
undertook a review of actuarial literature regarding credibility theory 
and consulted with experts on that topic from the Society of Actuaries. 
Based on that review and analysis, the proposed regulations set forth a 
method for developing substitute mortality tables that is materially 
different from the method that is required under the 2008 substitute 
mortality table regulations and the associated revenue procedure.
    The method for developing substitute mortality tables that is set 
forth in the proposed regulations is simpler than the method that 
applies under the 2008 substitute mortality table regulations, and also 
accommodates the use of substitute mortality tables by plans with 
smaller populations that have partially credible mortality experience. 
Comments are requested regarding additional simplifications that might 
be appropriate for use in developing substitute mortality tables.

I. Generally Applicable Mortality Tables

A. Base mortality tables

    The base mortality tables proposed for use under section 
430(h)(3)(A) are derived from the tables contained in the RP-2014 
Mortality Tables Report. In response to Notice 2013-49, commentators 
generally recommended that the RP-2014 mortality tables form the basis 
for the mortality tables used under section 430.\6\ After reviewing the 
RP-2014 mortality tables, the accompanying report published by the 
Society of Actuaries, and related public comments, Treasury and the IRS 
have determined that the experience study used to develop the RP-2014 
mortality tables is the best available study of the actual mortality 
experience of pension plan participants (other than disabled 
individuals). Accordingly, the RP-2014 mortality tables are the 
foundation for the base mortality tables used to project the mortality 
of pension plan participants under these proposed regulations.\7\
---------------------------------------------------------------------------

    \6\ These proposed regulations also apply the new generally 
applicable mortality tables under section 430 for purposes of 
determining the current liability of a multiemployer plan pursuant 
to section 431(c)(6)(D)(iv)(II) or a CSEC plan pursuant to section 
433(h)(3).
    \7\ Mortality tables that may be used as an alternative to the 
tables provided in these regulations with respect to certain 
disabled individuals are provided in Rev. Rul. 96-7, 1996-1 CB 59.
---------------------------------------------------------------------------

    Like the mortality tables provided in the 2008 general mortality 
table regulations, the mortality tables set forth in these proposed 
regulations are gender-distinct because of significant differences 
between expected male mortality and expected female mortality. In 
addition, as under the 2008 general mortality table regulations, these 
proposed regulations set forth separate mortality rates for annuitants 
and nonannuitants. This distinction has been made because these two 
groups have significantly different mortality experience. See chapter 3 
of the RP-2000 Mortality Tables Report, available at www.soa.org/research/experience-study/pension/research-rp-2000-mortality-tables.aspx.
    Under these proposed regulations, the annuitant mortality tables 
are applied to determine the present value of benefits for an 
annuitant. For a nonannuitant, the nonannuitant mortality tables are 
applied for the periods before the participant is projected to commence 
receiving benefits, and the annuitant mortality tables are used for 
later periods. With respect to a beneficiary of a participant, the 
annuitant mortality table applies for the period beginning with each 
assumed commencement of benefits for the participant. If the 
participant has died (or to the extent the participant is assumed to 
die before commencing benefits), the annuitant mortality table applies 
with respect to the beneficiary for the period beginning with each 
assumed commencement of benefits for the beneficiary.
    The proposed regulations set forth base tables that are to be used 
to develop the mortality tables for future years. These base tables 
have a base year of 2006 (the central year of the experience study used 
to develop the mortality tables in the RP-2014 Mortality Tables 
Report). These base tables generally have the same rates as the RP-2014 
mortality tables after factoring out the mortality improvements from 
2007 to 2014 (calculated using the Scale MP-2014 Rates). However, these 
base tables also include nonannuitant rates for ages below age 18 and 
above age 80 and annuitant rates for ages below age 50. This generally 
is the same approach that was used to develop the base tables included 
in the 2008 general mortality table regulations.
    The nonannuitant rates for ages above age 80 were developed by (1) 
using the annuitant rates from the base tables for ages 90 and older 
and (2) interpolating between the rates for age 80 and age 90 in order 
to produce a smooth transition between the age 80 rates from the 
nonannuitant tables to the age 90 rates from the annuitant tables. The 
interpolation uses increasing fractions

[[Page 95915]]

with a denominator of 55 to allocate the total difference between the 
rates at ages 80 and 90 over those 10 years. Thus, the rate at age 81 
is set equal to the rate at age 80 plus \1/55\ of the total difference, 
the age 82 rate is equal to the rate at age 81 plus \2/55\ of the total 
difference (so that the age 82 rate is equal to the rate at age 80 plus 
\3/55\ of the total difference), and so on for other ages.
    A similar approach was used to develop annuitant rates for ages 
below age 50. The annuitant rates for ages under age 50 were determined 
by (1) using the nonannuitant rates from the base tables for ages 18 to 
40, and (2) interpolating between the rates for age 40 and age 50, 
using the same methodology described in the prior paragraph. This 
method produces a smooth transition between the age 40 rates from the 
nonannuitant table and the age 50 rates from the annuitant table. For 
ages below age 18, both the annuitant and nonannuitant rates 
incorporate the juvenile rates from the RP-2014 Mortality Tables 
Report, after factoring out the mortality improvements from 2007 to 
2014 (calculated using the Scale MP-2014 Rates).

B. Reflection of Mortality Improvement

    The proposed regulations provide that expected trends in mortality 
experience must be taken into account through the use of either 
generational or annually updated static mortality tables. In accordance 
with section 430(h)(3)(B), the proposed regulations update the 
mortality improvement rates from the Scale AA Projection Factors that 
were set forth under the 2008 general mortality table regulations.
    In order to select up-to-date mortality improvement rates, Treasury 
and the IRS reviewed the Mortality Improvement Scale MP-2014 Report, 
related public comments, the data sources cited in those comments, the 
Mortality Improvement Scale MP-2015 Report, the Mortality Improvement 
Scale MP-2016 Report, and other published data sources.\8\ Pursuant to 
this review, Treasury and the IRS determined that the procedures that 
RPEC used to develop the Scale MP-2016 Rates generate the most 
appropriate currently available mortality improvement rates. 
Accordingly, the proposed regulations provide that, for valuation dates 
in 2018, the mortality tables for use under section 430(h)(3)(A) must 
reflect the mortality improvement rates contained in the Mortality 
Improvement Scale MP-2016 Report.
---------------------------------------------------------------------------

    \8\ See the August 2013 Literature Review and Assessment of 
Mortality Improvement Rates in the U.S. Population: Past Experience 
and Future Long-Term Trends, available at www.soa.org/Files/Research/Exp-Study/research-2013-lit-review.pdf; and the 2015 
Technical Panel on Assumptions and Methods Report to the Social 
Security Advisory Board, available at www.ssab.gov/Details-Page/ArticleID/656/2015-Technical-Panel-on-Assumptions-and-Methods-A-Report-to-the-Board-September-2015.
---------------------------------------------------------------------------

    The Scale MP-2016 Rates are structured as two-dimensional tables 
that contain mortality improvement rates that vary according to both 
age and calendar year (so that the mortality improvement rate for 
someone who is age 72 in 2020 is different than the mortality 
improvement rate for someone who is age 72 in 2030). RPEC provided for 
two-dimensional tables of mortality improvement rates in order to 
reflect differences in mortality improvement at different ages as well 
as mortality improvement trends that vary for different age cohorts. 
The proposed regulations include numerical examples illustrating how to 
apply these two-dimensional mortality improvement rates.
    As under the current regulations, the proposed regulations take 
into account the limitations of some current actuarial software that is 
not designed to use generational mortality tables. Accordingly, the 
proposed regulations continue to permit the use of static mortality 
tables. These static tables consist of a single table for each gender, 
updated annually, that approximates the effect of projected mortality 
improvement under the generational mortality tables. The static 
mortality tables that would be used for 2018 are included in these 
proposed regulations. For later years, updated static mortality tables 
will be set forth in guidance published in the Internal Revenue 
Bulletin. See Sec.  601.601(d)(2)(ii)(b) of this chapter.
    The static mortality tables that would be permitted to be used 
under the proposed regulations are constructed from the base tables 
that are used for purposes of the generational mortality tables. For 
each calendar year, the static mortality tables are based on a 
projection of mortality improvement applied to the mortality rates in 
the base tables for the period beginning with 2006 and ending with the 
year of the table, with a further projection from that year for a 
specified projection period. The rates in the static mortality tables 
are not the expected mortality rates for the current plan year, nor are 
they the mortality rates under the generational mortality tables that 
would apply for any current age. Instead, the projection period has 
been selected so that the use of the static mortality tables to 
calculate present values produces approximately the same results as 
would be calculated using the generational tables. Based on modeling of 
annuity values at different ages, Treasury and the IRS have selected a 
projection period of 8 years for males and 9 years for females, with a 
further adjustment based on age. For ages below 80, the projection 
period is increased by 1 year for each year below 80. For ages above 
80, the projection period is reduced (but not below zero) by \1/3\ year 
for each year above 80.
    These proposed regulations provide an option for smaller plans 
(plans for which the total number of active and inactive participants 
and beneficiaries of deceased participants is not more than 500 on the 
valuation date for the plan year) to use gender distinct blended static 
tables for all participants and beneficiaries--in lieu of the separate 
static tables for annuitants and nonannuitants--in order to simplify 
the actuarial valuation for these plans. These blended tables are 
constructed from the separate nonannuitant and annuitant static 
mortality tables using the same nonannuitant and annuitant weighting 
factors as in the 2008 general mortality table regulations.
    Treasury and the IRS understand that RPEC expects to issue updated 
mortality improvement rates that reflect new data for mortality 
improvement trends for the general population on an annual basis. 
Treasury and the IRS expect to take those updates into account in 
determining the mortality rates to be used under section 430(h)(3) for 
valuation dates in years after 2018. Those rates will be specified in 
guidance to be published in the Internal Revenue Bulletin. See Sec.  
601.601(d)(2)(ii)(b) of this chapter.

II. Plan-Specific Substitute Mortality Tables

A. Overview

    These proposed regulations contain a comprehensive set of rules 
regarding plan-specific substitute mortality tables. The proposed 
regulations contain many of the rules regarding substitute mortality 
tables from the 2008 substitute mortality table regulations. However, 
after analyzing the actuarial literature regarding credibility theory, 
Treasury and the IRS propose to make a number of changes to the 
regulations relating to the development of substitute mortality tables. 
Specifically, the proposed regulations would require a substitute 
mortality table to be constructed by multiplying the mortality rates 
from a projected version of the generally applicable base mortality 
table by a mortality ratio (that is, a ratio of the actual deaths for 
the plan population to

[[Page 95916]]

expected deaths determined using the standard mortality tables for that 
population).
    Use of mortality ratios (rather than providing for the graduation 
of raw mortality rates as under the 2008 substitute mortality table 
regulations) should make it easier for plan sponsors to develop the 
substitute tables, because it would eliminate the need to apply a 
graduation technique. It would also make it easier for the IRS to 
review applications to use substitute mortality tables. This 
simplification is particularly important in light of the other major 
change made in the proposed regulations, which would permit the use of 
substitute mortality tables for a plan that has only partially credible 
mortality information. Treasury and the IRS expect significantly more 
plan sponsors to request the use of substitute mortality tables after 
this change becomes effective.

B. Development of Substitute Mortality Tables for Plans With Full 
Credibility

    The substitute mortality table for a population with full 
credibility would be determined by applying projected mortality 
improvement to a base substitute mortality table which is developed 
using an experience study of the population. The proposed regulations 
would use the same requirements for an experience study as under the 
2008 substitute mortality table regulations. Specifically, the 
experience study would have to cover a period of at least 2 years (and 
no more than 5 years) that ends less than 3 years before the first day 
of the first plan year for which the substitute mortality tables are to 
apply. As under the 2008 substitute mortality table regulations, the 
calendar year that contains the day before the midpoint of the 
experience study is the base year for the base substitute mortality 
table. In addition, the proposed regulations include the rule in the 
2008 substitute mortality table regulations that requires an additional 
demonstration that the experience study results are predictive of 
future mortality for a plan population if the number of individuals in 
that population has changed by more than 20 percent compared to the 
average number of individuals in that population during the experience 
study period.
    The base substitute mortality table is determined by multiplying 
the mortality rates from the standard mortality table (that is, the 
generally applicable base mortality table projected with mortality 
improvement to the base year for the base substitute mortality table) 
by the plan's mortality ratio. For this purpose, the mortality 
improvement rates that apply for the calendar year during which the 
plan sponsor submits the request to use substitute mortality tables are 
used to project the generally applicable base mortality table to the 
base year for the base substitute mortality table.\9\ The mortality 
ratio is determined as a fraction, the numerator of which is the number 
of actual deaths during the experience study period (with each death 
weighted by the amount of benefit) and the denominator of which is the 
number of expected deaths during that period (determined using the 
standard mortality table) weighted by the amount of the benefit. For 
this purpose, the amount of benefit is the accrued benefit 
(substituting the current periodic payment in the case of individuals 
in pay status). Consistent with section 503 of the Bipartisan Budget 
Act of 2015 (and unlike Sec.  1.430(h)(3)-2(c)(2)(ii)(D) of the 2008 
substitute mortality table regulations, which provides that the 
Commissioner may permit the use of other recognized mortality tables to 
construct the base substitute mortality table), these proposed 
regulations provide that the standard mortality table that must be used 
for this purpose is the generally applicable base mortality table 
projected with mortality improvement to the base year for the base 
substitute mortality table.
---------------------------------------------------------------------------

    \9\ If the plan sponsor submits such a request during 2017, then 
the cumulative mortality improvement factors are determined using 
the Scale MP-2016 Rates.
---------------------------------------------------------------------------

C. Standards for Full Credibility

    The proposed regulations revise the standard for full credibility 
of a population under the 2008 substitute mortality table regulations 
(which is 1,000 actual deaths for the relevant population during the 
experience study period). This is because, under established actuarial 
credibility theory, that threshold (which is a rounding down of the 
1,082 actual deaths that would be needed for a 90% confidence level 
that the measured rate is within 5% of the underlying rate of 
mortality) should apply to the credibility for a single rate of 
mortality and not an entire mortality table.\10\ Moreover, the 1,000 
death threshold did not take into account the well-established 
actuarial principle that mortality experience within a population will 
vary predictably based on the amount of the annuity (or life insurance, 
as applicable). The base tables for the generally applicable mortality 
tables were constructed on an amounts-weighted basis (under which the 
individuals with higher benefit amounts have a greater weight in the 
computation of the mortality rate for a particular age); accordingly, 
substitute mortality tables should be constructed using the same 
principle.
---------------------------------------------------------------------------

    \10\ Note, however, the use of a graduation technique set forth 
in Rev. Proc. 2008-62 enables a plan to have credible mortality 
experience in order to establish a substitute mortality table even 
though there are fewer than 1000 deaths at each age.
---------------------------------------------------------------------------

    The variability of benefit amounts for different individuals in 
different populations within a plan means that a single 1,000 actual-
death standard that would apply to all populations is not appropriate. 
Instead, established actuarial credibility theory would require a plan-
specific calculation of the full-credibility standard that takes into 
account the dispersion of benefits within the plan.
    Under the proposed regulations, the number of deaths that are 
needed for the population within a plan to have fully credible 
mortality information is determined as the product of 1,082 and the 
benefit dispersion factor for the population.\11\ The benefit 
dispersion factor for a population is equal to the number of expected 
deaths for the population during the experience study period, times the 
sum of the mortality-weighted square of the benefits, divided by the 
square of the mortality-weighted benefits.\12\
---------------------------------------------------------------------------

    \11\ This is based on the assumption that the distribution of 
releases from liability due to deaths follows a compound Poisson 
model. See www.actuaries.ca/members/publications/2002/202037e.pdf.
    \12\ See Gavin Benjamin, Selecting Mortality Tables: A 
Credibility Approach, available at www.soa.org/Files/Research/Projects/research-2008-benjamin.pdf.
---------------------------------------------------------------------------

D. Partial Credibility

    The proposed regulations permit substitute mortality tables to be 
used for a plan that does not have sufficient deaths to have fully 
credible mortality information. In accordance with established 
actuarial credibility theory, such a plan would use a weighted average 
of the standard mortality table (projected with mortality improvement 
to the base year of the base substitute mortality table) and the 
mortality table that would be developed for the plan if it were to have 
fully credible mortality information. The weight for the mortality 
table that would apply if the plan were to have fully credible 
mortality information is the square-root of a fraction, the numerator 
of which is the actual number of deaths for the population within the 
experience study period and the denominator of which is the number of 
deaths needed for the

[[Page 95917]]

plan to have fully credible mortality information.
    In order to avoid the need to create a substitute mortality table 
for a plan with a relatively small population, the proposed regulations 
provide that a population does not have credible mortality information 
if the actual number of deaths for that population during the 
experience study period is less than 100. For this purpose, the length 
of the experience study period must be the same length as the longest 
experience study period for any plan in the controlled group.
    Treasury and the IRS chose the threshold of 100 deaths as a result 
of balancing the burdens of developing substitute mortality tables and 
the benefit of the use of those tables, in light of the requirement 
under section 430(h)(3)(C)(iv) that substitute mortality tables be used 
for all plans within a controlled group (and the exception to this 
requirement for plans that lack fully or partially credible mortality 
information). Comments are requested regarding whether this is the 
appropriate threshold or whether a different number of deaths should be 
used for this purpose.

E. Mortality Improvement Rates

    As required under the 2008 substitute mortality table regulations, 
the proposed regulations provide that substitute mortality tables must 
be generational mortality tables. These proposed regulations require 
that the mortality improvement rates that are used for the generally 
applicable mortality tables also be applied beginning with the base 
year of the base substitute mortality tables.

F. Other Rules Relating to the Use of Substitute Mortality Tables

1. Use of Separate Subpopulations Within a Gender Under Plan
    The proposed regulations continue to apply the rules under the 2008 
substitute mortality table regulations regarding the applicability of 
substitute mortality tables for separate populations within a plan. 
Specifically, separate substitute mortality tables must be developed 
for each gender under the plan. In addition, the regulations permit 
separate substitute mortality tables to be developed for separate 
subpopulations (such as hourly and salaried participants) within a 
gender under the plan in certain circumstances.
    As under the 2008 substitute mortality table regulations, 
permission to separate a gender into separate subpopulations is 
generally limited to situations in which each of the subpopulations 
have fully credible mortality information. However, that requirement 
does not apply if the separate subpopulations are annuitants and non-
annuitants. Comments are requested on whether there should be other 
exceptions to this rule. For example:
     Should the regulations allow separate sub-populations to 
be used if one subpopulation has full credibility while the other one 
has only partial credibility?
     Should the regulations provide for the use of separate 
sub-populations based on age, even if those groups have only partial 
credibility?
     Should there be a rule to ``normalize'' the mortality 
tables for separate sub-populations (so that the total number of 
expected deaths for the separate subpopulations is the same as the 
total number of expected deaths for the entire population without 
regard to the separation)?
2. Requirement To Use Substitute Mortality Tables for All Plans With 
Credible Mortality Information
    As under the 2008 substitute mortality table regulations, the 
proposed regulations provide that substitute mortality tables are 
permitted to be used for a plan for a plan year only if, for that plan 
year, substitute mortality tables are also approved and used for each 
other pension plan subject to the requirements of section 430 that is 
maintained by the plan sponsor or by a member of the sponsor's 
controlled group. However, this rule does not prohibit the use of 
substitute mortality tables for one plan if the only other plan or 
plans maintained by the plan sponsor (or by a member of the plan 
sponsor's controlled group) for which substitute mortality tables are 
not used are too small to have fully or partially credible mortality 
information for the plan year. Thus, if a sponsor's controlled group 
contains two pension plans that are subject to section 430, each of 
which has fully or partially credible mortality information for at 
least one gender, either the plan sponsors of both plans must obtain 
approval from the Commissioner to use substitute mortality tables or 
substitute mortality tables may not be used for either plan. In 
contrast, if for one of those plans neither males nor females have 
fully or partially credible mortality information, then the plan 
without credible mortality information will not prevent the use of 
substitute mortality tables for the plan with credible mortality 
information.
    As under the 2008 substitute mortality table regulations, the 
proposed regulations provide that the requirement that the plan sponsor 
demonstrate the lack of credible mortality information for both the 
male and female populations in other plans maintained by the plan 
sponsor (and by members of the plan sponsor's controlled group) for 
which substitute mortality tables are not used must be satisfied for 
each plan year for which substitute mortality tables are used. This 
demonstration is made for a plan population by showing that the 
population has not experienced at least 100 deaths over a time period 
that satisfies the requirements set forth in the regulations. In 
general, for each plan year in which substitute mortality tables are 
used for a plan, in order to demonstrate that a gender within a plan 
does not have credible mortality information for a plan year, the 
demonstration that the gender within the plan has fewer than 100 deaths 
must be made by analyzing the actual number of deaths over a period 
that is the same length as the period for the experience study on which 
the substitute mortality tables are based and that ends less than three 
years before the first day of the plan year.
3. Permitted Aggregation of Plans
    The proposed regulations retain the rules contained in the 2008 
substitute mortality table regulations regarding aggregation of plans 
for purposes of using substitute mortality tables. Under these rules, 
in order for a plan sponsor to use the same substitute mortality tables 
for two or more plans, the rules set forth in the regulations are 
applied by treating those plans as a single plan. In such a case, the 
substitute mortality tables must be used for all such plans and must be 
based on data collected with respect to all such plans. Although plans 
generally are not required to be aggregated for purposes of substitute 
mortality tables, the regulations require a plan to be aggregated with 
any plan that was previously spun off from that plan if the 
Commissioner determines that one purpose of the spinoff was to avoid 
the use of substitute mortality tables for any of the plans involved in 
the spinoff.
4. Special Rules for Newly-Acquired Plans
    If substitute mortality tables are used for at least one plan 
within a controlled group, in order for the plan sponsor to continue to 
use substitute mortality tables for that plan after a plan joins the 
controlled group, substitute mortality tables must be used for the 
newly affiliated plan unless the newly affiliated plan demonstrates 
that it lacks credible mortality information. However, the proposed 
regulations

[[Page 95918]]

provide for a transition period during which the standard mortality 
table is permitted to be used for a newly affiliated plan (without 
affecting the use of substitute mortality tables for other plans within 
the controlled group) even if the newly affiliated plan fails to 
demonstrate a lack of credible mortality information. Similarly, the 
use of substitute mortality tables for a newly affiliated plan is not 
affected during the transition period merely because the standard 
mortality tables are used for another plan within the controlled group 
despite the failure of that other plan to demonstrate a lack of 
credible mortality information. Notably, these rules do not change the 
requirement that the continued use of substitute mortality tables for 
any plan within the controlled group is permitted only if the other 
pre-affiliation plans within the controlled group for which substitute 
mortality tables are not used demonstrate a lack of credible mortality 
information.
    Like the 2008 substitute mortality table regulations, the proposed 
regulations do not require the use of pre-affiliation experience in 
order to establish whether a newly-affiliated plan has credible 
mortality information. If the pre-affiliation data is excluded and 
substitute mortality tables will be used for the plan, then the 
experience study period may be as short as one year (instead of two 
years). If the pre-affiliation data is excluded and substitute 
mortality tables will not be used for the plan, then the experience 
study period used to demonstrate that the plan does not have credible 
mortality information may also be shortened, provided that the period 
ends not more than one year and one day before the first day of the 
plan year.
5. Treatment of Mortality Experience With Respect to Disabled 
Individuals
    As under the 2008 substitute mortality table regulations, if 
separate mortality tables under section 430(h)(3)(D) are used for 
certain disabled individuals under a plan, then those individuals are 
disregarded for all purposes with respect to substitute mortality 
tables under section 430(h)(3)(C). Thus, if the mortality tables under 
section 430(h)(3)(D) are used for certain disabled individuals under a 
plan, mortality experience with respect to those individuals must be 
excluded in determining mortality rates for substitute mortality tables 
with respect to a plan.
6. Early Termination of Use of Substitute Mortality Tables
    The proposed regulations retain the rules from the 2008 substitute 
mortality table regulations regarding the early termination of use of 
substitute mortality tables. Under those rules, a plan's substitute 
mortality tables may not be used beginning with the earliest of: (1) 
For a plan for which substitute mortality tables are used for only one 
gender because of a lack of credible mortality information with respect 
to the other gender, the first plan year for which there is credible 
mortality information with respect to the gender that had lacked 
credible mortality information (unless the plan receives approval to 
use a substitute mortality table for that other gender); (2) the first 
plan year for which the requirements regarding use of substitute 
mortality tables by controlled group members are not satisfied; (3) the 
second plan year following the plan year for which there is a 
significant change in individuals covered by the plan (unless the 
plan's actuary certifies in writing to the satisfaction of the 
Commissioner that the substitute mortality tables used for the plan 
population continue to be accurately predictive of future mortality of 
that population (taking into account the effect of the change in the 
population)); (4) the first plan year following the plan year for which 
a substitute mortality table used for a plan population is no longer 
accurately predictive of future mortality of that population, as 
determined by the Commissioner or as certified by the plan's actuary to 
the satisfaction of the Commissioner; or (5) the date specified in 
guidance published in the Internal Revenue Bulletin in conjunction with 
a replacement of generally applicable mortality tables (other than 
annual updates to the static mortality tables or changes to the 
mortality improvement rates).
G. Procedures for Requesting Approval of Substitute Mortality Tables
    As under the 2008 substitute mortality table regulations, the 
proposed regulations provide that a plan sponsor that wishes to use 
substitute mortality tables for a plan must submit a request to the IRS 
for approval to use the proposed tables. In general, the request must 
be submitted at least 7 months before the first day of the plan year 
for which the proposed substitute tables would be used. If the IRS does 
not deny the request within 180 days (which may be extended as agreed 
to by the IRS and the plan sponsor), the request is deemed to have been 
approved.
    The IRS intends to issue an updated version of Rev. Proc. 2008-62 
after final regulations regarding substitute mortality tables are 
issued. If the timing of the release of those final regulations and the 
associated revenue procedure does not leave adequate time to submit an 
application to use substitute mortality tables for the plan year 
beginning in 2018, Treasury and the IRS expect that they would provide 
a transition rule that would permit extra time to submit such an 
application
    Before final regulations adopting the provisions set forth in these 
proposed regulations are issued, plan sponsors requesting the use of 
substitute mortality tables should continue to use the procedures set 
forth in Rev. Proc. 2008-62. During that period, the IRS will not 
evaluate whether a substitute mortality table for a population with 
only partially credible mortality information is appropriate.
Applicability Date
    These regulations are proposed to apply to plan years beginning on 
or after January 1, 2018. Under the proposed regulations, a plan 
sponsor may use a substitute mortality table for a plan year beginning 
on or after January 1, 2018 only if that substitute mortality table is 
approved as provided in these proposed regulations.
Statement of Availability of IRS Documents
    IRS Revenue Rulings, Revenue Procedures, and Notices cited in this 
document are published in the Internal Revenue Bulletin (or Cumulative 
Bulletin) and are available from the Superintendent of Documents, U.S. 
Government Printing Office, Washington, DC 20402, or by visiting the 
IRS Web site at www.irs.gov.
Special Analyses
    Certain IRS regulations, including this one, are exempt from the 
requirements of Executive Order 12866, as supplemented and reaffirmed 
by Executive Order 13563. Therefore, a regulatory assessment is not 
required. The regulations do not impose a collection of information on 
small entities, therefore the Regulatory Flexibility Act (5 U.S.C. 
chapter 6) does not apply. Pursuant to section 7805(f) of the Internal 
Revenue Code, this notice of proposed rulemaking will be submitted to 
the Chief Counsel for Advocacy of the Small Business Administration for 
comment on their impact on small business.
Comments and Public Hearing
    Before these proposed regulations are adopted as final regulations, 
consideration will be given to any comments that are submitted timely 
to Treasury and the IRS as prescribed in this preamble in the ADDRESSES 
section.

[[Page 95919]]

Treasury and the IRS request comments on all aspects of these proposed 
regulations. All comments will be available for public inspection and 
copying at www.regulations.gov or upon request.
    A public hearing on these proposed regulations has been scheduled 
for April 13, 2017 beginning at 10 a.m. in the Auditorium, Internal 
Revenue Service, 1111 Constitution Avenue NW., Washington, DC 20224. 
Due to building security procedures, visitors must enter at the 
Constitution Avenue entrance. In addition, all visitors must present 
photo identification to enter the building. Because of access 
restrictions, visitors will not be admitted beyond the immediate 
entrance area more than 30 minutes before the hearing starts. For 
information about having your name placed on the building access list 
to attend the hearing, see the FOR FURTHER INFORMATION CONTACT section 
of this preamble.
    The rules of 26 CFR 601.601(a)(3) apply to the hearing. Persons who 
wish to present oral comments at the hearing must submit written or 
electronic comments by March 29, 2017, and an outline of topics to be 
discussed and the amount of time to be devoted to each topic by March 
29, 2017. A period of 10 minutes will be allotted to each person for 
making comments. An agenda showing the scheduling of the speakers will 
be prepared after the deadline for receiving outlines has passed. 
Copies of the agenda will be available free of charge at the hearing.
Drafting Information
    The principal authors of these regulations are Thomas Morgan and 
Linda S. F. Marshall of Office of Associate Chief Counsel (Tax Exempt 
and Government Entities). However, other personnel from Treasury and 
the IRS participated in the development of these regulations.

List of Subjects in 26 CFR Part 1

    Income taxes, Reporting and recordkeeping requirements.

Amendments to the Regulations

    Accordingly, 26 CFR part 1 is proposed to be amended as follows:

PART 1--INCOME TAXES

0
Paragraph 1. The authority citation for part 1 continues to read, in 
part, as follows:

    Authority: 26 U.S.C. 7805 * * *

0
Par. 2. Section 1.430(h)(3)-1 is revised to read as follows:


Sec.  1.430(h)(3)-1  Mortality tables used to determine present value.

    (a) Basis for mortality tables--(1) In general. This section sets 
forth rules for the mortality tables to be used in determining present 
value or making any computation under section 430. Generally applicable 
mortality tables for participants and beneficiaries are set forth in 
this section pursuant to section 430(h)(3)(A). In general, either the 
generational mortality tables set forth in paragraph (a)(2) of this 
section or the static mortality tables set forth in paragraph (a)(3) of 
this section must be used for a plan. In lieu of using the mortality 
tables provided under this section with respect to participants and 
beneficiaries, plan-specific substitute mortality tables are permitted 
to be used for this purpose pursuant to section 430(h)(3)(C), provided 
that the requirements of Sec.  1.430(h)(3)-2 are satisfied. Mortality 
tables that may be used with respect to disabled individuals are to be 
provided in guidance published in the Internal Revenue Bulletin. See 
Sec.  601.601(d)(2)(ii)(b) of this chapter.
    (2) Generational mortality tables--(i) In general--(A) Use of 
generational mortality tables. The generational mortality tables that 
are permitted to be used under section 430(h)(3)(A) and paragraph 
(a)(1) of this section are determined using the base mortality tables 
described in paragraph (a)(2)(i)(B) of this section and the mortality 
improvement rates described in paragraph (a)(2)(i)(C) of this section.
    (B) Base mortality tables. The base mortality tables are set forth 
in paragraph (d) of this section. The base year for those tables is 
2006.
    (C) Mortality improvement rates. The mortality improvement rates 
for valuation dates occurring during 2018 are the mortality improvement 
rates contained in the Mortality Improvement Scale MP-2016 Report 
(issued by the Retirement Plans Experience Committee (RPEC) of the 
Society of Actuaries and available at www.soa.org/Research/Experience-Study/Pension/research-2016-mp.aspx). For later years, updated 
mortality improvement rates that take into account new data for 
mortality improvement trends of the general population are to be 
provided in guidance published in the Internal Revenue Bulletin. See 
Sec.  601.601(d)(2)(ii)(b) of this chapter.
    (D) Application of mortality improvement rates. Under the 
generational mortality tables described in this paragraph (a)(2), the 
probability of an individual's death at a particular age in the future 
is determined as the individual's base mortality rate that applies at 
that age (that is, the applicable mortality rate from the table set 
forth in paragraph (d) of this section for that age, gender, and status 
as an annuitant or a nonannuitant) multiplied by the cumulative 
mortality improvement factor for the individual's gender and for that 
age for the period from 2006 through the calendar year in which the 
individual is projected to reach the particular age. Paragraph 
(a)(2)(ii) of this section shows how the base mortality tables in 
paragraph (d) of this section and the mortality improvement rates for 
valuation dates occurring during 2018 are combined to determine 
projected mortality rates.
    (E) Cumulative mortality improvement factor. The cumulative 
mortality improvement factor for an age and gender for a period is the 
product of the annual mortality improvement factors for that age and 
gender for each year within that period.
    (F) Annual mortality improvement factor. The annual mortality 
improvement factor for an age and gender for a year is 1 minus the 
mortality improvement rate that applies for that age and gender for 
that year.
    (ii) Example of calculation--(A) Calculation of mortality rate. The 
mortality rate for 2018 that is applied to male annuitants who are age 
66 in 2018 is equal to the product of the mortality rate for 2006 that 
applied to male annuitants who were age 66 in 2006 (0.013855) and the 
cumulative mortality improvement factor for age 66 males from 2006 to 
2018. The cumulative mortality improvement factor for age 66 males for 
the period from 2006 to 2018 is 0.8929, and the mortality rate for 2018 
for male annuitants who are age 66 in that year would be 0.012371, as 
shown in the following table.

[[Page 95920]]



----------------------------------------------------------------------------------------------------------------
                                                                      Annual
                                                   Scale MP-2016     mortality      Cumulative
                                                     mortality      improvement      mortality
                  Calendar year                     improvement     factor (1-      improvement   Mortality rate
                                                       rate        scale MP-2016      factor
                                                                       rate)
----------------------------------------------------------------------------------------------------------------
2006............................................             n/a             n/a             n/a        0.013855
2007............................................          0.0237          0.9763          0.9763
2008............................................          0.0211          0.9789          0.9557
2009............................................          0.0180          0.9820          0.9385
2010............................................          0.0142          0.9858          0.9252
2011............................................          0.0099          0.9901          0.9160
2012............................................          0.0053          0.9947          0.9112
2013............................................          0.0043          0.9957          0.9072
2014............................................          0.0035          0.9965          0.9041
2015............................................          0.0030          0.9970          0.9014
2016............................................          0.0028          0.9972          0.8988
2017............................................          0.0030          0.9970          0.8961
2018............................................          0.0036          0.9964          0.8929        0.012371
----------------------------------------------------------------------------------------------------------------

    (B) Probability of survival for an individual. After the projected 
mortality rates are derived for each age for each year, the rates are 
used to calculate the present value of a benefit stream that depends on 
the probability of survival year-by-year. For example, for purposes of 
calculating the present value (for a 2018 valuation date) of future 
payments in a benefit stream payable for a male annuitant who is age 66 
in 2018, the probability of survival for the annuitant is based on the 
mortality rate for a male annuitant who is age 66 in 2018 (0.012371), 
and the projected mortality rate for a male annuitant who will be age 
67 in 2019 (0.013302), age 68 in 2020 (0.014321), and so on.
    (3) Static mortality tables. The static mortality tables that are 
permitted to be used under section 430(h)(3)(A) and paragraph (a)(1) of 
this section are updated annually by the IRS according to the 
methodology described in paragraph (c)(2) of this section. Paragraph 
(e) of this section sets forth static tables that are permitted to be 
used for valuation dates in 2018. For valuation dates in later years, 
static mortality tables are to be provided in guidance published in the 
Internal Revenue Bulletin. See Sec.  601.601(d)(2)(ii)(b) of this 
chapter.
    (b) Use of the tables--(1) Separate tables for annuitants and 
nonannuitants--(i) In general. Separate tables are provided for use for 
annuitants and nonannuitants. The nonannuitant mortality table is 
applied to determine the probability of survival for a nonannuitant for 
the period before the nonannuitant is projected to commence receiving 
benefits. The annuitant mortality table is applied to determine the 
present value of benefits for each annuitant. In addition, the 
annuitant mortality table is applied for each nonannuitant with respect 
to each assumed commencement of benefits for the period beginning with 
that assumed commencement. For purposes of this section, an annuitant 
means a plan participant who has commenced receiving benefits and a 
nonannuitant means a plan participant who has not yet commenced 
receiving benefits (for example, an active employee or a terminated 
vested participant). A participant whose benefit has partially 
commenced is treated as an annuitant with respect to the portion of the 
benefit that has commenced and treated as a nonannuitant with respect 
to the balance of the benefit. In addition, with respect to a 
beneficiary of a participant, the annuitant mortality table applies for 
the period beginning with each assumed commencement of benefits for the 
participant. If the participant has died (or to the extent the 
participant is assumed to die before commencing benefits), the 
annuitant mortality table applies with respect to the beneficiary for 
the period beginning with each assumed commencement of benefits for the 
beneficiary.
    (ii) Examples of calculation using separate annuitant and 
nonannuitant tables. With respect to a 45-year-old active participant 
who is projected to commence receiving an annuity at age 55, the 
funding target is determined using the nonannuitant mortality table for 
the period before the participant attains age 55 (so that, if the 
static mortality tables are used pursuant to paragraph (a)(3) of this 
section, the probability of an active male participant living from age 
45 to age 55 using the table that applies for a valuation date in 2018 
is 0.988857) and using the annuitant mortality table for the period 
ages 55 and above. Similarly, for a 45-year-old terminated vested 
participant who is projected to commence an annuity at age 65, the 
funding target is determined using the nonannuitant mortality table for 
the period before the participant attains age 65 and using the 
annuitant mortality table for ages 65 and above.
    (2) Small plan tables. If static mortality tables are used pursuant 
to paragraph (a)(3) of this section, as an alternative to the separate 
static tables specified for annuitants and nonannuitants pursuant to 
paragraph (b)(1) of this section, combined static tables that apply the 
same mortality rates to both annuitants and nonannuitants are permitted 
to be used for a small plan. For this purpose, a small plan is defined 
as a plan with 500 or fewer total participants (including both active 
and inactive participants and beneficiaries of deceased participants) 
on the valuation date. The combined static tables that are permitted to 
be used for small plans pursuant to this paragraph (b)(2) are 
constructed from the separate nonannuitant and annuitant static 
mortality tables using the weighting factors for small plans that are 
set forth in paragraph (d) of this section. The weighting factors are 
applied to develop these combined static tables using the following 
equation: Combined mortality rate = [nonannuitant rate * (1- weighting 
factor)] + [annuitant rate * weighting factor].
    (c) Static tables--(1) Source of rates. The static mortality tables 
that are used pursuant to paragraph (a)(3) of this section are 
determined using the base mortality tables described in paragraph 
(a)(2)(i)(B) of this section taking into account the mortality 
improvement rates described in paragraph (a)(2)(i)(C) of this section, 
in accordance with the rules set forth in paragraph (c)(3) of this 
section.
    (2) Selection of static tables. The static mortality tables that 
are used for a valuation date are the static mortality tables for the 
calendar year that contains the valuation date.

[[Page 95921]]

    (3) Projection of mortality improvements--(i) General rule. Except 
as provided in paragraph (c)(3)(iii) of this section, the static 
mortality tables for a calendar year are determined by multiplying the 
applicable mortality rate for each age from the base mortality tables 
by both--
    (A) The cumulative mortality improvement factor (determined under 
the rules of paragraph (a)(2) of this section) for the period from 2006 
through that calendar year; and
    (B) The cumulative mortality improvement factor (determined under 
the rules of paragraph (a)(2) of this section) for the period beginning 
in that calendar year and continuing beyond that calendar year for the 
number of years in the projection period described in paragraph 
(c)(3)(ii) of this section.
    (ii) Projection period for static mortality tables--(A) In general. 
The projection period is 8 years for males and 9 years for females, as 
adjusted based on age as provided in paragraph (c)(3)(ii)(B) of this 
section.
    (B) Age adjustment. For ages below 80, the projection period is 
increased by 1 year for each year below age 80. For ages above 80, the 
projection period is reduced (but not below zero) by \1/3\ year for 
each year above 80.
    (iii) Fractional projection periods. If for an age the number of 
years in the projection period determined under this paragraph (c)(3) 
is not a whole number, then the mortality rate for that age is 
determined by using linear interpolation between--
    (A) The mortality rate for that age that would be determined under 
paragraph (c)(3)(i) of this section if the number of years in the 
projection period were the next lower whole number; and
    (B) The mortality rate for that age that would be determined under 
paragraph (c)(3)(i) of this section if the number of years in the 
projection period were the next higher whole number.
    (iv) Example. For example, at age 85 the projection period for a 
male is 6\1/3\ years (8 years minus \1/3\ year for each of the 5 years 
above age 80). For a valuation date in 2018, the mortality rate in the 
static mortality table for an 85-year-old male is based on a projection 
of mortality improvement for 6\1/3\ years beyond 2018. Under paragraph 
(c)(3)(iii) of this section, the mortality rate for an 85-year-old male 
annuitant in the static mortality table for 2018 is \2/3\ times the 
projected mortality rate for a male annuitant that age in 2024 plus \1/
3\ times the projected mortality rate for a male annuitant that age in 
2025. Accordingly, the mortality rate for an 85-year-old male annuitant 
in the static mortality table for 2018 is 0.075196 (\2/3\ times the 
projected mortality rate for an 85-year old male annuitant in 2024 
(0.075447) plus \1/3\ times the projected mortality rate for an 85-year 
old male annuitant in 2025 (0.074693)).
    (d) Base mortality tables. The following are the base mortality 
tables. The base year for these tables is 2006.

----------------------------------------------------------------------------------------------------------------
                                                  Males                                    Females
----------------------------------------------------------------------------------------------------------------
                                                                  Weighting                            Weighting
                                            Non-                    factor       Non-                    factor
                  Age                    annuitant    Annuitant   for small   annuitant    Annuitant   for small
                                                                    plans                                plans
----------------------------------------------------------------------------------------------------------------
0.....................................     0.008878     0.008878          0     0.007278     0.007278          0
1.....................................     0.000515     0.000515          0     0.000451     0.000451          0
2.....................................     0.000348     0.000348          0     0.000295     0.000295          0
3.....................................     0.000289     0.000289          0     0.000220     0.000220          0
4.....................................     0.000225     0.000225          0     0.000165     0.000165          0
5.....................................     0.000197     0.000197          0     0.000149     0.000149          0
6.....................................     0.000177     0.000177          0     0.000137     0.000137          0
7.....................................     0.000156     0.000156          0     0.000127     0.000127          0
8.....................................     0.000132     0.000132          0     0.000117     0.000117          0
9.....................................     0.000107     0.000107          0     0.000109     0.000109          0
10....................................     0.000090     0.000090          0     0.000102     0.000102          0
11....................................     0.000095     0.000095          0     0.000105     0.000105          0
12....................................     0.000142     0.000142          0     0.000121     0.000121          0
13....................................     0.000187     0.000187          0     0.000137     0.000137          0
14....................................     0.000230     0.000230          0     0.000151     0.000151          0
15....................................     0.000274     0.000274          0     0.000165     0.000165          0
16....................................     0.000318     0.000318          0     0.000177     0.000177          0
17....................................     0.000364     0.000364          0     0.000187     0.000187          0
18....................................     0.000412     0.000412          0     0.000196     0.000196          0
19....................................     0.000463     0.000463          0     0.000202     0.000202          0
20....................................     0.000510     0.000510          0     0.000202     0.000202          0
21....................................     0.000552     0.000552          0     0.000197     0.000197          0
22....................................     0.000587     0.000587          0     0.000191     0.000191          0
23....................................     0.000599     0.000599          0     0.000190     0.000190          0
24....................................     0.000594     0.000594          0     0.000188     0.000188          0
25....................................     0.000545     0.000545          0     0.000186     0.000186          0
26....................................     0.000510     0.000510          0     0.000186     0.000186          0
27....................................     0.000486     0.000486          0     0.000188     0.000188          0
28....................................     0.000472     0.000472          0     0.000192     0.000192          0
29....................................     0.000468     0.000468          0     0.000198     0.000198          0
30....................................     0.000470     0.000470          0     0.000209     0.000209          0
31....................................     0.000480     0.000480          0     0.000222     0.000222          0
32....................................     0.000495     0.000495          0     0.000238     0.000238          0
33....................................     0.000514     0.000514          0     0.000257     0.000257          0
34....................................     0.000534     0.000534          0     0.000278     0.000278          0
35....................................     0.000557     0.000557          0     0.000301     0.000301          0
36....................................     0.000581     0.000581          0     0.000325     0.000325          0
37....................................     0.000611     0.000611          0     0.000355     0.000355          0
38....................................     0.000648     0.000648          0     0.000389     0.000389          0
39....................................     0.000694     0.000694          0     0.000428     0.000428          0
40....................................     0.000750     0.000750          0     0.000471     0.000471          0

[[Page 95922]]

 
41....................................     0.000814     0.000823      .0045     0.000518     0.000515          0
42....................................     0.000890     0.000969      .0091     0.000570     0.000603          0
43....................................     0.000982     0.001188      .0136     0.000628     0.000735          0
44....................................     0.001088     0.001480      .0181     0.000691     0.000911          0
45....................................     0.001207     0.001846      .0226     0.000758     0.001131      .0084
46....................................     0.001342     0.002285      .0272     0.000831     0.001395      .0167
47....................................     0.001487     0.002797      .0317     0.000908     0.001703      .0251
48....................................     0.001643     0.003382      .0362     0.000986     0.002055      .0335
49....................................     0.001807     0.004040      .0407     0.001065     0.002451      .0419
50....................................     0.001979     0.004771      .0453     0.001151     0.002891      .0502
51....................................     0.002159     0.005059      .0498     0.001242     0.002993      .0586
52....................................     0.002351     0.005343      .0686     0.001344     0.003124      .0744
53....................................     0.002539     0.005592      .0953     0.001458     0.003291      .0947
54....................................     0.002741     0.005839      .1288     0.001588     0.003499      .1189
55....................................     0.002967     0.006102      .2066     0.001735     0.003755      .1897
56....................................     0.003231     0.006399      .3173     0.001902     0.004065      .2857
57....................................     0.003548     0.006746      .3780     0.002091     0.004435      .3403
58....................................     0.003932     0.007155      .4401     0.002302     0.004869      .3878
59....................................     0.004396     0.007639      .4986     0.002537     0.005373      .4360
60....................................     0.004954     0.008211      .5633     0.002795     0.005942      .4954
61....................................     0.005616     0.008878      .6338     0.003080     0.006581      .5805
62....................................     0.006392     0.009646      .7103     0.003388     0.007283      .6598
63....................................     0.007291     0.010523      .7902     0.003724     0.008043      .7520
64....................................     0.008320     0.011514      .8355     0.004089     0.008870      .8043
65....................................     0.009486     0.012621      .8832     0.004482     0.009760      .8552
66....................................     0.010668     0.013855      .9321     0.005004     0.010731      .9118
67....................................     0.011973     0.015221      .9510     0.005575     0.011790      .9367
68....................................     0.013414     0.016736      .9639     0.006205     0.012952      .9523
69....................................     0.015006     0.018421      .9714     0.006898     0.014226      .9627
70....................................     0.016761     0.020288      .9740     0.007662     0.015628      .9661
71....................................     0.018690     0.022348      .9766     0.008507     0.017170      .9695
72....................................     0.020824     0.024638      .9792     0.009438     0.018861      .9729
73....................................     0.023176     0.027176      .9818     0.010470     0.020723      .9763
74....................................     0.025770     0.029992      .9844     0.011615     0.022780      .9797
75....................................     0.028623     0.033113      .9870     0.012887     0.025057      .9830
76....................................     0.031761     0.036585      .9896     0.014301     0.027590      .9864
77....................................     0.035214     0.040457      .9922     0.015885     0.030438      .9898
78....................................     0.039007     0.044778      .9948     0.017656     0.033653      .9932
79....................................     0.043169     0.049605      .9974     0.019639     0.037296      .9966
80....................................     0.047750     0.055022        1.0     0.021859     0.041440        1.0
81....................................     0.049804     0.061087        1.0     0.023791     0.046181        1.0
82....................................     0.053911     0.067902        1.0     0.027655     0.051564        1.0
83....................................     0.060072     0.075550        1.0     0.033451     0.057714        1.0
84....................................     0.068286     0.084162        1.0     0.041179     0.064709        1.0
85....................................     0.078554     0.093775        1.0     0.050838     0.072601        1.0
86....................................     0.090876     0.104507        1.0     0.062429     0.081490        1.0
87....................................     0.105251     0.116487        1.0     0.075952     0.091444        1.0
88....................................     0.121680     0.129770        1.0     0.091407     0.102470        1.0
89....................................     0.140162     0.144470        1.0     0.108794     0.114635        1.0
90....................................     0.160698     0.160698        1.0     0.128113     0.128113        1.0
91....................................     0.177741     0.177741        1.0     0.142619     0.142619        1.0
92....................................     0.195154     0.195154        1.0     0.157939     0.157939        1.0
93....................................     0.212642     0.212642        1.0     0.173886     0.173886        1.0
94....................................     0.230055     0.230055        1.0     0.190319     0.190319        1.0
95....................................     0.247257     0.247257        1.0     0.207191     0.207191        1.0
96....................................     0.265940     0.265940        1.0     0.225057     0.225057        1.0
97....................................     0.284940     0.284940        1.0     0.243507     0.243507        1.0
98....................................     0.304432     0.304432        1.0     0.262587     0.262587        1.0
99....................................     0.324272     0.324272        1.0     0.282171     0.282171        1.0
100...................................     0.344364     0.344364        1.0     0.302162     0.302162        1.0
101...................................     0.364420     0.364420        1.0     0.322282     0.322282        1.0
102...................................     0.384058     0.384058        1.0     0.342371     0.342371        1.0
103...................................     0.403188     0.403188        1.0     0.362210     0.362210        1.0
104...................................     0.421533     0.421533        1.0     0.381534     0.381534        1.0
105...................................     0.438903     0.438903        1.0     0.400321     0.400321        1.0
106...................................     0.455492     0.455492        1.0     0.418418     0.418418        1.0
107...................................     0.470810     0.470810        1.0     0.435390     0.435390        1.0
108...................................     0.484965     0.484965        1.0     0.451459     0.451459        1.0
109...................................     0.498023     0.498023        1.0     0.466408     0.466408        1.0
110...................................     0.509768     0.509768        1.0     0.480123     0.480123        1.0
111...................................     0.512472     0.512472        1.0     0.492664     0.492664        1.0

[[Page 95923]]

 
112...................................     0.509296     0.509296        1.0     0.503970     0.503970        1.0
113...................................     0.506193     0.506193        1.0     0.507361     0.507361        1.0
114...................................     0.503061     0.503061        1.0     0.503564     0.503564        1.0
115...................................     0.500000     0.500000        1.0     0.500000     0.500000        1.0
116...................................     0.500000     0.500000        1.0     0.500000     0.500000        1.0
117...................................     0.500000     0.500000        1.0     0.500000     0.500000        1.0
118...................................     0.500000     0.500000        1.0     0.500000     0.500000        1.0
119...................................     0.500000     0.500000        1.0     0.500000     0.500000        1.0
120...................................     1.000000     1.000000        1.0     1.000000     1.000000        1.0
----------------------------------------------------------------------------------------------------------------

    (e) Static tables for 2018. The following static mortality tables 
are used pursuant to paragraph (a)(3) of this section for determining 
present value or making any computation under section 430 with respect 
to valuation dates occurring during 2018.

----------------------------------------------------------------------------------------------------------------
                                                    Males                                 Females
----------------------------------------------------------------------------------------------------------------
                                                                Optional                               Optional
                                        Non-                    combined       Non-                    combined
                Age                  annuitant    Annuitant    table for    annuitant    Annuitant    table for
                                                              small plans                            small plans
----------------------------------------------------------------------------------------------------------------
0.................................     0.002420     0.002420     0.002420     0.002234     0.002234     0.002234
1.................................     0.000142     0.000142     0.000142     0.000140     0.000140     0.000140
2.................................     0.000097     0.000097     0.000097     0.000092     0.000092     0.000092
3.................................     0.000081     0.000081     0.000081     0.000070     0.000070     0.000070
4.................................     0.000064     0.000064     0.000064     0.000053     0.000053     0.000053
5.................................     0.000056     0.000056     0.000056     0.000048     0.000048     0.000048
6.................................     0.000051     0.000051     0.000051     0.000045     0.000045     0.000045
7.................................     0.000046     0.000046     0.000046     0.000042     0.000042     0.000042
8.................................     0.000039     0.000039     0.000039     0.000039     0.000039     0.000039
9.................................     0.000032     0.000032     0.000032     0.000037     0.000037     0.000037
10................................     0.000027     0.000027     0.000027     0.000035     0.000035     0.000035
11................................     0.000029     0.000029     0.000029     0.000036     0.000036     0.000036
12................................     0.000044     0.000044     0.000044     0.000042     0.000042     0.000042
13................................     0.000058     0.000058     0.000058     0.000048     0.000048     0.000048
14................................     0.000072     0.000072     0.000072     0.000053     0.000053     0.000053
15................................     0.000087     0.000087     0.000087     0.000059     0.000059     0.000059
16................................     0.000102     0.000102     0.000102     0.000064     0.000064     0.000064
17................................     0.000118     0.000118     0.000118     0.000068     0.000068     0.000068
18................................     0.000135     0.000135     0.000135     0.000072     0.000072     0.000072
19................................     0.000153     0.000153     0.000153     0.000075     0.000075     0.000075
20................................     0.000170     0.000170     0.000170     0.000076     0.000076     0.000076
21................................     0.000192     0.000192     0.000192     0.000078     0.000078     0.000078
22................................     0.000214     0.000214     0.000214     0.000080     0.000080     0.000080
23................................     0.000229     0.000229     0.000229     0.000084     0.000084     0.000084
24................................     0.000238     0.000238     0.000238     0.000087     0.000087     0.000087
25................................     0.000230     0.000230     0.000230     0.000090     0.000090     0.000090
26................................     0.000226     0.000226     0.000226     0.000094     0.000094     0.000094
27................................     0.000226     0.000226     0.000226     0.000099     0.000099     0.000099
28................................     0.000230     0.000230     0.000230     0.000105     0.000105     0.000105
29................................     0.000238     0.000238     0.000238     0.000111     0.000111     0.000111
30................................     0.000249     0.000249     0.000249     0.000120     0.000120     0.000120
31................................     0.000263     0.000263     0.000263     0.000130     0.000130     0.000130
32................................     0.000278     0.000278     0.000278     0.000142     0.000142     0.000142
33................................     0.000294     0.000294     0.000294     0.000155     0.000155     0.000155
34................................     0.000309     0.000309     0.000309     0.000168     0.000168     0.000168
35................................     0.000323     0.000323     0.000323     0.000182     0.000182     0.000182
36................................     0.000336     0.000336     0.000336     0.000196     0.000196     0.000196
37................................     0.000350     0.000350     0.000350     0.000213     0.000213     0.000213
38................................     0.000366     0.000366     0.000366     0.000231     0.000231     0.000231
39................................     0.000385     0.000385     0.000385     0.000251     0.000251     0.000251
40................................     0.000410     0.000410     0.000410     0.000273     0.000273     0.000273
41................................     0.000438     0.000443     0.000438     0.000298     0.000296     0.000298
42................................     0.000474     0.000516     0.000474     0.000326     0.000344     0.000326
43................................     0.000518     0.000627     0.000519     0.000358     0.000419     0.000358
44................................     0.000573     0.000779     0.000577     0.000395     0.000520     0.000395
45................................     0.000636     0.000973     0.000644     0.000436     0.000651     0.000438
46................................     0.000712     0.001213     0.000726     0.000484     0.000813     0.000489
47................................     0.000798     0.001502     0.000820     0.000538     0.001010     0.000550
48................................     0.000896     0.001844     0.000930     0.000597     0.001245     0.000619
49................................     0.001005     0.002248     0.001056     0.000661     0.001522     0.000697
50................................     0.001128     0.002719     0.001200     0.000734     0.001844     0.000790

[[Page 95924]]

 
51................................     0.001265     0.002963     0.001350     0.000814     0.001961     0.000881
52................................     0.001418     0.003224     0.001542     0.000903     0.002099     0.000992
53................................     0.001580     0.003481     0.001761     0.001003     0.002263     0.001122
54................................     0.001761     0.003751     0.002017     0.001114     0.002454     0.001273
55................................     0.001964     0.004040     0.002393     0.001235     0.002673     0.001508
56................................     0.002200     0.004357     0.002884     0.001367     0.002921     0.001811
57................................     0.002474     0.004704     0.003317     0.001509     0.003200     0.002084
58................................     0.002796     0.005088     0.003805     0.001661     0.003512     0.002379
59................................     0.003174     0.005515     0.004341     0.001823     0.003860     0.002711
60................................     0.003613     0.005989     0.004951     0.001994     0.004238     0.003106
61................................     0.004122     0.006516     0.005639     0.002181     0.004659     0.003619
62................................     0.004705     0.007100     0.006406     0.002381     0.005119     0.004188
63................................     0.005364     0.007742     0.007243     0.002600     0.005616     0.004868
64................................     0.006111     0.008457     0.008071     0.002842     0.006165     0.005515
65................................     0.006940     0.009234     0.008966     0.003107     0.006766     0.006236
66................................     0.007779     0.010103     0.009945     0.003465     0.007430     0.007080
67................................     0.008697     0.011056     0.010940     0.003863     0.008170     0.007897
68................................     0.009709     0.012114     0.012027     0.004308     0.008993     0.008770
69................................     0.010836     0.013302     0.013231     0.004806     0.009912     0.009722
70................................     0.012093     0.014637     0.014571     0.005366     0.010945     0.010756
71................................     0.013486     0.016126     0.016064     0.006001     0.012111     0.011925
72................................     0.015044     0.017799     0.017742     0.006711     0.013412     0.013230
73................................     0.016794     0.019693     0.019640     0.007521     0.014886     0.014711
74................................     0.018751     0.021823     0.021775     0.008439     0.016552     0.016387
75................................     0.020950     0.024237     0.024194     0.009485     0.018443     0.018291
76................................     0.023428     0.026986     0.026949     0.010678     0.020600     0.020465
77................................     0.026183     0.030081     0.030051     0.012035     0.023061     0.022949
78................................     0.029308     0.033645     0.033622     0.013582     0.025888     0.025804
79................................     0.032774     0.037661     0.037648     0.015347     0.029144     0.029097
80................................     0.036705     0.042295     0.042295     0.017347     0.032886     0.032886
81................................     0.038556     0.047291     0.047291     0.019058     0.036992     0.036992
82................................     0.042087     0.053009     0.053009     0.022345     0.041662     0.041662
83................................     0.047283     0.059466     0.059466     0.027251     0.047017     0.047017
84................................     0.054248     0.066860     0.066860     0.033811     0.053130     0.053130
85................................     0.062990     0.075196     0.075196     0.042053     0.060056     0.060056
86................................     0.073605     0.084646     0.084646     0.052009     0.067888     0.067888
87................................     0.086115     0.095308     0.095308     0.063725     0.076724     0.076724
88................................     0.100513     0.107196     0.107196     0.077205     0.086549     0.086549
89................................     0.116840     0.120431     0.120431     0.092462     0.097426     0.097426
90................................     0.135087     0.135087     0.135087     0.109484     0.109484     0.109484
91................................     0.150610     0.150610     0.150610     0.122541     0.122541     0.122541
92................................     0.166534     0.166534     0.166534     0.136397     0.136397     0.136397
93................................     0.182546     0.182546     0.182546     0.150811     0.150811     0.150811
94................................     0.198598     0.198598     0.198598     0.165818     0.165818     0.165818
95................................     0.214442     0.214442     0.214442     0.181360     0.181360     0.181360
96................................     0.232944     0.232944     0.232944     0.198746     0.198746     0.198746
97................................     0.251903     0.251903     0.251903     0.216930     0.216930     0.216930
98................................     0.271612     0.271612     0.271612     0.235921     0.235921     0.235921
99................................     0.291889     0.291889     0.291889     0.255617     0.255617     0.255617
100...............................     0.312680     0.312680     0.312680     0.275938     0.275938     0.275938
101...............................     0.333720     0.333720     0.333720     0.296628     0.296628     0.296628
102...............................     0.354570     0.354570     0.354570     0.317471     0.317471     0.317471
103...............................     0.375136     0.375136     0.375136     0.338385     0.338385     0.338385
104...............................     0.395172     0.395172     0.395172     0.358868     0.358868     0.358868
105...............................     0.413945     0.413945     0.413945     0.379183     0.379183     0.379183
106...............................     0.432145     0.432145     0.432145     0.398878     0.398878     0.398878
107...............................     0.449197     0.449197     0.449197     0.417703     0.417703     0.417703
108...............................     0.465497     0.465497     0.465497     0.435384     0.435384     0.435384
109...............................     0.480869     0.480869     0.480869     0.452108     0.452108     0.452108
110...............................     0.495080     0.495080     0.495080     0.467928     0.467928     0.467928
111...............................     0.500557     0.500557     0.500557     0.482562     0.482562     0.482562
112...............................     0.500454     0.500454     0.500454     0.496164     0.496164     0.496164
113...............................     0.500352     0.500352     0.500352     0.502110     0.502110     0.502110
114...............................     0.500201     0.500201     0.500201     0.500952     0.500952     0.500952
115...............................     0.500000     0.500000     0.500000     0.500000     0.500000     0.500000
116...............................     0.500000     0.500000     0.500000     0.500000     0.500000     0.500000
117...............................     0.500000     0.500000     0.500000     0.500000     0.500000     0.500000
118...............................     0.500000     0.500000     0.500000     0.500000     0.500000     0.500000
119...............................     0.500000     0.500000     0.500000     0.500000     0.500000     0.500000
120...............................     1.000000     1.000000     1.000000     1.000000     1.000000     1.000000
----------------------------------------------------------------------------------------------------------------


[[Page 95925]]

0
Par. 3. Section 1.430(h)(3)-2 is revised to read as follows:


Sec.  1.430(h)(3)-2  Plan-specific substitute mortality tables used to 
determine present value.

    (a) In general. This section sets forth rules for the use of 
substitute mortality tables under section 430(h)(3)(C) in determining 
any present value or making any computation under section 430 in 
accordance with Sec.  1.430(h)(3)-1(a)(1). In order to use substitute 
mortality tables, a plan sponsor must obtain approval to use substitute 
mortality tables for the plan in accordance with the procedures set 
forth in paragraph (b) of this section. Paragraph (c) of this section 
sets forth rules for the development of substitute mortality tables, 
including guidelines providing that a plan must have either full or 
partial credibility in order to have sufficient credible mortality 
information to use substitute mortality tables. Paragraph (d) of this 
section describes the requirements for full credibility. Paragraph (e) 
of this section describes the requirements for partial credibility. 
Paragraph (f) of this section provides special rules for newly 
affiliated plans not using substitute mortality tables. Paragraph (g) 
of this section specifies the effective date and applicability date of 
this section. The Commissioner may, in revenue rulings and procedures, 
notices or other guidance published in the Internal Revenue Bulletin 
(see Sec.  601.601(d)(2)(ii)(b) of this chapter), provide additional 
guidance regarding approval and use of substitute mortality tables 
under section 430(h)(3)(C) and related matters.
    (b) Procedures for obtaining approval to use substitute mortality 
tables--(1) Written request to use substitute mortality tables--(i) 
General requirements. In order to use substitute mortality tables, a 
plan sponsor must submit a written request to the Commissioner that 
demonstrates that those substitute mortality tables meet the 
requirements of section 430(h)(3)(C) and this section. This request 
must specify the first plan year and the term of years (not more than 
10) for which the tables are to apply.
    (ii) Time for written request. Substitute mortality tables may not 
be used for a plan year unless the plan sponsor submits the written 
request described in paragraph (b)(1)(i) of this section at least 7 
months prior to the first day of the first plan year for which the 
substitute mortality tables are to apply.
    (2) Commissioner's review of request--(i) In general. During the 
180-day period that begins on the date the plan sponsor submits a 
request to use substitute mortality tables for a plan pursuant to this 
section, the Commissioner will determine whether the request to use 
substitute mortality tables satisfies the requirements of this section 
(including any published guidance issued pursuant to paragraph (a) of 
this section), and will either approve or deny the request. The 
Commissioner will deny a request if the request fails to meet the 
requirements of this section or if the Commissioner determines that a 
substitute mortality table does not sufficiently reflect the mortality 
experience of the applicable plan population.
    (ii) Request for additional information. The Commissioner may 
request additional information with respect to the submission. Failure 
to provide that information on a timely basis constitutes grounds for 
denial of the request.
    (iii) Deemed approval. Except as provided in paragraph (b)(2)(iv) 
of this section, if the Commissioner does not issue a denial within the 
180-day review period, the request is deemed to have been approved.
    (iv) Extension of time permitted. The Commissioner and a plan 
sponsor may, before the expiration of the 180-day review period, agree 
in writing to extend that period, provided that any such agreement also 
specifies any revisions in the plan sponsor's request, including any 
change in the requested term of use of the substitute mortality tables.
    (c) Development of substitute mortality tables--(1) Substitute 
mortality tables must be used for all plans in controlled group--(i) 
General rule. Except as otherwise provided in this paragraph (c), 
substitute mortality tables are permitted to be used for a plan for a 
plan year only if, for that plan year (or any portion of that plan 
year), substitute mortality tables are also approved and used for each 
other pension plan subject to the requirements of section 430 that is 
maintained by the plan sponsor and by each member of the plan sponsor's 
controlled group. For purposes of this section, the term controlled 
group means any group treated as a single employer under paragraph (b), 
(c), (m), or (o) of section 414.
    (ii) Treatment of plans without credible mortality information. The 
rule of paragraph (c)(1)(i) of this section does not prohibit use of 
substitute mortality tables for one plan for a plan year if the only 
other plan or plans maintained by the plan sponsor (or by a member of 
the plan sponsor's controlled group) for which substitute mortality 
tables are not used are too small to have fully or partially credible 
mortality information for the plan year. For purposes of demonstrating 
that neither males nor females under a plan have credible mortality 
information for a plan year, the length of the experience study period 
must be the same length as the longest experience study period used for 
any plan within the controlled group.
    (2) Mortality experience requirements--(i) In general. Substitute 
mortality tables must reflect the actual mortality experience of the 
pension plan for which the tables are to be used and that mortality 
experience must be credible mortality information as described in 
paragraph (c)(2)(ii) of this section. Separate mortality tables must be 
established for each gender under the plan, and a substitute mortality 
table is permitted to be established for a gender only if the plan has 
credible mortality information with respect to that gender. See 
paragraph (d)(5) of this section for rules permitting the use of 
substitute mortality tables for populations within a gender that have 
full credibility.
    (ii) Credible mortality information--(A) In general. There is 
credible mortality information for a gender within a plan if and only 
if the mortality experience with respect to that gender satisfies the 
requirement for either--
    (1) Full credibility (as described in paragraph (d) of this 
section); or
    (2) Partial credibility (as described in paragraph (e) of this 
section).
    (B) Simplified rule. Whether there is credible mortality 
information for a gender may be determined by only taking into account 
people who are at least age 50 and less than age 100. If there is 
credible mortality information for a gender when applying this 
simplified rule, the entire gender (not just those who are at least age 
50 and less than age 100) has credible mortality information.
    (iii) Gender without credible mortality information--(A) In 
general. If for a plan, one gender has credible mortality information 
but for a plan year the other gender does not have credible mortality 
information, then the substitute mortality tables are established for 
the gender that does have credible mortality information and the 
mortality tables under Sec.  1.430(h)(3)-1 are used for the gender that 
does not have credible mortality information.
    (B) Demonstration of lack of credible mortality information for a 
gender. In general, in order to demonstrate that a gender within a plan 
does not have credible mortality information for a plan year, the 
demonstration that the gender within the plan has fewer than the

[[Page 95926]]

minimum number of actual deaths to have partial credibility, as 
described in paragraph (e)(1) of this section, must be made by 
analyzing the actual number of deaths over a period that is the same 
length as the period for the experience study on which the substitute 
mortality tables are based and that ends less than three years before 
the first day of the plan year.
    (3) Determination of substitute mortality tables--(i) Requirement 
to use generational mortality table. A plan's substitute mortality 
tables must be generational mortality tables. A plan's substitute 
mortality tables are determined using the plan's base substitute 
mortality tables developed pursuant to paragraph (d) or (e) of this 
section and the mortality improvement factors described in paragraph 
(c)(3)(ii) of this section.
    (ii) Determination of mortality improvement factors. The mortality 
improvement factor for an age and a gender is the cumulative mortality 
improvement factor determined under Sec.  1.430(h)(3)-1(a)(2)(i)(E) for 
the applicable period. The applicable period is the period beginning 
with the base year of the base substitute mortality table determined 
under paragraph (d) or (e) of this section and ending in the calendar 
year in which the individual attains the age for which the probability 
of death is being determined. The base year for the base substitute 
mortality table is the calendar year that contains the day before the 
midpoint of the experience study period.
    (4) Disabled individuals. Under section 430(h)(3)(D), separate 
mortality tables are permitted to be used for certain disabled 
individuals. If such separate mortality tables are used for those 
disabled individuals, then those individuals are disregarded for all 
purposes under this section. Thus, if the mortality tables under 
section 430(h)(3)(D) are used for disabled individuals under a plan, 
mortality experience with respect to those individuals must be excluded 
in developing mortality rates for substitute mortality tables under 
this section.
    (5) Aggregation--(i) Permissive aggregation of plans. A plan 
sponsor may use a set of substitute mortality tables for two or more 
its plans provided that the rules of this section are applied by 
treating those plans as a single plan. In such a case, the substitute 
mortality tables must be used for the aggregated plans and must be 
based on data collected with respect to those aggregated plans.
    (ii) Required aggregation of plans. In general, plans are not 
required to be aggregated for purposes of applying the rules of this 
section. However, for purposes of this section, a plan is required to 
be aggregated with any plan that was previously spun off from that plan 
if a purpose of the spinoff is to avoid the use of substitute mortality 
tables for any of the plans that were involved in the spinoff.
    (6) Duration of use of tables--(i) General rule. Except as provided 
in this paragraph (c)(6), substitute mortality tables are used for a 
plan for the term of consecutive plan years specified in the plan 
sponsor's written request to use such tables under paragraph (b)(1) of 
this section and approved by the Commissioner, or a shorter period 
prescribed by the Commissioner in the approval to use substitute 
mortality tables. Following the end of the approved term of use, or 
following any early termination of use described in this paragraph 
(c)(6), the mortality tables specified in Sec.  1.430(h)(3)-1 are used 
for the plan unless approval under paragraph (b)(1) of this section has 
been received by the plan sponsor to use substitute mortality tables 
for a further term.
    (ii) Early termination of use of tables. A plan's substitute 
mortality tables must not be used beginning with the earliest of--
    (A) For a plan using a substitute mortality table for only one 
gender because of a lack of credible mortality information with respect 
to the other gender, the first plan year for which there is credible 
mortality information with respect to the gender that had lacked 
credible mortality information (unless an approved substitute mortality 
table is used for that gender);
    (B) The first plan year for which the plan fails to satisfy the 
requirements of paragraph (c)(1) of this section (regarding use of 
substitute mortality tables by controlled group members);
    (C) The second plan year following the plan year for which there is 
a significant change in individuals covered by the plan as described in 
paragraph (c)(6)(iii) of this section;
    (D) The first plan year following the plan year for which a 
substitute mortality table used for a plan population is no longer 
accurately predictive of future mortality of that population, as 
determined by the Commissioner or as certified by the plan's actuary to 
the satisfaction of the Commissioner; or
    (E) The date specified in guidance published in the Internal 
Revenue Bulletin (see Sec.  601.601(d)(2)(ii)(b) of this chapter) in 
conjunction with a replacement of mortality tables specified under 
section 430(h)(3)(A) and Sec.  1.430(h)(3)-1 (other than annual updates 
to the static mortality tables issued pursuant to Sec.  1.430(h)(3)-
1(a)(3) or changes to the mortality improvement rates pursuant to Sec.  
1.430(h)(3)-1(a)(2)(i)(C)).
    (iii) Significant change in coverage--(A) Change in coverage from 
time of experience study. For purposes of applying the rules of 
paragraph (c)(6)(ii)(C) of this section, a significant change in the 
individuals covered by a substitute mortality table occurs if there is 
an increase or decrease in the number of individuals of at least 20 
percent compared to the average number of individuals in that 
population over the years covered by the experience study on which the 
substitute mortality tables are based. However, a change in coverage is 
not treated as significant if the plan's actuary certifies in writing 
to the satisfaction of the Commissioner that the substitute mortality 
tables used for the plan population continue to be accurately 
predictive of future mortality of that population (taking into account 
the effect of the change in the population).
    (B) Change in coverage from time of certification. For purposes of 
applying the rules of paragraph (c)(6)(ii)(C) of this section, a 
significant change in the individuals covered by a substitute mortality 
table occurs if there is an increase or decrease in the number of 
individuals covered by a substitute mortality table of at least 20 
percent compared to the number of individuals in a plan year for which 
a certification described in paragraph (c)(6)(iii)(A) of this section 
was made on account of a prior change in coverage. However, a change in 
coverage is not treated as significant if the plan's actuary certifies 
in writing to the satisfaction of the Commissioner that the substitute 
mortality tables used by the plan with respect to the covered 
population continue to be accurately predictive of future mortality of 
that population (taking into account the effect of the change in the 
plan population).
    (d) Full credibility--(1) In general. The mortality experience with 
respect to a gender or other population within a plan has full 
credibility if the actual number of deaths for that population during 
the experience study period described in paragraph (d)(2) of this 
section is at least the full credibility threshold described in 
paragraph (d)(3) of this section. Paragraph (d)(4) of this section 
provides rules for the creation of a base substitute mortality table 
from the experience study, which apply if the mortality experience for 
the population has full credibility.
    (2) Experience study period requirements. The base substitute

[[Page 95927]]

mortality table for a gender or other population within a plan must be 
developed from an experience study of the mortality experience of that 
population that is collected over an experience study period. The 
length of the experience study period must be at least 2 years and no 
more than 5 years. The last day of the final year reflected in the 
experience data must be less than 3 years before the first day of the 
first plan year for which the substitute mortality tables are to apply. 
For example, if July 1, 2019, is the first day of the first plan year 
for which the substitute mortality tables will be used, then an 
experience study using calendar year data must include data collected 
for a period that ends no earlier than December 31, 2016.
    (3) Full credibility threshold--(i) Threshold number of deaths. The 
full credibility threshold for a gender or other population within a 
plan is the product of 1,082 and the population's benefit dispersion 
factor. In calculating the population's benefit dispersion factor, for 
purposes of paragraphs (d)(3)(iii), (iv), and (v) of this section, the 
population is adjusted, as appropriate, for people who leave on account 
of reason other than death.
    (ii) Population's benefit dispersion factor. The population's 
benefit dispersion factor is equal to--
    (A) The number of expected deaths for the population during the 
experience study period (as defined in paragraph (d)(3)(iii) of this 
section); multiplied by
    (B) The mortality-weighted square of the benefits (as defined in 
paragraph (d)(3)(iv) of this section); divided by
    (C) The square of the mortality-weighted benefits (as defined in 
paragraph (d)(3)(v) of this section).
    (iii) Number of expected deaths. The number of expected deaths for 
a population during the experience study period is equal to the sum, 
for each year in the experience study period, of the expected number of 
deaths in the population during the year using the mortality rates from 
the standard mortality tables set forth in paragraph (d)(4)(iii) of 
this section.
    (iv) Mortality-weighted square of the benefits. The mortality-
weighted square of the benefits for a population is the sum, for each 
year in the experience study period, for all individuals for each age 
in the population at the beginning of the year, of the product of--
    (A) The probability of death of those individuals using the 
mortality rate for that age from the standard mortality table set forth 
in paragraph (d)(4)(iii) of this section; and
    (B) The sum of the square of the accrued benefits (substituting the 
current periodic payment in the case of individuals in pay status) for 
those individuals.
    (v) Square of the mortality-weighted benefits. The square of the 
mortality-weighted benefits is equal to the square of the sum, for each 
year in the experience study period, for all individuals for each age 
in the population at the beginning of the year, of the product of--
    (A) The probability of death of those individuals using the 
mortality rate for that age from the standard mortality table set forth 
in paragraph (d)(4)(iii) of this section; and
    (B) The sum of the accrued benefits (substituting the current 
periodic payment in the case of individuals in pay status) for those 
individuals.
    (4) Development of mortality rates--(i) In general. The mortality 
rates derived from the experience study must be amounts-weighted 
mortality rates that are derived by multiplying the mortality rate from 
the standard mortality table described in paragraph (d)(4)(iii) of this 
section by the mortality ratio determined under paragraph (d)(4)(ii) of 
this section. If the simplified rule of paragraph (c)(2)(ii)(B) of this 
section is used for the population, then the mortality ratio is 
determined only taking into account people who are at least 50 years 
old and less than 100 years old, but the mortality ratio is applied to 
all ages. Because amounts-weighted mortality rates for a plan cannot be 
determined without benefit amounts, the mortality experience study used 
to develop a base table must not include periods before the plan was 
established.
    (ii) Mortality ratio. The mortality ratio for a gender or other 
population within a plan is equal to the quotient determined by 
dividing--
    (A) The sum, for each year in the experience study period, of the 
accrued benefits (substituting the current periodic payment in the case 
of individuals in pay status) for all individuals in the population at 
the beginning of the year who died during the year, by
    (B) The sum, for each year in the experience study period, for all 
individuals for each age in the population at the beginning of the year 
(adjusted, as appropriate, for people who leave on account of reason 
other than death), of the product of--
    (1) The probability of death of those individuals using the 
mortality rate for that age from the standard mortality table set forth 
in paragraph (d)(4)(iii) of this section; and
    (2) The sum of the accrued benefits (substituting the current 
periodic payment in the case of individuals in pay status) for those 
individuals.
    (iii) Standard mortality table--(A) Projection of base table. The 
standard mortality table for a year is the mortality table determined 
by applying cumulative mortality improvement factors determined under 
Sec.  1.430(h)(3)-1(a)(2)(i)(E) to the base mortality table under Sec.  
1.430(h)(3)-1(d) for the period beginning with 2006 and ending in the 
base year for the base substitute mortality table determined under 
paragraph (d) or (e) of this section. For purposes of the previous 
sentence, the cumulative mortality improvement factors are determined 
using the mortality improvement rates described in Sec.  1.430(h)(3)-
1(a)(2)(i)(C) that apply for the calendar year during which the plan 
sponsor submits the request to use substitute mortality tables. If the 
plan sponsor submits such a request during 2017, then the cumulative 
mortality improvement factors are determined using the mortality 
improvement rates contained in the Mortality Improvement Scale MP-2016 
Report (issued by the Retirement Plans Experience Committee (RPEC) of 
the Society of Actuaries and available at www.soa.org/Research/Experience-Study/Pension/research-2016-mp.aspx).
    (B) Selection of base table. If the population consists solely of 
annuitants, the annuitant base mortality table under Sec.  1.430(h)(3)-
1(d) must be used for purposes of paragraph (d)(4)(iii)(A) of this 
section. If the population consists solely of nonannuitants, the 
nonannuitant base mortality table under Sec.  1.430(h)(3)-1(d) must be 
used for that purpose. If the population includes both annuitants and 
non-annuitants, a combination of the annuitant and nonannuitant base 
tables under Sec.  1.430(h)(3)-1(d) must be used for that purpose. The 
combined table is constructed using the weighting factors for small 
plans that are set forth in Sec.  1.430(h)(3)-1(d). The weighting 
factors are applied to develop the combined table using the following 
equation: Combined mortality rate = [nonannuitant rate * (1 - weighting 
factor)] + [annuitant rate * weighting factor].
    (iv) Change in number of individuals covered by table. Experience 
data may not be used to develop a base table if the number of 
individuals in the population covered by the table (for example, the 
male annuitant population) as of the last day of the plan year before 
the year the request to use substitute mortality tables is made, 
compared to the average number of individuals in that population over 
the years covered by the experience study on which the

[[Page 95928]]

substitute mortality tables are based, reflects a difference of 20 
percent or more, unless it is demonstrated to the satisfaction of the 
Commissioner that the experience data is accurately predictive of 
future mortality of that plan population (taking into account the 
effect of the change in individuals) after appropriate adjustments to 
the data are made (for example, excluding data from individuals with 
respect to a spun-off portion of the plan). For this purpose, a 
reasonable estimate of the number of individuals in the population 
covered by the table may be used.
    (5) Separate tables for specified populations--(i) In general. 
Except as provided in this paragraph (d)(5), separate substitute 
mortality tables are permitted to be used for separate populations 
within a gender under a plan only if--
    (A) All individuals of that gender in the plan are divided into 
separate populations;
    (B) Each separate population has mortality experience that has full 
credibility as determined under the rules of paragraph (d)(5)(iii) of 
this section; and
    (C) The separate base substitute mortality table for each separate 
population is developed applying the rules of paragraphs (d)(1) through 
(4) of this section using an experience study that takes into account 
solely members of that population.
    (ii) Annuitant and nonannuitant separate populations. 
Notwithstanding paragraph (d)(5)(i)(B) of this section, substitute 
mortality tables for separate populations of annuitants and 
nonannuitants within a gender may be used even if only one of those 
separate populations has credible mortality information. Similarly, if 
separate populations that satisfy paragraph (d)(5)(i)(B) of this 
section are established, then any of those populations may be further 
subdivided into separate annuitant and nonannuitant subpopulations, 
provided that at least one of the two resulting subpopulations has 
credible mortality experience. The standard mortality tables under 
Sec.  1.430(h)(3)-1 are used for a resulting subpopulation that does 
not have credible mortality information. For example, in the case of a 
plan with mortality experience for both its male hourly and salaried 
individuals that has full credibility, if the male salaried annuitant 
population has credible mortality information, substitute mortality 
tables may be used for the plan with respect to that population even if 
the standard mortality tables under Sec.  1.430(h)(3)-1 are used with 
respect to the male salaried nonannuitant population (because that 
nonannuitant population does not have credible mortality information).
    (iii) Credible mortality experience for separate populations. In 
determining whether the mortality experience for a separate population 
within a gender has full credibility, the requirements of paragraph 
(d)(1) of this section must be satisfied but, in applying that 
paragraph (d)(1), the separate population should be substituted for the 
particular gender. In demonstrating that an annuitant or nonannuitant 
population within a gender or within a separate population does not 
have credible mortality information, the requirements of paragraph 
(c)(2)(iii)(B) of this section must be satisfied but, in applying that 
paragraph, the annuitant (or nonannuitant) population should be 
substituted for the particular gender.
    (e) Partial credibility--(1) In general. The mortality experience 
with respect to a population has partial credibility if the actual 
number of deaths for that population during the experience study period 
described in paragraph (d)(2) of this section is at least equal to the 
partial credibility threshold of 100 and is less than the full 
credibility threshold described for the population in paragraph (d)(3) 
of this section. If the mortality experience for the population has 
partial credibility, then in lieu of creating a base substitute 
mortality table as described in paragraph (d) of this section, the base 
substitute mortality table is created as the sum of--
    (i) The product of--
    (A) The partial credibility weighting factor determined under 
paragraph (e)(2) of this section; and
    (B) The mortality rates that are derived from the experience study 
determined under paragraph (d)(4)(i) of this section, and
    (ii) The product of--
    (A) One minus the partial credibility weighting factor described in 
paragraph (e)(2) of this section; and
    (B) The mortality rate from the standard mortality tables described 
in paragraph (d)(4)(iii) of this section.
    (2) Partial credibility weighting factor. The partial credibility 
weighting factor is equal to the square root of the fraction--
    (i) The numerator of which is the actual number of deaths for the 
population during the experience study period, and
    (ii) The denominator of which is the full credibility threshold for 
the population described in paragraph (d)(3) of this section.
    (f) Special rules for newly affiliated plans--(1) In general. This 
paragraph (f) provides special rules that provide temporary relief from 
certain rules in this section in the case of a controlled group that 
includes a newly affiliated plan. Paragraph (f)(2) of this section 
provides a transition period during which the requirement in paragraph 
(c)(1) of this section (that is, the requirement that all plans within 
the controlled group that have credible mortality information must use 
substitute mortality tables) is not applicable. Paragraph (f)(3) of 
this section provides special rules that permit the use of a shorter 
experience study period in the case of a newly affiliated plan that 
excludes the mortality experience data for the period prior to the date 
the plan sponsor becomes maintained by a member of the new plan 
sponsor's controlled group. Paragraph (f)(4) of this section defines 
newly affiliated plan.
    (2) Transition period for newly affiliated plans. The use of 
substitute mortality tables for a plan within a controlled group is not 
prohibited merely because substitute mortality tables are not used 
during the transition period for a newly affiliated plan that fails to 
demonstrate a lack of credible mortality information during the that 
period. Similarly, during the transition period, the use of substitute 
mortality tables for a newly affiliated plan is not prohibited merely 
because substitute mortality tables are not used for another plan 
within the controlled group that fails to demonstrate a lack of 
credible mortality information during that period. The transition 
period runs through the last day of the plan year that contains the 
last day of the period described in section 410(b)(6)(C)(ii) for either 
of the plans, whichever is later.
    (3) Experience study period for newly affiliated plan--(i) In 
general. The mortality experience data for a newly affiliated plan may 
either include or exclude mortality experience data for the period 
prior to the date the plan becomes maintained by a member of the new 
plan sponsor's controlled group. If a plan sponsor excludes mortality 
experience data for the period prior to the date the plan becomes 
maintained within the new plan sponsor's controlled group, the 
exclusion must apply for all populations within the plan.
    (ii) Demonstration relating to lack of credible mortality 
experience. If the experience study for a newly affiliated plan 
excludes mortality experience data for the period prior to the date the 
plan becomes maintained by a member of the new plan sponsor's 
controlled group, then the demonstration that the plan does not have 
credible mortality information for a plan year that begins

[[Page 95929]]

after the transition period can be made using a shorter experience 
study period than would otherwise be permitted under paragraph 
(c)(2)(iii)(B) of this section, provided that the experience study 
period begins with the date the plan becomes maintained within the 
sponsor's controlled group and ends not more than one year and one day 
before the first day of the plan year.
    (iii) Demonstration relating to credible mortality experience. If 
the experience study for a newly affiliated plan excludes mortality 
experience data for the period prior to the date the plan becomes 
maintained by a member of the new plan sponsor's controlled group and 
the plan fails to demonstrate that it does not have credible mortality 
information for the plan year under the rules of paragraph (f)(3)(ii) 
of this section, then other plans within the controlled group can 
continue to use substitute mortality tables only if substitute 
mortality tables are used for the newly affiliated plan the plan year. 
In such a case, the experience study period can be a shorter period 
than the period in paragraph (d)(2) of this section, provided that the 
period is at least one year.
    (4) Definition of newly affiliated plan. For purposes of this 
paragraph (f), a plan is treated as a newly affiliated plan if it 
becomes maintained by the plan sponsor (or by a member of the plan 
sponsor's controlled group) in connection with a merger, acquisition, 
or similar transaction described in Sec.  1.410(b)-2(f). A plan also is 
treated as a newly affiliated plan for purposes of this section if the 
plan is established in connection with a transfer of assets and 
liabilities from another employer's plan in connection with a merger, 
acquisition, or similar transaction described in Sec.  1.410(b)-2(f).
    (g) Effective/applicability date. This section applies for plan 
years beginning on or after January 1, 2018, and any substitute 
mortality table used for a plan for such a plan year must comply with 
the rules of this section.
0
Par. 4. Section 1.431(c)(6)-1 is revised to read as follows:


Sec.  1.431(c)(6)-1  Mortality tables used to determine current 
liability.

    (a) Mortality tables used to determine current liability. The 
mortality assumptions that apply to a defined benefit plan for the plan 
year pursuant to section 430(h)(3)(A) and Sec.  1.430(h)(3)-1(a) are 
used to determine a multiemployer plan's current liability for purposes 
of applying the rules of section 431(c)(6). Either the generational 
mortality tables used pursuant to Sec.  1.430(h)(3)-1(a)(2) or the 
static mortality tables used pursuant to Sec.  1.430(h)(3)-1(a)(3) are 
permitted to be used for a multiemployer plan for this purpose. 
However, for this purpose, substitute mortality tables under Sec.  
1.430(h)(3)-2 are not permitted to be used for a multiemployer plan.
    (b) Effective/applicability date. This section applies for plan 
years beginning on or after January 1, 2018. For rules that apply to 
plan years beginning before January 1, 2018 and on or after January 1, 
2008, see Sec.  1.431(c)(6)-1 (as contained in 26 CFR part 1 revised 
April 1, 2015).
0
Par. 5. Section 1.433(h)(3)-1 is added to read as follows:


Sec.  1.433(h)(3)-1   Mortality tables used to determine current 
liability.

    (a) Mortality tables used to determine current liability. In 
accordance with section 433(h)(3)(B), the mortality assumptions that 
apply to a defined benefit plan for the plan year pursuant to section 
430(h)(3)(A) and Sec.  1.430(h)(3)-1(a) are used to determine a CSEC 
plan's current liability for purposes of applying the rules of section 
433(c)(7)(C). Either the static mortality tables used pursuant to Sec.  
1.430(h)(3)-1(a)(3) or generational mortality tables used pursuant to 
Sec.  1.430(h)(3)-1(a)(2) are permitted to be used for a CSEC plan for 
this purpose, but substitute mortality tables under Sec.  1.430(h)(3)-2 
are not permitted to be used for this purpose.
    (b) Effective/applicability date. This section applies for plan 
years beginning on or after January 1, 2018.

John Dalrymple,
Deputy Commissioner for Services and Enforcement.
[FR Doc. 2016-30906 Filed 12-28-16; 8:45 am]
BILLING CODE 4830-01-P