Safety Standard for Recreational Off-Highway Vehicles (ROVs), 68963-69031 [2014-26500]

Agencies

• Consumer Product Safety Commission

[Federal Register Volume 79, Number 223 (Wednesday, November 19, 2014)]
[Proposed Rules]
[Pages 68963-69031]
From the Federal Register Online via the Government Printing Office [www.gpo.gov]
[FR Doc No: 2014-26500]



[[Page 68963]]

Vol. 79

Wednesday,

No. 223

November 19, 2014

Part II





 Consumer Product Safety Commission





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





16 CFR Part 1422





 Safety Standard for Recreational Off-Highway Vehicles (ROVs); Proposed 
Rule

Federal Register / Vol. 79 , No. 223 / Wednesday, November 19, 2014 / 
Proposed Rules

[[Page 68964]]


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

CONSUMER PRODUCT SAFETY COMMISSION

16 CFR Part 1422

RIN 3041-AC78
[Docket No. CPSC-2009-0087]


Safety Standard for Recreational Off-Highway Vehicles (ROVs)

AGENCY: Consumer Product Safety Commission.

ACTION: Notice of Proposed Rulemaking.

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

SUMMARY: The U.S. Consumer Product Safety Commission has determined 
preliminarily that there may be an unreasonable risk of injury and 
death associated with recreational off-highway vehicles (ROVs). To 
address these risks, the Commission proposes a rule that includes: 
lateral stability and vehicle handling requirements that specify a 
minimum level of rollover resistance for ROVs and require that ROVs 
exhibit sublimit understeer characteristics; occupant retention 
requirements that would limit the maximum speed of an ROV to no more 
than 15 miles per hour (mph), unless the seat belts of both the driver 
and front passengers, if any, are fastened, and would require ROVs to 
have a passive means, such as a barrier or structure, to limit further 
the ejection of a belted occupant in the event of a rollover; and 
information requirements.

DATES: Submit comments by February 2, 2015.

ADDRESSES: You may submit comments, identified by Docket No. CPSC-2009-
0087, by any of the following methods:
    Electronic Submissions: Submit electronic comments to the Federal 
eRulemaking Portal at: https://www.regulations.gov. Follow the 
instructions for submitting comments. The Commission does not accept 
comments submitted by electronic mail (email), except through 
www.regulations.gov. The Commission encourages you to submit electronic 
comments by using the Federal eRulemaking Portal, as described above.
    Written Submissions: Submit written submissions by mail/hand 
delivery/courier to: Office of the Secretary, Consumer Product Safety 
Commission, Room 820, 4330 East West Highway, Bethesda, MD 20814; 
telephone (301) 504-7923.
    Instructions: All submissions received must include the agency name 
and docket number for this notice. All comments received may be posted 
without change, including any personal identifiers, contact 
information, or other personal information provided, to: https://www.regulations.gov. Do not submit confidential business information, 
trade secret information, or other sensitive or protected information 
that you do not want to be available to the public. If furnished at 
all, such information should be submitted in writing.
    Docket: For access to the docket to read background documents or 
comments received, go to: https://www.regulations.gov, and insert the 
docket number CPSC-2009-0087, into the ``Search'' box, and follow the 
prompts.
    Submit comments related to the Paperwork Reduction Act (PRA) 
aspects of the proposed rule to the Office of Information and 
Regulatory Affairs, Attn: OMB Desk Officer for the CPSC or by email: 
OIRA_submission@omb.eop.gov or fax: 202-395-6881. In addition, comments 
that are sent to OMB also should be submitted electronically at https://www.regulations.gov, under Docket No. CPSC-2009-0087.

FOR FURTHER INFORMATION CONTACT: Caroleene Paul, Project Manager, 
Directorate for Engineering Sciences, Consumer Product Safety 
Commission, 5 Research Place, Rockville, MD 20850; telephone: 301-987-
2225; email: cpaul@cpsc.gov.

SUPPLEMENTARY INFORMATION:

I. Background

    The U.S. Consumer Product Safety Commission (Commission or CPSC) is 
proposing a standard for recreational off-highway vehicles (ROVs).\1\ 
ROVs are motorized vehicles that combine off-road capability with 
utility and recreational use. Reports of ROV-related fatalities and 
injuries prompted the Commission to publish an advance notice of 
proposed rulemaking (ANPR) in October 2009 to consider whether there 
may be unreasonable risks of injury and death associated with ROVs. (74 
FR 55495 (October 28, 2009)). The ANPR began a rulemaking proceeding 
under the Consumer Product Safety Act (CPSA). The Commission received 
116 comments in response to the ANPR. The Commission is now issuing a 
notice of proposed rulemaking (NPR) that would establish requirements 
for lateral stability, vehicle handling, and occupant protection 
performance, as well as information requirements. The information 
discussed in this preamble is derived from CPSC staff's briefing 
package for the NPR and from CPSC staff's supplemental memorandum to 
the Commission, which are available on CPSC's Web site at https://www.cpsc.gov//Global/Newsroom/FOIA/CommissionBriefingPackages/2014/SafetyStandardforRecreationalOff-HighwayVehicles-ProposedRule.pdf and 
https://www.cpsc.gov//Global/Newsroom/FOIA/CommissionBriefingPackages/2015/SupplementalInformation-ROVs.pdf.
---------------------------------------------------------------------------

    \1\ The Commission voted (3-2) to publish this notice in the 
Federal Register. Chairman Elliot F. Kaye and Commissioners Robert 
S. Adler and Marietta S. Robinson voted to approve publication of 
the proposed rule. Commissioners Ann Marie Buerkle and Joseph P. 
Mohorovic voted against publication of the proposed rule.
---------------------------------------------------------------------------

II. The Product

A. Products Covered

    ROVs are motorized vehicles designed for off-highway use with the 
following features: Four or more pneumatic tires designed for off-
highway use; bench or bucket seats for two or more occupants; 
automotive-type controls for steering, throttle, and braking; and a 
maximum vehicle speed greater than 30 miles per hour (mph). ROVs are 
also equipped with rollover protective structures (ROPS), seat belts, 
and other restraints (such as doors, nets, and shoulder barriers) for 
the protection of occupants.
    ROVs and All-Terrain Vehicles (ATVs) are similar in that both are 
motorized vehicles designed for off-highway use, and both are used for 
utility and recreational purposes. However, ROVs differ significantly 
from ATVs in vehicle design. ROVs have a steering wheel instead of a 
handle bar for steering; foot pedals instead of hand levers for 
throttle and brake control; and bench or bucket seats rather than 
straddle seating for the occupant(s). Most importantly, ROVs only 
require steering wheel input from the driver to steer the vehicle, and 
the motion of the occupants has little or no effect on vehicle control 
or stability. In contrast, ATVs require riders to steer with their 
hands and to maneuver their body front to back and side to side to 
augment the ATV's pitch and lateral stability.
    Early ROV models emphasized the utility aspects of the vehicles, 
but the recreational aspects of the vehicles have become very popular. 
Currently, there are two varieties of ROVs: Utility and recreational. 
Models emphasizing utility have larger cargo beds, higher cargo 
capacities, and lower top speeds. Models emphasizing recreation have 
smaller cargo beds, lower cargo capacities, and higher top speeds. Both 
utility and recreational ROVs with maximum speed greater than 30 mph 
are covered by the scope of this NPR.

B. Similar or Substitute Products

    There are several types of off-road vehicles that have some 
characteristics

[[Page 68965]]

that are similar to those of ROVs and may be considered substitutes for 
some purposes.
    Low-Speed Utility vehicles (UTVs)--Although ROVs can be considered 
to be a type of utility vehicle, their maximum speeds of greater than 
30 mph distinguish them from low-speed utility vehicles, which have 
maximum speeds of 25 mph or less. Like ROVs, low-speed utility vehicles 
have steering wheels and bucket or bench seating capable of carrying 
two or more riders. All utility vehicles have both work and 
recreational uses. However, low-speed utility vehicles might not be 
good substitutes for ROVs in recreational uses where speeds higher than 
30 mph are important.
    All-terrain vehicles (ATVs)--Unlike ROVs, ATVs make use of 
handlebars for steering and hand controls for operating the throttle 
and brakes. The seats on ATVs are intended to be straddled, unlike the 
bucket or bench seats on ROVs. Some ATVs are intended for work or 
utility applications, as well as for recreational uses; others are 
intended primarily for recreational purposes. ATVs are usually narrower 
than ROVs. This means that ATVs can navigate some trails or terrain 
that some ROVs might not be able to navigate.
    Unlike ROVs, ATVs are rider interactive. When riding an ATV, the 
driver must shift his or her weight from side to side while turning, or 
forward or backward when ascending or descending a hill or crossing an 
obstacle. Most ATVs are designed for one rider (the driver). On ATVs 
that are designed for more than one rider, the passenger sits behind 
the driver and not beside the driver as on ROVs.
    Go-Karts--Go-karts (sometimes called ``off-road buggies'') are 
another type of recreational vehicle that has some similarities to 
ROVs. Go-karts are usually intended solely for recreational purposes. 
Some go-karts with smaller engines are intended to be driven by 
children 12 and younger. Some go-karts are intended to be driven 
primarily on prepared surfaces. These go-karts would not be substitutes 
for ROVs. Other go-karts have larger engines, full suspensions, can 
reach maximum speeds in excess of 30 mph, and can be used on more 
surfaces. These go-karts could be close substitutes for ROVs in some 
recreational applications.

III. Risk of Injury

A. Incident Data

    As of April 5, 2013, CPSC staff is aware of 550 reported ROV-
related incidents that occurred between January 1, 2003 and April 5, 
2013; there were 335 reported fatalities and 506 reported injuries 
related to these incidents. To analyze hazard patterns related to ROVs, 
a multidisciplinary team of CPSC staff reviewed incident reports that 
CPSC received by December 31, 2011 concerning incidents that occurred 
between January 1, 2003 and December 31, 2011. CPSC received 428 
reports of ROV-related incidents that occurred between January 1, 2003 
and December 31, 2011, from the Injury and Potential Injury Incident 
(IPII) and In-Depth Investigation (INDP) databases.
    ROV-related incidents can involve more than one injury or fatality 
because the incidents often involve both a driver and passengers. There 
were a total of 826 victims involved in the 428 incidents. Of the 428 
ROV-related incidents, there were a total of 231 reported fatalities 
and 388 reported injuries. Seventy-five of the 388 injuries (19 
percent) could be classified as severe; that is, based on the 
information available, the victim has lasting repercussions from the 
injuries received in the incident. The remaining 207 victims were 
either not injured or their injury information was not known.
    Of the 428 ROV-related incidents, 76 incidents involved drivers 
under 16 years of age (18 percent); 227 involved drivers 16 years of 
age or older (53 percent); and 125 involved drivers of unknown age (29 
percent). Of the 227 incidents involving adult drivers, 86 (38 percent) 
are known to have involved the driver consuming at least one alcoholic 
beverage before the incident; 52 (23 percent) did not involve alcohol; 
and 89 (39 percent) have an unknown alcohol status of the driver.
    Of the 619 victims who were injured or killed, most (66 percent) 
were in a front seat of the ROV, either as a driver or passenger, when 
the incidents occurred. The remaining victims were in the rear of the 
ROV or in an unspecified location of the ROV.
    In many of the ROV-related incidents resulting in at least one 
death, the Commission was able to obtain more detailed information on 
the events surrounding the incident through an In-Depth Investigation 
(IDI). Of the 428 ROV-related incidents, 224 involved at least one 
death. This includes 218 incidents resulting in one fatality, five 
incidents resulting in two fatalities, and one incident resulting in 
three fatalities, for a total of 231 fatalities. Of the 224 fatal 
incidents, 145 (65 percent) occurred on an unpaved surface; 38 (17 
percent) occurred on a paved surface; and 41 (18 percent) occurred on 
unknown terrain.

B. Hazard Characteristics

    After CPSC staff determined that a reported incident resulting in 
at least one death or injury was ROV-related, a multidisciplinary team 
reviewed all the documents associated with the incident. The 
multidisciplinary team was made up of a human factors engineer, an 
economist, a health scientist, and a statistician. As part of the 
review process, each member of the review team considered every 
incident and coded victim characteristics, the characteristics of the 
vehicle involved, the environment, and the events of the incident.\2\ 
Below, we discuss the key hazard characteristics that the review 
identified.
---------------------------------------------------------------------------

    \2\ The data collected for the Commission's study are based on 
information reported to the Commission through various sources. The 
reports are not a complete set of all incidents that have occurred, 
nor do they constitute a statistical sample representing all ROV-
related incidents with at least one death or injury resulting. 
Additionally, reporting is ongoing for ROV-related incidents that 
occurred in the specified time frame. The Commission is expecting 
additional reports and information on ROV-related incidents that 
resulted in a death or injury and that occurred in the given time 
frame.
---------------------------------------------------------------------------

1. Rollover
    Of the 428 reported ROV-related incidents, 291 (68 percent) 
involved rollover of the vehicle, more than half of which occurred 
while the vehicle was in a turn (52 percent). Of the 224 fatal 
incidents, 147 (66 percent) involved rollover of the vehicle, and 56 of 
those incidents (38 percent) occurred on flat terrain. The slope of the 
terrain is unknown in 39 fatal incidents.
    A total of 826 victims were involved in the 428 reported incidents, 
including 231 fatalities and 388 injuries. Of the 231 reported 
fatalities, 150 (65 percent) died in an incident involving lateral 
rollover of the ROV. Of the 388 injured victims, 75 (19 percent) were 
classified as being severely injured; 67 of these victims (89 percent) 
were injured in incidents that involved lateral rollover of the ROV.
2. Occupant Ejection and Seat Belt Use
    From the 428 ROV-related incidents reviewed by CPSC, 817 victims 
were reported to be in or on the ROV during the incident, and 610 (75 
percent) were known to have been injured or killed. Seatbelt use is 
known for 477 of the 817 victims; of these, 348 (73 percent) were not 
wearing a seatbelt at the time of the incident.
    Of the 610 fatally and nonfatally injured victims who were in or on 
the ROV, 433 (71 percent) were partially or fully ejected from the ROV; 
and 269 (62 percent) of these victims were struck by

[[Page 68966]]

a part of the vehicle, such as the roll cage or side of the ROV, after 
ejection. Seat belt use is known for 374 of the 610 victims; of these, 
282 (75 percent) were not wearing a seat belt.
    Of the 225 fatal victims who were in or on the ROV at the time of 
the incident, 194 (86 percent) were ejected partially or fully from the 
vehicle, and 146 (75 percent) were struck by a part of the vehicle 
after ejection. Seat belt use is known for 155 of the 194 ejected 
victims; of these, 141 (91 percent) were not wearing a seat belt.

C. NEISS Data

    To estimate the number of nonfatal injuries associated with ROVs 
that were treated in a hospital emergency department, CPSC undertook a 
special study to identify cases that involved ROVs that were reported 
through the National Electronic Injury Surveillance System (NEISS) from 
January 1, 2010 to August 31, 2010.\3\
---------------------------------------------------------------------------

    \3\ NEISS is a stratified national probability sample of 
hospital emergency departments that allows the Commission to make 
national estimates of product-related injuries. The sample consists 
of about 100 of the approximately 5,400 U.S. hospitals that have at 
least six beds and provide 24-hour emergency service. Consumer 
product-related injuries treated in emergency departments of the 
NEISS-member hospitals are coded from the medical record. As such, 
information about the injury is extracted, but specifics about the 
product and its use are often not available.
---------------------------------------------------------------------------

    NEISS does not contain a separate category or product code for 
ROVs. Injuries associated with ROVs are usually assigned to an ATV 
product category (NEISS product codes 3286--3287) or to the utility 
vehicle (UTV) category (NEISS product code 5044). A total of 2,018 
injuries that were related to ATVs or UTVs were recorded in NEISS 
between January 1, 2010 and August 31, 2010. The Commission attempted 
follow-up interviews with each victim (or a relative of the victim) to 
gather more information about the incidents and the vehicles involved. 
CPSC determined whether the vehicle involved was an ROV based on the 
make and model of the vehicle reported in the interviews. If the make 
and model of the vehicle was not reported, staff did not count the case 
as involving an ROV.
    A total of 688 surveys were completed, resulting in a 33 percent 
response rate for this survey. Of the 688 completed surveys, 16 were 
identified as involving an ROV based on the make and model of the 
vehicle involved. It is possible that more cases involved an ROV, but 
it was not possible to identify them due to lack of information on the 
vehicle make and model.
    The estimated number of emergency department-treated ROV-related 
injuries occurring in the United States between January 1, 2010 and 
August 31, 2010, is 2,200 injuries. Extrapolating for the year 2010, 
the estimated number of emergency department-treated, ROV-related 
injuries is 3,000, with a corresponding 95 percent confidence interval 
of 1,100 to 4,900.

D. Yamaha Rhino Repair Program

    CPSC staff began investigating ROVs following reports of serious 
injuries and fatalities associated with the Yamaha Rhino. In March 
2009, CPSC staff negotiated a repair program on the Yamaha Rhino 450, 
660, and 700 model ROVs to address stability and handling issues with 
the vehicles.\4\ CPSC staff investigated more than 50 incidents, 
including 46 driver and passenger deaths related to the Yamaha Rhino. 
The manufacturer voluntarily agreed to design changes through a repair 
program that would increase the vehicle's lateral stability and change 
the vehicle's handling characteristic from oversteer to understeer. The 
repair consisted of the following: (1) Addition of 50-mm spacers on the 
vehicle's rear wheels to increase the track width, and (2) the removal 
of the rear stabilizer bar to effect understeer characteristics.
---------------------------------------------------------------------------

    \4\ CPSC Release #09-172, March 31, 2009, Yamaha Motor Corp. 
Offers Free Repair for 450, 660, and 700 Model Rhino Vehicles.
---------------------------------------------------------------------------

    CPSC staff reviewed reports of ROV-related incidents reported to 
the CPSC between January 1, 2003 and May 31, 2012, involving Yamaha 
Rhino model vehicles. (The data are only those reported to CPSC staff 
and are not representative of all incidents.) The number of incidents 
that occurred by quarters of a year are shown below in Figure 1.

[[Page 68967]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.000

    After the repair program was initiated in March 2009, the number of 
reported incidents involving a Yamaha Rhino ROV decreased noticeably.
    CPSC staff also analyzed the 242 Yamaha Rhino-related incidents 
reported to CPSC and identified 46 incidents in which a Yamaha Rhino 
vehicle rolled over during a turn on flat or gentle terrain. Staff 
identified forty-one of the 46 incidents as involving an unrepaired 
Rhino vehicle. In comparison, staff identified only two of the 46 
incidents in which a repaired Rhino vehicle rolled during a turn, and 
each of these incidents occurred on terrain with a 5 to 10 degree 
slope. Among these 41 reported incidents, there were no incidents 
involving repaired Rhinos rolling over on flat terrain during a turn.
    The Commission believes the decrease in Rhino-related incidents 
after the repair program was initiated can be attributed to the vehicle 
modifications made by the repair program. Specifically, correction of 
oversteer and improved lateral stability can reduce rollover incidents 
by reducing the risk of sudden and unexpected increases in lateral 
acceleration during a turn, and increasing the amount of force required 
to roll the vehicle over. CPSC believes that lateral stability and 
vehicle handling have the most effect on rollovers during a turn on 
level terrain because the rollover is caused primarily by lateral 
acceleration generated by friction during the turn. Staff's review of 
rollover incidents during a turn on level ground indicates that 
repaired Rhino vehicles are less likely than unrepaired vehicles to 
roll over. CPSC believes this is further evidence that increasing 
lateral stability and correcting oversteer to understeer contributed to 
the decrease in Yamaha Rhino incidents.

IV. Statutory Authority

    ROVs are ``consumer products'' that can be regulated by the 
Commission under the authority of the CPSA. See 15 U.S.C. 2052(a). 
Section 7 of the CPSA authorizes the Commission to promulgate a 
mandatory consumer product safety standard that sets forth certain 
performance requirements for a consumer product or that sets forth 
certain requirements that a product be marked or accompanied by clear 
and adequate warnings or instructions. A performance, warning, or 
instruction standard must be reasonably necessary to prevent or reduce 
an unreasonable risk or injury. Id.
    Section 9 of the CPSA specifies the procedure the Commission must 
follow to issue a consumer product safety standard under section 7. In 
accordance with section 9, the Commission may commence rulemaking by 
issuing an ANPR; as noted previously, the Commission issued an ANPR on 
ROVs in October 2009. Section 9 authorizes the Commission to issue an 
NPR including the proposed rule and a preliminary regulatory analysis 
in accordance with section 9(c) of the CPSA and request comments 
regarding the risk of injury identified by the Commission, the 
regulatory alternatives being considered, and other possible 
alternatives for addressing the risk. Id. 2058(c). Next, the Commission 
will consider the comments received in response to the proposed rule 
and decide whether to issue a final rule along with a final regulatory 
analysis. Id. 2058(c)-(f). The Commission also will provide an 
opportunity for interested persons to make oral presentations of the 
data, views, or arguments, in accordance with section 9(d)(2) of the 
CPSA. Id. 2058(d)(2).
    According to section 9(f)(1) of the CPSA, before promulgating a 
consumer product safety rule, the Commission must consider, and make 
appropriate

[[Page 68968]]

findings to be included in the rule, concerning the following issues: 
(1) The degree and nature of the risk of injury that the rule is 
designed to eliminate or reduce; (2) the approximate number of consumer 
products subject to the rule; (3) the need of the public for the 
products subject to the rule and the probable effect the rule will have 
on utility, cost, or availability of such products; and (4) the means 
to achieve the objective of the rule while minimizing adverse effects 
on competition, manufacturing, and commercial practices. Id. 
2058(f)(1).
    According to section 9(f)(3) of the CPSA, to issue a final rule, 
the Commission must find that the rule is ``reasonably necessary to 
eliminate or reduce an unreasonable risk of injury associated with such 
product'' and that issuing the rule is in the public interest. Id. 
2058(f)(3)(A)&(B). In addition, if a voluntary standard addressing the 
risk of injury has been adopted and implemented, the Commission must 
find that: (1) The voluntary standard is not likely to eliminate or 
adequately reduce the risk of injury, or that (2) substantial 
compliance with the voluntary standard is unlikely. Id. 2058(f)(3(D). 
The Commission also must find that expected benefits of the rule bear a 
reasonable relationship to its costs and that the rule imposes the 
least burdensome requirements that would adequately reduce the risk of 
injury. Id. 2058(f)(3)(E)&(F).
    Other provisions of the CPSA also authorize this rulemaking. 
Section 27(e) provides the Commission with authority to issue a rule 
requiring consumer product manufacturers to provide the Commission with 
such performance and technical data related to performance and safety 
as may be required to carry out the CPSA and to give such performance 
and technical data to prospective and first purchasers. Id. 2076(e). 
This provision bolsters the Commission's authority under section 7 to 
require provision of safety-related information, such as hang tags.

V. Overview of Proposed Requirements

    Based on incident data, vehicle testing, and experience with the 
Yamaha Rhino repair program, the Commission believes that improving 
lateral stability (by increasing rollover resistance) and improving 
vehicle handling (by correcting oversteer to understeer) are the most 
effective approaches to reducing the occurrence of ROV rollover 
incidents. ROVs with higher lateral stability are less likely to roll 
over because more lateral force is necessary to cause rollover than an 
ROV with lower lateral stability. ROVs exhibiting understeer during a 
turn are less likely to rollover because steering control is stable and 
the potential for the driver to lose control is low.
    The Commission believes that when rollovers do occur, improving 
occupant protection performance (by increasing seat belt use) will 
mitigate injury severity. CPSC's analysis of ROV incidents indicates 
that 91 percent of fatally ejected victims were not wearing a seat belt 
at the time of the incident. Increasing seat belt use, in conjunction 
with better shoulder retention performance, will significantly reduce 
injuries and deaths associated with an ROV rollover event.
    To address these hazards, the Commission is proposing requirements 
for:
     A minimum level of rollover resistance of the ROV when 
tested using the J-turn test procedure;
     A hang tag providing information about the vehicle's 
rollover resistance on a progressive scale;
     Understeer performance of the ROV when tested using the 
constant radius test procedure;
     Limited maximum speed of the ROV when tested with occupied 
front seat belts unbuckled; and
     A minimum level of passive shoulder protection when using 
a probe test.

VI. CPSC Technical Analysis and Basis for Proposed Requirements

A. Overview of Technical Work

    In February 2010, the Commission contracted SEA, Limited (SEA) to 
conduct an in-depth study of vehicle dynamic performance and static 
rollover measures for ROVs. SEA evaluated a sample of 10 ROVs that 
represented the recreational and utility oriented ROVs available in the 
U.S. market that year. SEA tested and measured several characteristics 
and features that relate to the rollover performance of the vehicles 
and to the vehicle's handling characteristics.
    In 2011, SEA designed and built a roll simulator to measure and 
analyze occupant response during quarter-turn roll events of a wide 
range of machines, including ROVs. The Commission contracted with SEA 
to conduct occupant protection performance evaluations of seven ROVs 
with differing occupant protection designs.\5\
---------------------------------------------------------------------------

    \5\ SEA's reports are available on CPSC's Web site at: https://www.cpsc.gov/en/Research-Statistics/Sports-Recreation/ATVs/Technical-Reports/.
---------------------------------------------------------------------------

B. Lateral Stability

1. Definitions
    Following are definitions of basic terms used in this section.
     Lateral acceleration: acceleration that generates the 
force that pushes the vehicle sideways. During a turn, lateral 
acceleration is generated by friction between the tires and surface. 
Lateral acceleration is expressed as a multiple of free-fall gravity 
(g).
     Two-wheel lift: point at which the inside wheels of a 
turning vehicle lift off the ground, or when the uphill wheels of a 
vehicle on a tilt table lift off the table. Two-wheel lift is a 
precursor to a rollover event. We use the term ``two-wheel lift'' 
interchangeably with ``tip-up.''
     Threshold lateral acceleration: minimum lateral 
acceleration of the vehicle at two-wheel lift.
     Untripped rollover: rollover that occurs during a turn due 
solely to the lateral acceleration generated by friction between the 
tires and the road surface.
     Tripped rollover: rollover that occurs when the vehicle 
slides and strikes an object that provides a pivot point for the 
vehicle to roll over.
2. Static Measures to Evaluate ROV Lateral Stability
    CPSC and SEA evaluated the static measurements of the static 
stability factor (SSF) and tilt table ratio (TTR) to compare lateral 
stability of a group of 10 ROVS.
a. Static Stability Factor (SSF)
    SSF approximates the lateral acceleration in units of gravitational 
acceleration (g) at which rollover begins in a simplified vehicle that 
is assumed to be a rigid body without suspension movement or tire 
deflections. NHTSA uses rollover risk as determined by dynamic test 
results and SSF values to evaluate passenger vehicle rollover 
resistance for the New Car Assessment Program (NCAP).\6\ SSF relates 
the track width of the vehicle to the height of the vehicle center of 
gravity (CG), as shown in Figure 2. Loading condition is important 
because CG height and track width vary, depending on the vehicle load 
condition. Mathematically, the relationship is track width (T) divided 
by two times the CG height (H), or SSF=T/2H. Higher values for SSF 
indicate higher lateral stability, and lower SSF values indicate lower 
lateral stability.
---------------------------------------------------------------------------

    \6\ NHTSA, 68 FR 59250, ``Consumer Information; New Car 
Assessment Program; Rollover Resistance,'' (Oct. 14, 2003).

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

[[Page 68969]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.001

    SEA measured track width and CG height values for the sample group 
of 10 ROVs. SEA used their Vehicle Inertia Measurement Facility (VIMF), 
which incorporates the results of five different tests to determine the 
CG height. SEA has demonstrated that VIMF CG height measurements are 
repeatable within 0.5 percent of the measured values.\7\ 
Using the CG height and track width measurement, SEA calculated SSF 
values for several different load conditions. (See Table 1).
---------------------------------------------------------------------------

    \7\ Heydinger, Gary J., et al, The Design of a Vehicle Inertia 
Measurement Facility, SAE 950309, 1995.

                           Table 1--SSF Values
------------------------------------------------------------------------
                   Vehicle rank  (SSF)                          SSF
------------------------------------------------------------------------
F.......................................................           0.881
A.......................................................           0.887
H.......................................................           0.918
B.......................................................           0.932
D.......................................................           0.942
J.......................................................           0.962
E.......................................................           0.965
C.......................................................           0.991
G.......................................................           1.031
I.......................................................           1.045
------------------------------------------------------------------------

b. Tilt Table Ratio (TTR)
    SEA conducted tilt table tests on the ROV sample group. In this 
test, the vehicles in various loaded conditions were placed on a rigid 
platform, and the angle of platform tilt was increased (see Figure 3) 
until both upper wheels of the vehicle lifted off the platform. The 
platform angle at two-wheel lift is the Tilt Table Angle (TTA). The 
trigonometric tangent of the TTA is the Tilt Table Ratio (TTR). TTA and 
TTR are used to evaluate the stability of the vehicle. Larger TTA and 
TTR generally correspond to better lateral stability, except these 
measures do not account for dynamic tire deflections or dynamic 
suspension compliances. Tilt testing is a quick and simple static test 
that does not require sophisticated instrumentation. Tilt testing is 
used as a rollover metric in the voluntary standards created by the 
Recreational Off-Highway Vehicle Association (ROHVA) and the Outdoor 
Power Equipment Institute (OPEI). TTA and TTR values measured by SEA 
are shown in Table 2.\8\
---------------------------------------------------------------------------

    \8\ ROHVA developed ANSI/ROHVA 1 for recreation-oriented ROVs 
and OPEI developed ANSI/OPEI B71.0 for utility-oriented ROVs.
[GRAPHIC] [TIFF OMITTED] TP19NO14.002


[[Page 68970]]



                       Table 2--TTA and TTR Values
------------------------------------------------------------------------
                                    TTA        Vehicle rank
      Vehicle rank  (TTA)          (deg.)          (TTR)           TTR
------------------------------------------------------------------------
A..............................      33.0   A.................     0.650
B..............................      33.6   B.................     0.664
D..............................      33.7   D.................     0.667
I..............................      35.4   I.................     0.712
H..............................      35.9   H.................     0.724
J..............................      36.1   J.................     0.730
F..............................      36.4   F.................     0.739
E..............................      38.1   E.................     0.784
C..............................      38.8   C.................     0.803
G..............................      39.0   G.................     0.810
------------------------------------------------------------------------

    Because ROVs are designed with long suspension travel and soft 
tires for off-road performance, staff was concerned that SSF and TTR 
would not accurately characterize the dynamic lateral stability of the 
vehicle. Therefore, CPSC's contractor, SEA, conducted dynamic J-turn 
tests to determine whether SSF or TTR measurement corresponded with 
actual dynamic measures for lateral stability.
3. Dynamic Test To Measure ROV Lateral Stability--the J-Turn Test
    In 2001, NHTSA evaluated the J-turn test (also called drop-throttle 
J-turn testing and step-steer testing) as a method to measure rollover 
resistance of automobiles. NHTSA found the J-turn test to be the most 
objective and repeatable method for vehicles with low rollover 
resistance. Specifically, the J-turn test is objective because a 
programmable steering machine turns the steering wheel during the test, 
and the test results show that the vehicle speed, lateral acceleration, 
and roll angle data observed during J-turn tests were highly 
repeatable.\9\ However, NHTSA determined that although the J-turn test 
is the most objective and repeatable method for vehicles with low 
rollover resistance, the J-turn test is unable to measure the high 
rollover resistance of most passenger automobiles.\10\ On pavement 
where a high-friction surface creates high lateral accelerations, 
vehicles with high rollover resistance (such as passenger automobiles) 
will lose tire traction and slide in a severe turn rather than roll 
over. The threshold lateral acceleration cannot be measured because 
rollover does not occur. In contrast, vehicles with low rollover 
resistance exhibit untripped rollover on a pavement during a J-turn 
test, and the lateral acceleration at rollover threshold can be 
measured. Thus, the J-turn test is the most appropriate method to 
measure the rollover resistance of ROVs because ROVs exhibit untripped 
rollover during the test.
---------------------------------------------------------------------------

    \9\ Forkenbrock, G. and Garrott, W. (2002). A Comprehensive 
Experimental Evaluation of Test Maneuvers That May Induce On-Road, 
Untripped, Light Vehicle Rollover Phase IV of NHTSA's Light Vehicle 
Rollover Research Program. DOT HS 809 513.
    \10\ Forkenbrock, G. and Garrott, W. (2002). A Comprehensive 
Experimental Evaluation of Test Maneuvers That May Induce On-Road, 
Untripped, Light Vehicle Rollover Phase IV of NHTSA's Light Vehicle 
Rollover Research Program. DOT HS 809 513.
---------------------------------------------------------------------------

    J-turn tests are conducted by driving the test vehicle in a 
straight path, releasing (dropping) the throttle, and rapidly turning 
the steering wheel to a specified angle once the vehicle slows to a 
specified speed. The steering wheel angle and vehicle speed are 
selected to produce two-wheel lift of the vehicle. Outriggers, which 
are beams that extend to either side of a vehicle, allow the vehicle to 
roll but prevent full rollover. The sequence of events in the test 
procedure is shown in Figure 4. SEA conducted drop-throttle J-turn 
tests to measure the minimum lateral accelerations necessary to cause 
two-wheel lift (shown in Step 3 of Figure 4) for each vehicle. Side 
loading of the vehicle occurs naturally as a result of the lateral 
acceleration that is created in the J-turn and this lateral 
acceleration can be measured and recorded. The lateral acceleration 
produced in the turn is directly proportional to the side loading force 
acting to overturn the vehicle according to the equation F = 
(m)(Ay), where F is force, m is the mass of the vehicle, and 
Ay is lateral acceleration.

[[Page 68971]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.003

    SEA conducted the J-turn testing at 30 mph. A programmable steering 
controller input the desired steering angles at a steering rate of 500 
degrees per second for all vehicles. The chosen steering rate of 500 
degrees per second is high enough to approximate a step input, but 
still within the capabilities of a driver. (A step input is one that 
happens instantly and requires no time to complete. For steering input, 
time is required to complete the desired steering angle, so a steering 
step input is approximated by a high angular rate of steering input.) 
SEA conducted preliminary tests by starting with a relatively low 
steering angle of 80 to 90 degrees and incrementally increasing the 
steering angle until two-wheel lift was achieved. When SEA determined 
the steering angle that produced a two-wheel lift, SEA conducted the 
test run for that vehicle load condition. For each test run, SEA 
recorded the speed, steering angle, roll rate, and acceleration in 
three directions (longitudinal, lateral, and vertical). SEA processed 
and plotted the data to determine the minimum lateral acceleration 
required for two-wheel lift of the vehicle.
    The J-turn test is a direct measure of the minimum or threshold 
lateral acceleration required to initiate a rollover event, or tip-up 
of the test vehicle when turning. ROVs that exhibit higher threshold 
lateral acceleration have a higher rollover resistance or are more 
stable than ROVs with lower threshold lateral accelerations. Each of 
the 10 ROVs tested in the study by SEA exhibited untripped rollover in 
the J-turn tests at steering wheel angles ranging from 93.8 to 205 
degrees and lateral accelerations ranging from 0.625 to 0.785 g. Table 
3 shows the vehicles arranged in ascending order for threshold lateral 
acceleration (Ay) at tip up, SSF, TTA, and TTR. Table 3 
illustrates the lack of correlation of the static metrics (SSF, TTA, or 
TTR) with the direct dynamic measure of threshold lateral acceleration 
(Ay) at tip up.

[[Page 68972]]



                                 Table 3
------------------------------------------------------------------------
        Vehicle rank (A)y             Ay(g)         SSF          TTR
------------------------------------------------------------------------
D................................        0.625        0.942        0.667
B................................        0.655        0.932        0.664
A................................        0.670        0.887        0.650
J................................        0.670        0.962        0.730
I................................        0.675        1.045        0.712
F................................        0.690        0.881        0.739
E................................        0.700        0.965        0.784
H................................        0.705        0.918        0.724
C................................        0.740        0.991        0.803
G................................        0.785        1.031        0.810
------------------------------------------------------------------------
Adapted from: Heydinger, G. (2011). Vehicle Characteristics Measurements
  of Recreational Off-Highway Vehicles--Additional Results for Vehicle
  J. Retrieved from https://www.cpsc.gov/PageFiles/93928/rovj.pdf.

    SEA also conducted J-turn tests on four ROVs to measure the 
repeatability of the lateral acceleration measurements and found the 
tests to be very repeatable.\11\ The results of the repeatability tests 
indicate the standard deviation for sets of 10 test runs (conducted in 
opposite directions and left/right turn directions) ranged from 0.002 g 
to 0.013 g.
---------------------------------------------------------------------------

    \11\ Heydinger, G. (2013). Repeatability of J-Turn Testing of 
Four Recreational Off-Highway Vehicles. Retrieved from https://www.cpsc.gov//Global/Research-and-Statistics/Injury-Statistics/Sports-and-Recreation/ATVs/SEAReporttoCPSCRepeatabilityTestingSeptember%202013.pdf.
---------------------------------------------------------------------------

    Comparison of the SSF, TTR, and Ay values for each ROV 
indicate that there is a lack of correspondence between the static 
metrics (SSF and TTR) and the direct measurement of threshold lateral 
acceleration at rollover. Static metrics cannot be used to evaluate ROV 
rollover resistance because static tests are unable to account fully 
for the dynamic tire deflections and suspension compliance exhibited by 
the ROVs during a J-turn maneuver. Therefore, the Commission believes 
that the lateral acceleration threshold at rollover is the most 
appropriate metric to use when measuring and comparing rollover 
resistance for ROVs.

C. Vehicle Handling

1. Basic Terms
     Understeer: Path of vehicle during a turn in which the 
vehicle steers less into a turn than the steering wheel angle input by 
the driver. If the driver does not correct for the understeer path of 
the vehicle, the vehicle continues on a straighter path than intended 
(see Figure 5).
     Oversteer: Path of vehicle during a turn in which the 
vehicle steers more into a turn than the steering wheel angle input by 
the driver. If the driver does not correct for the oversteer path of 
the vehicle, the vehicle spirals into the turn more than intended (see 
Figure 5).
     Sub-limit understeer or sub-limit oversteer: Steering 
condition that occurs while the tires have traction on the driving 
surface.
     Limit understeer or limit oversteer: Steering condition 
that occurs when the traction limits of the tires have been reached and 
the vehicle begins to slide.
[GRAPHIC] [TIFF OMITTED] TP19NO14.004

2. Staff's Technical Work
a. Constant Radius Test
    SAE International (formerly Society of Automotive Engineers) 
standard, SAE J266, Surface Vehicle Recommended Practice, Steady-State 
Directional Control Test Procedures for Passenger Cars and Light 
Trucks, establishes test procedures to measure the vehicle handling 
properties of passenger cars and light trucks. ROVs obey the same 
principles of motion as automobiles because ROVs and automobiles share 
key characteristics, such as pneumatic

[[Page 68973]]

tires, a steering wheel, and spring-damper suspension that contribute 
to the dynamic response of the vehicle.\12\ Thus, the test procedures 
to measure the vehicle handling properties of passenger cars and light 
trucks are also applicable to ROVs.
---------------------------------------------------------------------------

    \12\ See Tab A of the CPSC staff's briefing package.
---------------------------------------------------------------------------

    SEA used the constant radius test method, described in SAE J266, to 
evaluate the sample ROVs' handling characteristics. The test consists 
of driving each vehicle on a 100 ft. radius circular path from very low 
speeds, up to the speed where the vehicle experiences two-wheel lift or 
cannot be maintained on the path of the circle. The test vehicles were 
driven in the clockwise and counterclockwise directions. For a constant 
radius test, ``understeer'' is defined as the condition when the 
steering wheel angle required to maintain the circular path increases 
as the vehicle speed increases because the vehicle is turning less than 
intended. ``Neutral steer'' is defined as the condition when the 
steering wheel angle required to maintain the circular path is 
unchanged as the vehicle speed increases. ``Oversteer'' is defined as 
the condition when the average steering wheel input required to 
maintain the circular path decreases as the vehicle speed increases 
because the vehicle is turning more than intended.
    SEA tested 10 ROVs; five of those vehicles (A, D, F, I, and J) 
exhibited sub-limit transitions to oversteer when tested on asphalt 
(see Figure 6). The five remaining vehicles (B, C, E, G, and H) 
exhibited a sub-limit understeer condition for the full range of the 
test.
[GRAPHIC] [TIFF OMITTED] TP19NO14.005

b. Slowly Increasing Steer (SIS) Test
    SAE J266, Surface Vehicle Recommended Practice, Steady-State 
Directional Control Test Procedures for Passenger Cars and Light 
Trucks, also establishes test procedures for the Constant Speed 
Variable Steer Angle Test. SEA calls this test the ``constant speed 
slowly increasing steer (SIS) test.'' During the SIS test, the ROV 
driver maintains a constant speed of 30 mph, and the vehicle's steering 
wheel angle is slowly increased at a rate of 5 degrees per second until 
the ROV reaches a speed limiting condition or tip-up. A programmable 
steering controller (PSC) was used to increase the steering angle at a 
constant rate of 5 degrees per second. During the test, instrumentation 
for speed, steering angle, lateral acceleration, roll angle, and yaw 
rate were recorded. SEA conducted SIS tests on the sample of 10 ROVs.
    Figure 7 shows SIS test data plotted of lateral acceleration versus 
time for Vehicle A and Vehicle H. Vehicle H is the same model vehicle 
as Vehicle A, but Vehicle H is a later model year, where the sub-limit 
oversteer has been corrected to understeer.
    Plots from the ROV SIS tests in Figure 7 illustrate a sudden 
increase in lateral acceleration that is found only in vehicles that 
exhibit sub-limit oversteer. The sudden increase in lateral 
acceleration is exponential and represents a dynamically unstable

[[Page 68974]]

condition.\13\ This condition is undesirable because it can cause a 
vehicle with high lateral stability (such as a passenger car) to spin 
out of control, or it can cause a vehicle with low lateral stability 
(such as an ROV) to roll over suddenly.
---------------------------------------------------------------------------

    \13\ (Gillespie, T. (1992). Fundamentals of Vehicle Dynamics. 
Society of Automotive Engineers, Inc. p. 204-205.)
[GRAPHIC] [TIFF OMITTED] TP19NO14.006

    When Vehicle A reached its dynamically unstable condition, the 
lateral acceleration suddenly increased from 0.50 g to 0.69 g 
(difference of 0.19 g) in less than 1 second, and the vehicle rolled 
over. (Outriggers on the vehicle prevented full rollover of the 
vehicle.) In contrast, Vehicle H never reached a point where the 
lateral acceleration increases exponentially because the condition does 
not develop in understeering vehicles.\14\ The increase in Vehicle H's 
lateral acceleration remains linear, and the lateral acceleration 
increase from 0.50 g to 0.69 g (same difference of 0.19 g) occurs in 
5.5 seconds.
---------------------------------------------------------------------------

    \14\ Gillespie, T. (1992). Fundamentals of Vehicle Dynamics. 
Society of Automotive Engineers.
---------------------------------------------------------------------------

    SEA test results indicate that ROVs that exhibited sub-limit 
oversteer also exhibited a sudden increase in lateral acceleration that 
caused the vehicle to roll over. An ROV that exhibits this sudden 
increase in lateral acceleration is directionally unstable and 
uncontrollable.\15\
---------------------------------------------------------------------------

    \15\ Gillespie, T. (1992). Fundamentals of Vehicle Dynamics. 
Society of Automotive Engineers, Inc. p. 204-205; Bundorf, R. T. 
(1967). The Influence of Vehicle Design Parameters on Characteristic 
Speed and Understeer. SAE 670078; Segel, L. (1957). Research in the 
Fundamentals of Automobile Control and Stability. SAE 570044.
---------------------------------------------------------------------------

    Plots of the vehicle path during SIS tests illustrate further how 
an oversteering ROV (Vehicle A) will roll over earlier in a turn than 
an understeering ROV (Vehicle H), when the vehicles are operated at the 
same speed and steering rate (see Figure 8). Vehicle A and Vehicle H 
follow the same path until Vehicle A begins to oversteer and its turn 
radius becomes smaller. Vehicle A becomes dynamically unstable, its 
lateral acceleration increases exponentially, and the vehicle rolls 
over suddenly. In contrast, Vehicle H continues to travel 300 more feet 
in the turn before the vehicle reaches its threshold lateral 
acceleration and rolls over. A driver in Vehicle H has more margin (in 
time and distance) to correct the steering to prevent rollover than a 
driver in Vehicle A because Vehicle H remains in understeer during the 
turn, while Vehicle A transitions to oversteer and becomes dynamically 
unstable.

[[Page 68975]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.007

    The Commission believes that tests conducted by SEA provide strong 
evidence that sub-limit oversteer in ROVs is an unstable condition that 
can lead to a rollover incident, especially given the low rollover 
resistance of ROVs. All ROVs that exhibited sub-limit oversteer reached 
a dynamically unstable condition during a turn where the increase in 
lateral acceleration suddenly became exponential. The CPSC believes 
this condition can contribute to ROV rollover on level ground, and 
especially on pavement.

D. Occupant Protection

1. Overview and Basic Terms
    The open compartment configuration of ROVs is intentional and 
allows for easy ingress and egress, but the configuration also 
increases the likelihood of complete or partial ejection of the 
occupants in a rollover event. ROVs are equipped with a ROPS, seat 
belts, and other restraints for the protection of occupants (see Figure 
9). Occupants who remain in the ROV and surrounded by the ROPS, an area 
known as the protective zone, are generally protected from being 
crushed by the vehicle during a quarter-turn rollover. Seat belts are 
the primary restraint for keeping occupants within the protective zone 
of the ROPS.

[[Page 68976]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.008

NHTSA evaluates the occupant protection performance of passenger 
vehicles with tests that simulate vehicle collisions and tests that 
simulate vehicle rollover.\16\ The NHTSA tests use anthropometric test 
devices (ATDs), or crash test dummies, to evaluate occupant excursion 
and injury severity during the simulation tests. The occupant movement 
during these tests is called occupant kinematics. Occupant kinematics 
is defined as the occupant's motion during a crash event, including the 
relative motion between various body parts. Occupant kinematics is an 
important element of dynamic tests because forces act on an occupant 
from many different directions during a collision or rollover.
---------------------------------------------------------------------------

    \16\ Federal Motor Vehicle Safety Standard (1971) 49 CFR 
571.208.
---------------------------------------------------------------------------

    There are no standardized tests to evaluate the occupant protection 
performance of ROVs. However, a test to evaluate occupant protection 
performance in ROVs should be based on simulations of real vehicle 
rollover. In a rollover event, the vehicle experiences lateral 
acceleration and lateral roll. A valid simulation of an ROV rollover 
will reproduce the lateral acceleration and the roll rate experienced 
by an ROV during a real rollover event.
2. Seat belts
a. Seat Belt Use in Incidents
    From the 428 ROV-related incidents reviewed by the Commission, 817 
victims were reported to be in or on the ROV at the time of the 
incident, and 610 (75 percent) were known to have been injured or 
killed. Seatbelt use is known for 477 of the 817 victims; of these, 348 
(73 percent) were not wearing a seatbelt at the time of the incident.
    Of the 610 fatal and nonfatal victims who were in or on the ROV at 
the time of the incident, 433 (71 percent) were ejected partially or 
fully from the ROV, and 269 (62 percent) of these victims were struck 
by a part of the vehicle, such as the roll cage or side of the ROV, 
after ejection. Seat belt use is also known for 374 of the 610 victims; 
of these, 282 (75 percent) were not wearing a seat belt.
    Of the 225 fatal victims who were in or on the ROV at the time of 
the incident, 194 (86 percent) were ejected partially or fully from the 
vehicle, and 146 (75 percent) were struck by a part of the vehicle 
after ejection. Seat belt use is known for 155 of the 194 ejected 
victim; of these, 141 (91 percent) were not wearing a seat belt.
    A total of 826 victims were involved in the 428 ROV-related 
incidents reviewed the Commission's multidisciplinary team. Of these 
victims, 353 (43 percent) were known to be driving the ROV, and 203 (24 
percent) were known to be a passenger in the front seat of the ROV. Of 
the 231 reported fatalities, 141 (61 percent) were the driver of the 
ROV, and 49 (21 percent) were the right front passenger in an ROV.
    ROHVA also performed an analysis of hazard and risk issues 
associated with ROV-related incidents and determined that lack of seat 
belt use is the top incident factor.\17\ ROHVA has stated: ``Based on 
the engineering judgment of its members and its review of ROV incident 
data provided by the CPSC, ROHVA concludes that the vast majority of 
hazard patterns associated with ROV rollover would be eliminated 
through proper seat belt use alone.'' \18\
---------------------------------------------------------------------------

    \17\ Heiden, E. (2009). Summary of Recreational Off-Highway 
Vehicle (ROV) Hazard Analysis. Memorandum from E. Heiden to P. 
Vitrano. Docket No. CPSC-2009-0087. Regulations.gov.
    \18\ Yager, T. (2011) Letter to Caroleene Paul. 18 Apr. 2011. 
Recreational Off-Highway Vehicle Association (ROHVA) written 
response to CPSC staff's ballot on proposed American National 
Standard ANSI/ROHVA 1-201X.
---------------------------------------------------------------------------

a. Literature Review (Automotive)
    CPSC staff reviewed the substantial body of literature on seat belt 
use in automobiles. (See Tab I of staff's briefing package.) Although 
seat belts are one of the most effective strategies for avoiding death 
and injury in motor vehicle crashes, seat belts are only effective if 
they are used.
    Strategies for increasing seat belt use in passenger vehicles date 
to January 1, 1972, when NHTSA required all new cars to be equipped 
with passive restraints or with a seat belt reminder system that used a 
visual flashing light and audible buzzer that activated continuously 
for one minute if the vehicle was placed in gear with occupied front 
seat belts not belted. In 1973, NHTSA required that all new cars be 
equipped with an ignition interlock that allowed the vehicle to start 
only if the driver was belted. The ignition interlock was meant to be 
an interim measure until passive airbag technology matured, but public 
opposition to the technology led Congress to rescind the legislation 
and to prohibit NHTSA from requiring either ignition interlocks or 
continuous audible warnings that last more than 8 seconds. NHTSA then 
revised the Federal Motor Vehicle Safety Standard (FMVSS) to require a

[[Page 68977]]

seat belt reminder with warning light and audible buzzer that lasts 4 
seconds to 8 seconds when front seat belts are not fastened at the time 
of ignition. This standard still applies today (15 U.S.C. 1410 (b)).
    Work by NHTSA indicates seat belt users can be separated loosely 
into three categories: Full-time users, part-time users, and nonusers. 
Part-time users and nonusers give different reasons for not wearing 
seat belts. Part-time seat belt users consistently cite forgetfulness 
and perceived low risk, such as driving short distances or on familiar 
roads, as reasons for not using seat belts.\19\
---------------------------------------------------------------------------

    \19\ Block, 1998; Bradbard et al., 1998; Harrison and 
Senserrick, 2000; Bentley et al., 2003; Boyle and Vanderwolf, 2003; 
Eby et al., 2005; Boyle and Lampkin, 2008.
---------------------------------------------------------------------------

    One approach to increasing vehicle occupant seat belt use is to 
provide in-vehicle reminders to encourage occupants to fasten their 
seat belts. However, possible systems vary considerably in design, 
intrusiveness, and, most importantly, effectiveness.
    Observational studies of cars equipped with the original NHTSA-
required seat belt reminders found no significant difference in seat 
belt use among vehicles equipped with the continuous one minute visual-
audio system and vehicles not equipped with the reminder system.\20\ 
After NHTSA adopted the less stringent 4-second to 8-second visual and 
audio reminder system requirements, NHTSA conducted observational and 
phone interview studies and concluded that the less intrusive reminder 
system was also not effective in increasing seat belt use.\21\
---------------------------------------------------------------------------

    \20\ Robertson, L. S. and Haddon, W. (1974). The Buzzer-Light 
Reminder System and Safety Belt Use. American Journal of Public 
Health, Vol. 64, No. 8, pp. 814-815.; Robertson, L. S. (1975). 
Safety Belt Use in Automobiles with Starter-Interlock and Buzzer-
Light Reminder Systems. American Journal of Public Health, Vol. 65, 
No. 12, pp. 1319-1325.
    \21\ Westefeld, A. and Phillips, B. M. (1976). Effectiveness of 
Various Safety Belt Warning Systems. (DOT HS 801 953). Washington, 
DC: National Highway Traffic Safety Administration, U.S. Department 
of Transportation.
---------------------------------------------------------------------------

    A national research project by the University of Michigan 
Transportation Research Institute endeavored to promote safety belt use 
in the United States by developing an effective in-vehicle safety belt 
reminder system.\22\ The project authors performed literature reviews 
and conducted surveys and focus groups to design an optimal safety belt 
reminder system. The authors concluded that principles for an optimal 
safety belt reminder system include the following:
---------------------------------------------------------------------------

    \22\ Eby, D. W., Molnar, L. J., Kostyniuk, L. P., and Shope, J. 
T. (2005). Developing an Effective and Acceptable Safety Belt 
Reminder System. 19th International Technical Conference on the 
Enhanced Safety of Vehicles, Washington, DC, June 6-9, 2005. https://www-nrd.nhtsa.dot....01/esv/esv19/05-0171-O.pdf.
---------------------------------------------------------------------------

    1. The full-time safety belt user should not notice the system.
    2. It should be more difficult to cheat on the system than to use 
the safety belt.
    3. Permanent disconnection of the system should be difficult.
    4. The system should be reliable and have a long life.
    5. Crash and injury risk should not be increased as a result of the 
system.
    6. System design should be based on what is known about the 
effectiveness and acceptability of system types and elements.
    7. System design should be compatible with the manufacturer's 
intended purpose/goals for the system.
    NHTSA conducted a study of enhanced seatbelt reminder (ESBR) 
effectiveness that compared results of controlled experiments with 
field observations of actual seat belt use. Among the findings of the 
ESBR effectiveness report are: (1) Systems with only visual reminders 
are not effective; (2) ESBR systems, in general, promote greater seat 
belt use by 3 to 4 percentage points; (3) more annoying systems are 
more effective, but that creates the challenge of designing an 
effective system that is acceptable; (4) potential gains in seat belt 
use not only come from simply reminding users, but also from motivating 
users, such as equating seat belt use with elimination of an annoyance; 
and (5) the positive effects of ESBRs on belt use were more pronounced 
for the low belt-use propensity groups.\23\
---------------------------------------------------------------------------

    \23\ Lerner, N., Singer, J., Huey, R., Jenness, J. (2007). 
Acceptability and Potential Effectiveness of Enhanced Seat Belt 
Reminder System Features. (DOT HS 810 848). Washington, DC: National 
Highway Traffic Safety Administration, U.S. Department of 
Transportation. Freedman, M., Lerner, N., Zador, P., Singer, J., and 
Levi, S. (2009). Effectiveness and Acceptance of Enhanced Seat Belt 
Reminder Systems: Characteristics of Optimal Reminder Systems. (DOT 
HS 811 097). Washington, DC: National Highway Traffic Safety 
Administration, U.S. Department of Transportation.
---------------------------------------------------------------------------

c. Innovative Technologies
    Automobiles. Researchers developed more innovative in-vehicle 
technology, beyond visual and audible warnings, to study the 
effectiveness of systems that hindered a vehicle function if the 
driver's seat belt was not buckled. One system allowed drivers to start 
the vehicle but delayed the driver's ability to place the vehicle in 
gear if the seat belt was not buckled.\24\ Follow-up systems made it 
more difficult for the driver to depress the gas pedal when the vehicle 
exceeded 20-25 mph if the driver's seat belt was not buckled. Study 
participants were more receptive to the latter system, which was a 
consistent and forceful motivator to buckle the seat belt without 
affecting the general operation of the vehicle.\25\
---------------------------------------------------------------------------

    \24\ Van Houten, R., Malenfant, J.E.L., Reagan, I., Sifrit, K., 
Compton, R., & Tenenbaum, J. (2010). Increasing Seat Belt Use in 
Service Vehicle Drivers with a Gearshift Delay. Journal of Applied 
Behavior Analysis, 43, 369-380.
    \25\ Van Houten, R., Hilton, B., Schulman, R., and Reagan, I. 
(2011). Using Haptic Feedback to Increase Seat Belt Use of Service 
Vehicle Drivers. (DOT HS 811 434). Washington, DC: National Highway 
Traffic Safety Administration, U.S. Department of Transportation.
---------------------------------------------------------------------------

    ROVs. In 2010, Bombardier Recreation Products (BRP) introduced the 
Can-Am Commander 1000 ROV with a seat belt speed limiter system that 
restricts the vehicle speed to 9 mph if the driver's seat belt is not 
buckled. CPSC staff performed dynamic tests to verify that the 
vehicle's speed was limited when the driver's seat belt was not 
buckled. On level ground, the vehicle's speed was limited to 6 to 9 mph 
when the driver was unbelted, depending on the ignition key and 
transmission mode selected.
    In 2013, BRP introduced the Can-Am Maverick vehicle as a sport-
oriented ROV that also includes a seat belt speed limiter system. CPSC 
staff did not test the Maverick vehicle because a sample vehicle was 
not available for testing.
    In 2014, Polaris Industries (Polaris) announced that model year 
2015 Ranger and RZR ROVs will include a seatbelt system that limits the 
speed of the vehicle to 15 mph if the seatbelt is not engaged. 
(Retrieved at: https://www.weeklytimesnow.com.au/machine/sidebyside-vehicles-soon-to-get-safety-improvements/story-fnkerd6b-1227023275396.) 
The Commission has not tested these vehicles because they are not yet 
available on the market.
d. User Acceptance of Innovative Technologies in ROVs
    Studies of seat belt reminder systems on automobiles are an 
appropriate foundation for ROV analysis because ROVs are typically 
driven by licensed drivers and the seating environment is similar to an 
automobile. Staff decided to obtain data on ROV users' experience and 
acceptance of seat belt reminders to validate the analysis.
    CPSC staff was not aware of any studies that provide data on the 
effectiveness of seat belt reminder systems on ROVs or user acceptance 
of such technologies. Therefore, the CPSC contracted Westat, Inc. 
(Westat), to conduct focus groups with ROV users to explore their 
opinions of seat belt speed-limitation systems on ROVs. Phase 1 of the 
effort involved

[[Page 68978]]

conducting focus groups of ROV users and asking questions about ROV use 
and user opinions of the Can-Am speed-limitation system that were shown 
in a video to the participants. Results from Phase 1 were used to 
develop the protocol for Phase 2. Phase 2 of the effort conducts focus 
groups of ROV users who provide feedback after driving and interacting 
with an ROV equipped with a speed-limitation system.
    Results of Phase 1 of the Westat study indicate that participants:
     Admit to being part-time seat belt users;
     cite familiarity and low-risk perception as reasons for 
not wearing seat belts;
     value easy ROV ingress and egress over seat belt use;
     generally travel around 5 mph when driving on their own 
property, and overall, drive 15 to 30 mph for typical use;
     had a mixed reaction to the speed-limitation technology at 
10 mph;
     were more accepting of the speed-limitation technology if 
the speed was raised to 15 mph or if the system was tied to a key 
control.
    Phase 2 of the Westat study is ongoing, and a report of the results 
is expected by December 2014. The results will provide data on ROV 
users' acceptance of a seat belt speed limitation technology with a 
threshold speed of 10 mph, 15 mph, and 20 mph. CPSC believes the 
results will provide additional rationale for determining a threshold 
speed for a seat belt speed limitation technology that balances users 
acceptance (as high a speed as possible) with safe operation of the ROV 
without seat belt use (as low a speed as possible).
3. CPSC's Technical Work
    To explore occupant protection performance testing for a product 
for which no standard test protocol exists, CPSC staff contracted 
Active Safety Engineering (ASE) to conduct two exploratory pilot 
studies to evaluate potential test methods. After completion of the 
pilot studies, CPSC staff contracted SEA, Limited (SEA) to conduct 
occupant protection performance evaluation tests, based on a more 
advanced test device designed by SEA.\26\
---------------------------------------------------------------------------

    \26\ The ASE and SEA reports are available on CPSC's Web site 
at: https://www.cpsc.gov/en/Research-Statistics/Sports-Recreation/ATVs/Technical-Reports/.
---------------------------------------------------------------------------

a. Pilot Study 1
    ASE used a HYGE \TM\ accelerator sled to conduct dynamic rollover 
simulations on sample ROVs, occupied by a Hybrid III 50th percentile 
male anthropomorphic test device (ATD). The HYGE \TM\ system causes a 
stationary vehicle, resting on the test sled, to roll over by imparting 
a short-duration lateral acceleration to the test sled. The torso of an 
unbelted ATD ejected partially from the ROV during a simulated 
rollover. In comparison, the torso of a belted ATD remained in the ROV 
during a simulated rollover. The tests demonstrated that use of a seat 
belt prevented full ejection of the ATD's torso.
b. Pilot Study 2
    In a follow-up pilot study, ASE used a deceleration platform sled 
rather than a HYGE \TM\ accelerator sled to impart the lateral 
acceleration to the test vehicle. The deceleration sled is more 
accurate than the HYGETM sled in re-creating the lower energy rollovers 
associated with ROVs.
    An unbelted ATD ejected fully from the vehicle during tests 
conducted at the rollover threshold of the ROV. In comparison, a belted 
ATD partially ejected from the vehicle during tests conducted at the 
same lateral acceleration. These exploratory tests with belted and 
unbelted occupants indicate the importance of using seat belts to 
prevent full ejection of the occupant during a rollover event.
c. SEA Roll Simulator
    SEA designed and built a roll simulator to measure and analyze 
occupant response during quarter-turn roll events of a wide range of 
machines, including ROVs. The SEA roll simulator produces lateral 
accelerations using a deceleration sled and produces roll rates using a 
motor to rotate the test sled (see Figure 10).

[[Page 68979]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.009

SEA validated the roll simulator as an accurate simulation of ROV 
rollover and occupant kinematics by comparing roll rates, lateral 
accelerations, and ATD ejections that were created by the simulator 
with actual values measured during autonomous rollover. Results show 
that the roll simulator accurately re-creates the conditions of an ROV 
rollover. CPSC believes that the vehicle kinematics on the SEA rollover 
simulator accurately represent real-world events because SEA validated 
the sled kinematics against full-vehicle, real-world rollover events.
    SEA simulated tripped and untripped rollovers of seven sample ROVs 
using belted and unbelted ATD occupants. Plots of the head excursion 
data indicate how well the vehicle's occupant protection features 
retain the occupant inside the protective zone of the ROPS during a 
roll simulation (see Figure 11). Head displacement plots above the ROPS 
Plane indicate the occupant's head stayed inside the ROPS zone, and 
plots below the ROPS Plane indicate that the occupant's head moved 
outside the ROPS zone.
[GRAPHIC] [TIFF OMITTED] TP19NO14.010

    The SEA roll simulator test results indicate that five of the seven 
ROVs tested allowed a belted occupant's head to eject outside the ROPS 
of the vehicle during a quarter-turn rollover simulation. The occupant 
protection

[[Page 68980]]

performance of belted occupants varied from vehicle to vehicle, 
depending on seat belt design, passive hip and shoulder coverage, 
whether the rollover was tripped or untripped, and ROPS dimensions and 
geometry.
    CPSC staff analysis of the SEA roll simulator test results 
indicates that vehicles with the best occupant protection performance 
restricted movement of the occupant with combinations of quick-locking 
seat belts, passive coverage in the hip and shoulder areas of the 
occupant, and large ROPS zones around the occupant's head. Rollover 
tests indicate that a seat belt is effective at preventing full 
occupant ejection, but in some cases where the seat belt does not lock 
quickly, partial occupant ejection still occurs. However, when a seat 
belt is used in conjunction with a passive shoulder barrier restraint, 
testing indicates that the occupant remains within the protective zone 
of the vehicle's ROPS during quarter-turn rollover events.
    The SEA roll simulator test results also indicate that unbelted 
occupants are partially or fully ejected from all vehicles, regardless 
of the presence of other passive restraints, such as hip restraints or 
shoulder restraints. Although passive shoulder barriers may not provide 
substantial benefit for occupant protection in unbelted rollovers, the 
roll simulator test results indicate that shoulder restraints 
significantly improved occupant containment when used in conjunction 
with a seat belt.
    Although the SEA roll simulator is the most advanced test equipment 
viewed by the Commission, to date, and the test results provide clear 
evidence of occupant head excursion, not enough test data have been 
generated to base dynamic occupant protection performance test 
requirements on a device like the roll simulator. Therefore, the 
Commission is using the roll simulator test results to focus on 
occupant protection requirements that maximize occupant retention 
through seat belt use with passive shoulder restraint.
d. ANSI/ROHVA 1-2011 Occupant Protection Tests
    CPSC staff tested 10 sample ROVs to the occupant retention system 
(ORS) zone requirements specified in ANSI/ROHVA 1-2011. Requirements 
are specified for Zone 1--Leg/Foot, Zone 2--Shoulder/Hip, Zone 3--Arm/
Hand, and Zone 4--Head/Neck. CPSC focused on the requirements for Zone 
2 because occupant ejection occurs in this zone.\27\
---------------------------------------------------------------------------

    \27\ See Tab H of the briefing package.
---------------------------------------------------------------------------

    ANSI/ROHVA Zone 2--Shoulder/Hip requirements allow the vehicle to 
pass one of two different test methods to meet that zone's requirement. 
Under the first option, a construction-based method defines an area 
near the occupant's side that must be covered by a passive barrier. The 
test involves applying a 163-lbf. load at a point in the defined test 
area without failure or deformation of the barrier. Under the second 
option, a performance-based method specifies a tilt table test with a 
vehicle occupied by a belted test dummy. When the vehicle is tilted to 
45 degrees on the tilt table, the ejection of the dummy must not exceed 
5 inches beyond the vehicle width.
    Results of CPSC tests indicate that only four of 10 vehicles passed 
the construction-based test requirements, and eight of 10 vehicles 
passed the performance-based test requirements.\28\ CPSC analysis 
identified a primary weakness with the performance-based tilt table 
tests. The performance-based test criteria measure the torso excursion 
outside the vehicle width, not the excursion outside the protective 
zone of the ROPS. An occupant must remain inside the envelope of the 
ROPS to be protected; therefore, the requirement allows an inherently 
unsafe condition where the occupant moves outside the protective zone 
of the vehicle's ROPS.
---------------------------------------------------------------------------

    \28\ See Tab H of the briefing package.
---------------------------------------------------------------------------

    CPSC measured the difference between the outermost point of the ROV 
and the outermost point on the ROPS near the occupant's head (see 
Figure 12). On one vehicle, the vehicle's maximum width was 6.75 inches 
outside the maximum ROPS width near the occupant's head. Because the 
requirement is based on a 5-inch limitation beyond the vehicle width, 
the occupant's torso could be 11.75 inches (6.75 inches plus 5 inches) 
outside of the vehicle ROPS and still meet the performance-based 
requirement.
[GRAPHIC] [TIFF OMITTED] TP19NO14.011


[[Page 68981]]


    CPSC also compared the occupant head excursion relative to the 
torso excursion during the tilt table tests. Due to occupant rotation 
during the tests, the maximum head displacement exceeded the torso 
displacement by up to 3 inches. The discrepancy between head and torso 
displacement and between the vehicle width and ROPS' width can result 
in occupant head ejection that is 14.75 inches (11.75 inches plus 3 
inches) outside the protective zone of the ROPS and still meet the 
performance-based requirement.

VII. Relevant Existing Standards

A. Background

    Two different organizations developed separate voluntary standards 
for ROVs. The Recreational Off-Highway Vehicle Association (ROHVA) 
developed ANSI/ROHVA 1, American National Standard for Recreational 
Off-Highway Vehicles, and the Outdoor Power Equipment Institute (OPEI) 
developed ANSI/OPEI B71.9, American National Standard for Multipurpose 
Off-Highway Utility Vehicles.
    ROHVA member companies include: Arctic Cat, BRP, Honda, John Deere, 
Kawasaki, Polaris, and Yamaha. Work on ANSI/ROHVA 1 started in 2008, 
and work completed with the publication of ANSI/ROHVA 1-2010. The 
standard was immediately opened for revision, and a revised standard, 
ANSI/ROHVA 1-2011, was published in July 2011.
    OPEI member companies include: Honda, John Deere, Kawasaki, and 
Yamaha. Work on ANSI/OPEI B71.9 was started in 2008, and work was 
completed with the publication of ANSI/OPEI B71.9-2012 in March 2012.
    Both voluntary standards address design, configuration, and 
performance aspects of ROVs, including requirements for accelerator and 
brake controls; service and parking brake/parking mechanism 
performance; lateral and pitch stability; lighting; tires; handholds; 
occupant protection; labels; and owner's manuals.
    CPSC staff participated in the canvass process used to develop 
consensus for ANSI/ROHVA 1 and ANSI/OPEI B71.9. From June 2009 to the 
present, CPSC staff has engaged actively with ROHVA and OPEI through 
actions that include the following:
     Sending correspondence to ROHVA and OPEI with comments on 
voluntary standard ballots that outlined CPSC staff's concerns that the 
voluntary standard requirements for lateral stability are too low, that 
requirements for vehicle handling are lacking, and that requirements 
for occupant protection are not robust;
     Participating in public meetings with ROHVA and OPEI to 
discuss development of the voluntary standard and to discuss static and 
dynamic tests performed by contractors on behalf of CPSC staff;
     Sharing all CPSC contractor reports with test results of 
static and dynamic tests performed on ROVs by making all reports 
available on the CPSC Web site;
     Requesting copies of test reports on dynamic tests 
performed on ROVs by ROHVA for CPSC staff to review;
     Demonstrating dynamic test procedures and data collection 
to ROHVA and OPEI at a public meeting at an outdoor test facility in 
East Liberty, OH; and
     Submitting suggested changes and additions to the ANSI/
ROHVA 1-2011 voluntary standard to improve lateral stability, vehicle 
handling, and occupant protection (OPEI was copied).
    ANSI/ROHVA 1-2011 was published in July 2011, without addressing 
CPSC staff's concerns. CPSC staff requested, but has not received 
reports or test results of static or dynamic tests conducted by 
contractors on behalf of ROHVA.
    ANSI/OPEI B71.9-2012 was published in March 2012, without 
addressing CPSC staff's concerns.
    On August 29, 2013, CPSC staff sent a letter to ROHVA with 
suggested modifications to the voluntary standard requirements to 
address staff's concerns. CPSC staff sent a courtesy copy of the August 
29, 2013 recommendation letter to OPEI. On November 27, 2013, ROHVA 
responded that ROHVA plans to adopt less stringent versions of CPSC 
staff's suggested requirements to improve the lateral stability and 
occupant protection performance of ROVs. On March 13, 2014, ROHVA sent 
CPSC staff the Canvass Draft of proposed revisions to ANSI/ROHVA 1-
2011. Staff responded to the Canvass Draft on May 23, 2014, and 
summarized why staff believes ROHVA's proposed requirements will not 
reduce the number of deaths and injuries from ROVs. The discussion 
below also provides that explanation. On September 24, 2014, ANSI 
approved the proposed revisions to ANSI/ROHVA 1-2011, which is 
identical to the Canvass Draft. ROHVA has advised that the revised 
standard will soon be published as ANSI/ROHVA 1-2014. In addition, CPSC 
staff met with representatives from ROHVA and OPEI on October 23, 2014. 
Following is a link to the video of this meeting: https://www.cpsc.gov/en/Newsroom/Multimedia/?vid=70952.
    On February 21, 2014, OPEI sent a letter to CPSC staff requesting 
that the CPSC exclude from CPSC's rulemaking efforts multipurpose off-
highway utility vehicles (MOHUVs) that meet the ANSI/OPEI B71.9-12 
standard requirements. We address this request in the response to 
comments section of this preamble (Section VIII).

B. Voluntary Standards Provisions Related to the Proposed Rule

    In this section, we summarize the provisions of the voluntary 
standards that are related to the specific requirements the Commission 
is proposing and we assess the adequacy of these voluntary standard 
provisions.
1. Lateral Stability
    ANSI/ROHVA 1-2011 and ANSI/OPEI B71.9 include similar provisions to 
address static lateral stability and differing provisions to address 
dynamic lateral stability:
    Voluntary Standard Requirement: ANSI/ROHVA 1-2011 Section 8.2 
Stability Coefficient (Kst) and ANSI/OPEI B71.9-2012 Section 
8.6 Stability Coefficient (Kst) specify a stability 
coefficient, Kst, which is calculated from the vehicle's 
center of gravity location and track-width dimensions. The value of 
Kst for a vehicle at curb weight (without occupants) is 
required to be no less than 1.0.
    Adequacy: The Commission believes the stability coefficient 
requirement does not adequately address lateral stability in ROVs 
because static tests are unable to account fully for the dynamic tire 
deflections and suspension compliance exhibited by ROVs in a dynamic 
maneuver. For practical purposes, Kst and SSF values provide 
the same information for ROVs because the difference in front and rear 
track widths are averaged in the SSF calculation. Table 4 shows the 
results of SSF measurements made by SEA for driver-plus-passenger load 
condition. A comparison of how the vehicles would rank if the SSF (or 
Kst) were used instead of the threshold lateral acceleration 
at rollover (Ay) illustrates how poorly a stability 
coefficient correlates to the actual rollover resistance of the 
vehicle. The stability coefficient does not account for dynamic effects 
of tire compliance, suspension compliance, or vehicle handling, which 
are important factors in the vehicle's lateral stability.

[[Page 68982]]



            Table 4--Vehicle Ascending Rank Order Ay vs. SSF
                     [Operator plus passenger load]
------------------------------------------------------------------------
                                               Vehicle  rank
      Vehicle  rank  (Ay)         Ay  (g)          (SSF)           SSF
------------------------------------------------------------------------
D..............................     0.625   F.................     0.881
B..............................     0.655   A.................     0.887
A..............................     0.670   H.................     0.918
J..............................     0.670   B.................     0.932
I..............................     0.675   D.................     0.942
F..............................     0.690   J.................     0.962
E..............................     0.700   E.................     0.965
H..............................     0.705   C.................     0.991
C..............................     0.740   G.................     1.031
G..............................     0.785   I.................     1.045
------------------------------------------------------------------------
Adapted from: Heydinger, G. (2011) Vehicle Characteristics Measurements
  of Recreational Off-Highway Vehicles--Additional Results for Vehicle
  J. Retrieved from https://www.cpsc.gov/PageFiles/93928/rovj.pdf.

    Furthermore, all of the ROVs tested pass the Kst minimum 
of 1.0 for an unoccupied vehicle, as specified by ANSI/ROHVA 1-2011 and 
ANSI/OPEI B71.9-12. The Kst value of an ROV with no 
occupants is of limited value because an ROV in use has at least one 
occupant. The Commission believes the ANSI/ROHVA and ANSI/OPEI 
stability coefficient requirement is a requirement that all ROVs can 
pass, does not reflect the actual use of ROVs, does not promote 
improvement in lateral stability, and does not correspond to the actual 
rollover resistance of ROVs. The Commission believes that the threshold 
lateral acceleration at rollover is a direct measure for rollover 
resistance, and its use would eliminate the need for a stability 
coefficient requirement.
    Voluntary Standard Requirement: ANSI/ROHVA 1-2011 Section 8.1 Tilt 
Table Test and ANSI/OPEI Section 8.7 Tilt Table Stability specify tilt 
table tests in the driver-plus-passenger load condition and the gross 
vehicle weight rating (GVWR) load condition. The minimum tilt table 
angle (TTA) requirement for an ROV with a driver-plus-passenger load 
condition is 30 degrees, and the minimum TTA for GVWR load condition is 
24 degrees.
    Adequacy: The CPSC believes the tilt table requirement does not 
adequately address lateral stability in ROVs because static tests are 
unable to account fully for the dynamic tire deflections and suspension 
compliance exhibited by ROVs in a dynamic maneuver. Table 5 shows the 
results of tilt table measurements made by SEA for driver-plus-
passenger load condition. A comparison of how the vehicles would rank 
if the TTA were used instead of the direct measurement of threshold 
lateral acceleration at rollover (Ay) illustrates how poorly 
the TTA corresponds to the actual rollover resistance of the vehicle. 
The tilt table test does not account for dynamic effects of tire 
compliance, suspension compliance, or vehicle handling, which are 
important factors in the vehicle's lateral stability.
    Furthermore, all of the ROVs tested passed the minimum 30 degree 
TTA requirement specified by ANSI/ROHVA 1-2011. The ROV with the lowest 
rollover resistance, as directly measured by threshold lateral 
acceleration at rollover (Vehicle D, Ay = 0.625 g, TTA = 
33.7 degrees), exceeds the voluntary standard TTA requirement by 3.7 
degrees, or 12 percent above the 30 degree minimum. The ROV that was 
part of a repair program to increase its roll resistance, Vehicle A, 
exceeds the TTA requirement by 3.0 degrees, or 10 percent above the 30 
degree minimum.

            Table 5--Vehicle Ascending Rank Order Ay vs. TTA
                     [Operator plus passenger load]
------------------------------------------------------------------------
                                               Vehicle  rank       TTA
      Vehicle  rank  (Ay)         Ay  (g)          (TTA)         (deg.)
------------------------------------------------------------------------
D..............................     0.625   A.................      33.0
B..............................     0.655   B.................      33.6
A..............................     0.670   D.................      33.7
J..............................     0.670   I.................      35.4
I..............................     0.675   H.................      35.9
F..............................     0.690   J.................      36.1
E..............................     0.700   F.................      36.4
H..............................     0.705   E.................      38.1
C..............................     0.740   C.................      38.8
G..............................     0.785   G.................      39.0
------------------------------------------------------------------------
Source: Heydinger, G. (2011) Vehicle Characteristics Measurements of
  Recreational Off-Highway Vehicles--Additional Results for Vehicle J.
  Retrieved from https://www.cpsc.gov/PageFiles/93928/rovj.pdf.

    The CPSC believes the ANSI/ROHVA and ANSI/OPEI tilt table 
requirement does not detect inadequate rollover resistance. The TTA 
requirement in the voluntary standard does not correlate to the actual 
rollover resistance of ROVs, allows a vehicle that was part of repair 
program to pass the test without having undergone the repair, and 
provides no incentive for manufacturers to improve the lateral 
stability of ROVs. The CPSC believes the threshold lateral acceleration 
at rollover is a direct measure of rollover resistance, and its use 
would eliminate the need for a tilt table test requirement.
    Voluntary Standard Requirement: ANSI/ROHVA 1-2011 Section 8.3 
Dynamic Stability specifies a dynamic stability test based on a 
constant steer angle test performed on pavement. The standard describes 
the method for driving the vehicle around a 25-foot radius circle and 
slowly increasing the speed until 0.6 g of lateral acceleration is 
achieved; or 0.6 g lateral acceleration cannot be achieved because the 
vehicle experiences two-wheel lift of the inside wheels, or the vehicle 
speed is limited and will not increase with further throttle input. The 
vehicle passes the dynamic test if at least eight out of 10 test runs 
do not result in two-wheel lift.
    Adequacy: The CPSC does not believe the ANSI/ROHVA requirement 
accurately characterizes the lateral stability of an ROV because it 
does not measure the threshold lateral acceleration at rollover. The 
Commission is not aware of any standards, recognized test protocols, or 
real-world significance that supports using a constant steer angle test 
to assess dynamic lateral stability.
    CPSC staff contracted SEA to conduct constant steer angle testing, 
as specified by the ROHVA standard, on vehicles A, F, and J of the ROV 
study.\29\ Table 6 shows the results of the tests.
---------------------------------------------------------------------------

    \29\ Heydinger, G. J. (2011) Results from Proposed ROHVA and 
OPEI Dynamic Maneuvers--Vehicles A, F, and J. Retrieved from: https://www.cpsc.gov/Global/Research-and-Statistics/Technical-Reports/Sports-and-Recreation/ATV-ROV/ProposedROHVAandOPEIDynamicManeuvers.pdf.)

                      Table 6--Summary of Constant Steer Angle Test for 25 ft. Radius Path
----------------------------------------------------------------------------------------------------------------
                                    Turn direction  (CW =
              Vehicle                  clockwise  CCW =     Test end condition/   ROHVA Test  pass/fail outcome
                                      counter-clockwise)      limit response
----------------------------------------------------------------------------------------------------------------
Vehicle A.........................  Right (CW)...........  Two-wheel lift......  Fail.
                                    Left (CCW)...........  Two-wheel lift......  Fail.
Vehicle F.........................  Right (CW)...........  Maximum Speed*......  Pass.**
                                    Left (CCW)...........  Maximum Speed*......  Pass.**

[[Page 68983]]

 
Vehicle J.........................  Right (CW)...........  Two-wheel lift......  Fail.
                                    Left (CCW)...........  Maximum Speed/        Pass.
                                                            Spinout.
----------------------------------------------------------------------------------------------------------------
* Maximum speed occurred very near 0.6 g of corrected lateral acceleration for Vehicle F.
** Two-wheel lift occurred for Vehicle F after the driver slowed from maximum speed at the end of the test.
Source: Heydinger, G. (2011) Results from Proposed ROHVA and OPEI Dynamic Maneuvers--Vehicles A, F, and J.
  Retrieved from https://www.cpsc.gov/Global/Research-and-Statistics/Technical-Reports/Sports-and-Recreation/ATV-ROV/ProposedROHVAandOPEIDynamicManeuvers.pdf.

    The Commission is concerned that ROVs with low lateral stability 
can pass ROHVA's dynamic stability requirement because the small turn 
radius limits the ROV's speed and prevents generation of the lateral 
accelerations necessary to assess rollover resistance (as shown by the 
results for Vehicle F). The Commission is also concerned that the 
effects of oversteer can allow an ROV to pass the test because maximum 
speed is reached by vehicle spinout (as shown by the results for 
Vehicle J).
    NHTSA evaluated the J-turn test protocol as a method to measure the 
rollover resistance of automobiles.\30\ NHTSA determined that the J-
turn test is the most objective and repeatable method for vehicles with 
low rollover resistance. Vehicles with low rollover resistance exhibit 
untripped rollover on pavement during a J-turn test and the lateral 
acceleration at the rollover threshold can be measured. Lateral 
acceleration is the accepted measure by vehicle engineers for assessing 
lateral stability or rollover resistance.\31\ This value is commonly 
used by engineers to compare rollover resistance from one vehicle to 
another. The ANSI/ROHVA test protocol does not measure the lateral 
acceleration at two-wheel lift, and the parameters of the test appear 
tuned to allow most vehicles to pass. Based on CPSC's testing and 
review, the Commission does not believe the ANSI/ROHVA dynamic 
stability requirement is a true measure of rollover resistance, and the 
CPSC does not believe the requirement will improve the lateral 
stability of ROVs.
---------------------------------------------------------------------------

    \30\ Forkenbrock, G. and Garrott, W. (2002). A Comprehensive 
Experimental Evaluation of Test Maneuvers That May Induce On-Road, 
Untripped, Light Vehicle Rollover Phase IV of NHTSA's Light Vehicle 
Rollover Research Program. DOT HS 809 513.
    \31\ Gillespie, T. (1992). Fundamentals of Vehicle Dynamics. 
Society of Automotive Engineers, Inc. p. 309-319.
---------------------------------------------------------------------------

    Voluntary Standard Requirement: ANSI/OPEI B71.9-2012 Section 8.8 
Dynamic Stability specifies a dynamic stability test based on a 20 mph 
J-turn maneuver performed on pavement. At a steering input of 180 
degrees in the right and left directions, the vehicle shall not exhibit 
two-wheel lift.
    Adequacy: The Commission does not believe the ANSI/OPEI requirement 
accurately characterizes the lateral stability of an ROV because the 
ANSI/OPEI requirement does not measure the threshold lateral 
acceleration at rollover. The Commission is not aware of any standards 
or recognized test protocols that support using a J-turn maneuver with 
180 degrees of steering wheel input to assess dynamic lateral stability 
of an ROV.
    OPEI's use of the J-turn maneuver does not measure the lateral 
acceleration at two-wheel lift that produces ROV rollover. There is no 
correspondence between the proposed ANSI/OPEI dynamic stability 
requirement and ROV lateral stability because the 180-degree steering 
wheel input does not correspond to a turning radius. For example, an 
ROV with a low steering ratio will make a sharper turn at 180 degrees 
of steering wheel input than an ROV with a high steering ratio. (The 
steering ratio relates the amount that the steering wheel is turned to 
the amount that the wheels of the vehicle turns. A higher steering 
ratio means the driver turns the steering wheel more to get the vehicle 
wheels to turn, and a lower steering ratio means the driver turns the 
steering wheel less to get the vehicle wheels to turn.) In the proposed 
ANSI/ROHVA J-turn test, a vehicle with a larger steering ratio will 
make a wider turn and generate less lateral acceleration than a vehicle 
with a smaller steering ratio.
    The steering ratio is set by the ROV manufacturer and varies 
depending on make and model. SEA measured the steering ratios of the 10 
sample ROVs that were tested (see Figure 13). If the dynamic lateral 
stability requirement is defined by a steering wheel angle input, a 
manufacturer could increase the steering ratio of a vehicle to meet the 
requirement rather than improve the vehicle's stability.

[[Page 68984]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.012

    CPSC staff contracted SEA to conduct J-turn testing, as specified 
by the ANSI/OPEI standard, on vehicles A, F, and J (see Table 7).

                                     Table 7--Summary of J-Turn Test Results
                              [20 mph with 180 degrees steering wheel angle input]
----------------------------------------------------------------------------------------------------------------
                                                           Speed required for 2-   OPEI 20 mph test  pass/fail
              Vehicle                   Turn direction             wheel                     outcome
----------------------------------------------------------------------------------------------------------------
Vehicle A.........................  Right................  22 mph..............  Pass.
                                    Left.................  21 mph..............  Pass.
Vehicle F.........................  Right................  21 mph..............  Pass.
                                    Left.................  22 mph..............  Pass.
Vehicle J.........................  Right................  21 mph..............  Pass.
                                    Left.................  23 mph..............  Pass.
----------------------------------------------------------------------------------------------------------------
Source: Heydinger, G. (2011) Results from Proposed ROHVA and OPEI Dynamic Maneuvers--Vehicles A, F, and J.
  Retrieved from https://www.cpsc.gov/Global/Research-and-Statistics/Technical-Reports/Sports-and-Recreation/ATV-ROV/ProposedROHVAandOPEIDynamicManeuvers.pdf.

    CPSC is concerned that ROVs with low lateral stability can pass 
OPEI's dynamic stability requirement because an ROV that was part of a 
repair program (Vehicle A) to increase its roll resistance passed the 
ANSI/OPEI stability test. When the ANSI/OPEI J-turn maneuver was 
conducted just one mile above the requirement at 21 mph, Vehicle A 
failed. Similarly, when the maneuver was conducted at 22 mph, Vehicle F 
and Vehicle J failed. These results indicate that the parameters of the 
test protocol allow most ROVs to pass.
    NHTSA evaluated the J-turn test protocol as a method to measure 
rollover resistance of automobiles and determined that the J-turn test 
is the most objective and repeatable method for vehicles with low 
rollover resistance.\32\ Vehicles with low rollover resistance exhibit 
untripped rollover on pavement during a J-turn test and the lateral 
acceleration at the rollover threshold can be measured. Lateral 
acceleration is the accepted measure by vehicle engineers for assessing 
lateral stability or rollover resistance.\33\ This value is commonly 
used by engineers to compare rollover resistance from one vehicle to 
another. The ANSI/OPEI test protocol does not measure the lateral 
acceleration at two-wheel lift, and the parameters of the test appear 
tuned to allow most vehicles to pass. Based on CPSC's testing and 
review, the CPSC does not believe the ANSI/OPEI dynamic stability 
requirement is a true measure of rollover resistance, and the CPSC does 
not believe the requirement will improve the lateral stability of ROVs.
---------------------------------------------------------------------------

    \32\ Forkenbrock, G. and Garrott, W. (2002). A Comprehensive 
Experimental Evaluation of Test Maneuvers That May Induce On-Road, 
Untripped, Light Vehicle Rollover Phase IV of NHTSA's Light Vehicle 
Rollover Research Program. DOT HS 809 513.
    \33\ Gillespie, T. (1992). Fundamentals of Vehicle Dynamics. 
Society of Automotive Engineers, Inc. p. 309-319.
---------------------------------------------------------------------------

2. Vehicle Handling
    ANSI/ROHVA 1-2011 and ANSI/OPEI B71.9 both lack provisions to 
address vehicle handling:
    Voluntary Standard Requirement: ANSI/ROHVA 1-2011 ANSI/OPEI B71.9-
2012 do not specify a vehicle handling requirement.
    Adequacy: CPSC's testing and review indicate that a requirement for 
sub-limit understeer is necessary to reduce ROV rollovers that may be 
produced by sub-limit oversteer in ROVs. Tests conducted by SEA show 
that ROVs in sub-limit oversteer transition to a condition where the 
lateral acceleration increases suddenly and exponentially.

[[Page 68985]]

The CPSC believes this condition can lead to untripped ROV rollovers or 
cause ROVs to slide into limit oversteer and experience tripped 
rollover.
    ROVs that understeer in sub-limit conditions do not exhibit a 
sudden increase in lateral acceleration. Therefore, the CPSC concludes 
that ROVs should be required to operate in understeer at sub-limit 
conditions based on the associated inherent dynamic stability of 
understeering ROVs and the smaller burden of steering correction it 
places on the average driver who is familiar with driving a passenger 
vehicle that operates in sub-limit understeer.
    SIS tests conducted by SEA that illustrate the sudden increase in 
lateral acceleration that is found only in vehicles that exhibit sub-
limit oversteer. The sudden increase in lateral acceleration is 
exponential and represents a dynamically unstable condition. This 
condition is undesirable because it can cause a vehicle with low 
lateral stability (such as an ROV) to roll over suddenly.
    In Figure 14, Vehicle A is an ROV that transitions to oversteer; 
Vehicle H is the same model ROV, but a later model year in which the 
oversteer has been corrected to understeer.
[GRAPHIC] [TIFF OMITTED] TP19NO14.013

    When Vehicle A reached its dynamically unstable condition, the 
lateral acceleration suddenly increased in less than 1 second, and the 
vehicle rolled over. In contrast, Vehicle H never reaches a dynamically 
unstable condition because the condition does not develop in 
understeering vehicles. The increase in Vehicle H's lateral 
acceleration remains linear, and Vehicle H rolls over more than 5 
seconds later than Vehicle A.
3. Occupant Protection
    ANSI/ROHVA 1-2011and ANSI/OPEI B71.9 include similar provisions to 
address occupant retention during a rollover event.
    Voluntary Standard Requirement: ANSI/ROHVA 1-2011 Section 11.2 Seat 
Belt Reminder and ANSI/OPEI B71.9-2012 Section 5.1.3.2 Seat Belt 
Reminder System specify that ROVs shall be equipped with a seat belt 
reminder system that activates a continuous or flashing warning light 
visible to the operator for at least 8 seconds after the vehicle is 
started.
    Adequacy: The CPSC believes the requirement for an 8-second 
reminder light is not adequate to increase meaningfully seat belt use 
rates in ROVs because the system is not intrusive enough to motivate 
drivers and passengers to wear their seat belts. Results from past 
studies on automotive seat belt reminders conclude that visual 
reminders are ineffective. Numerous studies also conclude that reminder 
systems must be intrusive enough to motivate users to buckle their seat 
belts. The more intrusive reminders are more effective at changing user 
behavior, as long as the reminder is not so intrusive that users bypass 
the system.
    The Commission's analysis of ROV-related incidents indicates that 
91 percent of fatal victims, and 73 percent of all victims (fatal and 
nonfatal), were not wearing a seat belt at the time of the incident. 
Without seat belt use, occupants experience partial to full ejection 
from the ROV, and many occupants are struck by the ROV after ejection. 
Based on review of ROV incident data and CPSC's testing described 
above, the Commission believes that many ROV deaths and injuries can be 
eliminated if occupants are wearing seat belts.
    Automotive researchers have developed technology that motivates 
drivers to buckle seat belts by making it more difficult to drive 
faster than 20-25 mph if the driver's seat belt is not buckled.\34\ 
This concept shows promise in increasing seat belt use because the 
technology was acceptable to users and was 100 percent effective in 
motivating drivers to buckle their seat belts. One ROV manufacturer has 
also introduced a technology that limits the vehicle speed if the 
driver's seat belt is not buckled. ROVs with the speed-limitation 
technology have been in the market since 2010.
---------------------------------------------------------------------------

    \34\ Van Houten, R., Hilton, B., Schulman, R., and Reagan, I. 
(2011). Using Haptic Feedback to Increase Seat Belt Use of Service 
Vehicle Drivers. (DOT HS 811 434). Washington, DC: National Highway 
Traffic Safety Administration, U.S. Department of Transportation. 
Hilton, Bryan W. (2012). The Effect of Innovative Technology on 
Seatbelt Use. Masters Theses. Paper 103.

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

[[Page 68986]]

    Given the low seat belt use rate in ROV-related incidents, as well 
as the substantial potential reduction in injuries and deaths if seat 
belt use were higher, the CPSC believes that the requirement for seat 
belt reminders should be more stringent and should incorporate the most 
recent advances in technology developed in the automotive and ROV 
market.
    Voluntary Standard Requirement: ANSI/ROHVA 1-2011 Section 11.3 ORS 
Zones specifies construction and performance requirements for four 
zones that cover the leg/foot, shoulder/hip, arm/hand, and head/neck 
areas of an occupant. (Occupant retention system (ORS) is defined in 
ANSI/ROHVA 1-2011 as a system, including three-point seat belts, for 
retaining the occupant(s) of a vehicle to reduce the probability of 
injury in the event of an accident.) The construction requirements 
specify a force application test to set minimum guidelines for the 
design of doors, nets, and other barriers that are intended to keep 
occupants within the protection zone of the ROPS. The performance 
requirements use a tilt table and a Hybrid III 50th percentile male 
anthropomorphic test device (ATD) to determine occupant excursion when 
the vehicle is tilted 45 degrees laterally.
    Adequacy: The CPSC believes the tilt table performance requirements 
for Zone 2--Shoulder/Hip are not adequate to ensure that occupants 
remain within the protective zone of the vehicle's ROPS during a 
rollover event. The tilt table test method measures the torso ejection 
outside the vehicle width, not the ejection outside the protective zone 
of the ROPS. The CPSC's test results indicate the tilt table test 
allows unacceptable occupant head excursion beyond the protective zone 
of the vehicle ROPS. The Commission also believes the tilt table test 
method is not an accurate simulation of an ROV rollover event because 
the test method does not reproduce the lateral acceleration and roll 
experienced by the vehicle, and by extension, the occupants, during a 
rollover.
    CPSC staff also believes the construction-based test method for 
Zone 2 is inadequate because the specified point of application (a 
single point) and 3-inch diameter test probe do not accurately 
represent contact between an occupant and the vehicle during a rollover 
event. Specifying a single point does not ensure adequate coverage 
because a vehicle with a passive barrier at only that point would pass 
the test. Similarly, a 3 inch diameter probe does not represent the 
upper arm of an occupant and therefore does not ensure adequate 
coverage.
    Voluntary Standard Requirement: ANSI/OPEI B71.9-2012 Section 5.1.4 
Occupant Side Retention Devices specifies ROVs shall be equipped with 
occupant side retention devices that reduce the probability of 
entrapment of a properly belted occupant's head, upper torso, and limbs 
between the vehicle and the terrain, in the event of a lateral 
rollover. Physical barriers or design features of the vehicle may be 
used to comply with the requirement, but no performance tests are 
specified to determine compliance with the requirement.
    Adequacy: The Commission believes the occupant side retention 
requirements are not adequate because they lack performance 
requirements to gauge occupant protection performance. Performance 
requirements, based on occupant protection performance tests of ROV 
rollovers, are needed to ensure that occupants remain within the 
protective zone of the vehicle's ROPS during a rollover event.

VIII. Response to Comments

    In this section, we describe and respond to comments to the ANPR 
for ROVs. We present a summary of each of the commenter's topics, 
followed by the Commission's response. The Commission received 116 
comments. The comments can be viewed on: www.regulations.gov, by 
searching under the docket number of the ANPR, CPSC-2009-0087. Letters 
with multiple and detailed comments were submitted by the following:
    [ssquf] Joint comments submitted on behalf of Arctic Cat Inc., 
Bombardier Recreational Products Inc., Polaris Industries Inc., and 
Yamaha Motor Corporation, U.S.A. (Companies);
    [ssquf] Carr Engineering, Inc. (CEI);
    [ssquf] The OPEI/ANSI B 71.9 Committee (Committee); and
    [ssquf] ROHVA.
    The respondents were ROV manufacturers and their associations, 
consultants to ROV manufacturers, and more than 110 consumers. Eighteen 
commenters supported developing regulatory standards for ROVs. The 
other commenters opposed rulemaking action. The commenters raised 
issues in five areas:
     Voluntary standard activities,
     Static stability metrics,
     Vehicle handling,
     Occupant protection, and
     Consumer behavior.
    The comment topics are separated by category.
Voluntary Standard Activities
    1. Comment: Comments from the Companies, ROHVA, and several 
individuals state that the CPSC should work with ROHVA to develop a 
consensus voluntary standard for ROVs.
    Response: As described in detail in the previous section of this 
preamble, CPSC staff has been engaged actively with ROHVA since 2009, 
to express staff's concerns about the voluntary standard and to provide 
specific recommendations for the voluntary standard and supply ROHVA 
with CPSC's test results and data supporting the staff's 
recommendations.
    CPSC believes the history of engagement with ROHVA, as detailed 
above, shows that CPSC staff has tried to work with ROHVA to improve 
the voluntary standard requirements to address low lateral stability, 
lack of vehicle handling requirements, and inadequate occupant 
protection requirements. The Commission does not believe deferring to 
ROHVA will address those areas of concern because, although ROHVA has 
made changes to the voluntary standard, the requirements still do not 
improve the lateral stability of ROVs, do not eliminate sub-limit 
oversteer handling, and do not improve occupant protection in a 
rollover event.
    2. Comment: Comments from the Committee and ROHVA state that the 
Commission should defer to the current voluntary standards for ROVs. 
Several comments state that the current voluntary standards are 
adequate.
    Response: In the previous section of this preamble, we explain in 
detail why the requirements in ANSI/ROHVA 1-2011 and ANSI/OPEI B71.9-
2012 do not adequately address the risk of injury and death associated 
with ROVs. We summarize that explanation below.
    Lateral Stability. The Commission believes the static stability 
requirements and the dynamic lateral stability requirements specified 
in both voluntary standards do not measure the vehicle's resistance to 
rollover. Static and dynamic tests conducted by SEA on a sample of ROVs 
available in the U.S. market indicate that the tests specified in ANSI/
ROHVA 1-2011 and the ANSI/OPEI B71.9 will not promote improvement in 
the rollover resistance of ROVs.
    Vehicle Handling. In addition, ANSI/ROHVA 1-2011 and ANSI/OPEI 
B71.9-2012 do not have requirements for vehicle handling. The 
Commission believes that a requirement for sub-limit understeer is 
necessary to reduce ROV rollovers that may be produced by sub-limit 
oversteer in ROVs. Tests

[[Page 68987]]

conducted by SEA show that ROVs in sub-limit oversteer transition to a 
condition where the lateral acceleration increases suddenly and 
exponentially. The Commission believes this runaway increase in lateral 
acceleration can lead to untripped ROV rollovers or cause ROVs to slide 
into limit oversteer and experience tripped rollover.
    Occupant Protection. ANSI/ROHVA 1-2011 and ANSI/OPEI B71.9--2012 
require only an 8-second reminder light to motivate users to buckle 
seat belts. This requirement is similar to the Federal Motor Vehicle 
Safety Standard (FMVSS) seat belt reminder requirements for 
automobiles. Manufacturers in the automotive industry have long since 
exceeded such minimal seat belt reminder requirements because numerous 
studies have proven that the FMVSS requirements, and indeed visual-only 
reminders, are not effective.\35\
---------------------------------------------------------------------------

    \35\ Westefeld, A. and Phillips, B.M. (1976). Effectiveness of 
Various Safety Belt Warning Systems. (DOT HS 801 953). Washington, 
DC: National Highway Traffic Safety Administration, U.S. Department 
of Transportation.
---------------------------------------------------------------------------

    Lastly, the occupant protection requirements in ANSI/ROHVA 1-2011 
and ANSI/OPEI B71.9-2012 are not based on valid occupant protection 
performance tests that simulate conditions of vehicle rollover. ANSI/
OPEI B71.9-2012 does not include any performance requirements for 
occupant protection. ANSI/ROHVA 1-2011 includes performance 
requirements based on static tilt tests that allow unacceptable 
occupant head ejection beyond the protective zone of the vehicle ROPS.
    3. Comment: On February 21, 2014, OPEI sent a letter to CPSC staff 
requesting that the CPSC exclude multipurpose off-highway utility 
vehicles (MOHUVs) from CPSC's rulemaking efforts. OPEI states that 
there are key differences between work-utility vehicles and 
recreational vehicles. The differences include: Maximum vehicle speed, 
engine and powertrain design, cargo box configuration and capacity, 
towing provisions, and vehicle usage.
    Response: The Commission's proposed requirements for lateral 
stability, vehicle handling, and occupant protection are intended to 
reduce deaths and injuries caused by ROV rollover and occupant 
ejection. ROVs are motorized vehicles that are designed for off-highway 
use and have four or more tires, steering wheel, non-straddle seating, 
accelerator and brake pedals, ROPS, restraint system, and maximum 
vehicle speed greater than 30 mph.
    ``MOHUVs,'' as defined by ANSI/OPEI B71.9-2012, are vehicles with 
four or more wheels, a steering wheel, non-straddle seating, and 
maximum speed between 25 and 50 mph. Therefore, the Commission believes 
that an MOHUV that exceeds 30 mph is an ROV that is subject to the 
scope of the proposed rulemaking. The differences cited by OPEI between 
work-utility vehicles and recreational vehicles, e.g., the cargo 
capacity or the powertrain of a vehicle, do not exclude these ROVs from 
the hazard of rollover and occupant ejection.

Static Stability Metrics

    1. Comment: Comments from CEI state that the Static Stability 
Factor (SSF), defined as T/2H, is not an appropriate metric for 
stability because there is no correlation between SSF values and ROV 
rollovers.
    Response: The Commission agrees that the SSF is not an appropriate 
metric for ROV lateral stability because CPSC staff compared the actual 
lateral acceleration at rollover threshold of several ROVs, as measured 
by the J-turn test, and found that static measures (whether 
Kst, SSF, or TTA) are not accurate predictors of the 
vehicle's rollover resistance. The static tests are unable to account 
fully for the dynamic tire deflections and suspension compliance 
exhibited by ROVs. The Commission believes that the threshold lateral 
acceleration at rollover (Ay) is the most appropriate metric to use 
because it is a direct measure of the vehicle's resistance to rollover.
    2. Comment: Comments from the Companies and the Committee state 
that NHTSA decided not to implement a minimum SSF standard for on-road 
vehicles because it would have forced the radical redesign of the 
characteristics of many, and in some cases, all vehicles of certain 
classes, which would have raised issues of public acceptance and 
possibly even the elimination of certain classes of vehicles.
    Response: Contrary to the comment's implication that setting a 
minimum lateral stability (in this case SSF) is detrimental to vehicle 
design, and that NHTSA abandoned the use of SSF, NHTSA concluded that 
there is a causal relationship between SSF and rollover, and NHTSA has 
incorporated the SSF in its New Car Assessment Program (NCAP) rating of 
vehicles. In June 1994, NHTSA terminated rulemaking to establish a 
minimum standard for rollover resistance because it would be difficult 
to develop a minimum stability standard that would not disqualify whole 
classes of passenger vehicles (light trucks and sport utility vehicles) 
that consumers demand. Instead, by January 2001, NHTSA concluded that 
consumer information on the rollover risk of passenger cars would 
influence consumers to purchase vehicles with a lower rollover risk and 
inspire manufacturers to produce vehicles with a lower rollover 
risk.\36\ NHTSA found consistently that given a single-vehicle crash, 
the SSF is a good statistical predictor of the likelihood that the 
vehicle will roll over.\37\ The number of single-vehicle crashes was 
used as an index of exposure to rollover because this method eliminates 
the additional complexity of multi-vehicle impacts and because about 82 
percent of light vehicle rollovers occur in single-vehicle crashes. 
NHTSA decided to use the SSF to indicate the risk of rollover in 
single-vehicle crashes and to incorporate the new rating into NHTSA's 
New Car Assessment Program (NCAP). Based on NHTSA's statistical 
analysis of single-vehicle crash data and vehicle SSF value, the NCAP 
provides a 5-star rating system. One star represents a 40 percent or 
higher risk of rollover in a single vehicle crash; two stars represent 
a risk of rollover between 30 percent and 40 percent; three stars 
represent a risk of rollover between 20 percent and 29 percent; four 
stars represent a risk of rollover between 10 percent and 19 percent; 
and five stars represent a risk of rollover of less than 10 percent.
---------------------------------------------------------------------------

    \36\ Walz, M. C. (2005). Trends in the Static Stability Factor 
of Passenger Cars, Light Trucks, and Vans. DOT HS 809 868. Retrieved 
from https://www.nhtsa.gov/cars/rules/regrev/evaluate/809868/pages/.
    \37\ Rollover Prevention Docket No. NHTSA-2000-6859 RIN 2127-
AC64. Retrieved from https://www.nhtsa.gov/cars/rules/rulings/rollover/Chapt05.html.
---------------------------------------------------------------------------

    A subsequent study of SSF trends in automobiles found that SSF 
values increased for all vehicles after 2001, particularly SUVs, and 
SUVs tended to have the worst SSF values in the earlier years. NHTSA's 
intention that manufacturers improve the lateral stability of passenger 
vehicles was achieved through the NCAP rating, a rating based 
predominantly on the SSF value of the vehicle.
    Based on dynamic stability tests conducted by SEA and improvements 
in the Yamaha Rhino after the repair program was initiated, the 
Commission believes that setting a minimum rollover resistance value 
for ROVs can improve the lateral stability of the current market of 
ROVs, without forcing radical designs or elimination of any models. The 
Commission also believes continued increase in ROV lateral stability 
can be achieved by making the value of each model vehicle's threshold 
lateral

[[Page 68988]]

acceleration at rollover available to consumers. Publication of an ROV 
model's rollover resistance value on a hang tag will allow consumers to 
make informed purchasing decisions regarding the comparative lateral 
stability of ROVs. In addition, publication of rollover resistance will 
provide a competitive incentive for manufacturers to improve the 
rollover resistance of their ROVs.
    3. Comment: Comments from the Companies and the Committee state 
that Kst is the more appropriate stability factor than SSF 
because it accounts for differences in the rear and track width, as 
well as differences in the fore and aft location of the vehicle's 
center of gravity.
    Response: Kst is a three-dimensional calculation of the 
two-dimensional SSF, and when the front and rear track widths are 
equal, Kst equals SSF. For practical purposes, 
Kst and SSF provide the same information on ROVs. Occupant-
loaded values of Kst and SSF are informative to the design 
process of ROVs; however, Kst and SSF values do not account 
for all the dynamic factors that affect actual rollover resistance. 
Therefore, they do not represent the best stability metric for ROVs.
    The Commission compared the actual lateral acceleration at rollover 
threshold of several ROVs, as measured by the J-turn test, and found 
that the static measures (whether Kst, SSF, or TTA) are not 
accurate predictors of the vehicle's actual lateral stability. Direct 
dynamic measurement of the vehicle's resistance to rollover is possible 
with ROVs. Therefore, the Commission believes that J-turn testing to 
determine the threshold lateral acceleration at rollover should be used 
as the standard requirement to determine lateral stability.
    4. Comment: Comments from CEI and the Companies state that tilt 
table angle or tilt table ratio should be used as a measure of lateral 
stability.
    Response: As stated above, the staff compared the actual lateral 
acceleration at rollover threshold of several ROVs, as measured by the 
J-turn test, and found that the static measures (whether it is 
Kst or SSF or TTA) are not accurate predictors of the 
vehicle's actual lateral stability.
    The Commission believes that the tilt table requirement in ANSI/
ROHVA 1-2011 does not adequately address lateral stability in ROVs. A 
comparison of how the vehicles would rank if the TTA were used instead 
of the direct measurement of lateral acceleration at rollover 
(Ay) illustrates how poorly the TTA correlates to the actual 
rollover resistance of the vehicle. The tilt table test does not 
account for dynamic effects of tire compliance, suspension compliance, 
and vehicle handling, which are important factors in the vehicle's 
lateral stability.
    Direct dynamic measurement of the vehicle's resistance to rollover 
is possible with ROVs. Therefore, the Commission believes that J-turn 
testing to determine the threshold lateral acceleration at rollover 
should be used as the standard requirement to determine lateral 
stability.
    5. Comment: Comments from the Companies state that the ANSI/ROHVA 
1, American National Standard for Recreational Off-Highway Vehicles, 
lateral stability requirement of Kst = 1 and TTA = 30 
degrees is adequate and should be adopted by CPSC.
    Response: SEA tested 10 representative ROV samples to the tilt 
table requirements in ANSI/ROHVA 1-2011. All of the ROVs tested pass 
the minimum 30-degree TTA, which indicates that the tilt table 
requirement is a status quo test. Vehicle D, the vehicle with the 
lowest rollover resistance (Ay = 0.625 g, TTA = 33.7 
degrees), exceeds the TTA requirement by 3.7 degrees, or 12 percent 
above the 30-degree minimum requirement. Vehicle A, the ROV that was 
part of a repair program to increase its roll resistance, exceeds the 
TTA requirement by 3.0 degrees, or 10 percent above the 30-degree 
minimum.
    CPSC believes the ANSI/ROHVA and ANSI/OPEI tilt table requirement 
is a requirement that all ROVs can pass and will not promote 
improvement among vehicles that have lower rollover resistance. The TTA 
requirement in the voluntary standard does not correlate to the actual 
rollover resistance of ROVs; the requirement allows the Yamaha Rhino to 
pass the test without having undergone the repair; and the requirement 
provides no incentive for manufacturers to improve the lateral 
stability of ROVs. The Commission believes that the threshold lateral 
acceleration at rollover value is a direct measure for rollover 
resistance, and its use would eliminate the need for tilt table testing 
as a requirement.
    6. Comment: Comments from the Companies, the Committee, and several 
individuals state that the SSF values recommended by CPSC staff for 
ROVs would make the vehicles unusable for off-road use and would 
eliminate this class of vehicle.
    Response: Based on the testing and data discussed in this preamble, 
CPSC staff no longer recommends using the SSF value as a measure of an 
ROV's rollover resistance. The SSF value of a vehicle represents the 
best theoretical lateral stability that the vehicle can achieve. CPSC 
staff compared the actual lateral acceleration at rollover threshold of 
several ROVs, as measured by the J-turn test, and found that the static 
measures (whether it is Kst, or SSF, or TTA) are not 
accurate predictors of the vehicle's actual lateral stability due to 
the extreme compliance in the vehicle's suspension and tires. 
Therefore, the Commission believes that neither the Kst, nor 
the SSF is an accurate measure of an ROV's lateral stability. Rather, 
the vehicle's actual lateral acceleration at rollover threshold is the 
appropriate measure of the vehicle's lateral stability.
Vehicle Handling
    1. Comment: Comments from CEI and the Companies state that 
measurements of understeer/oversteer made on pavement are not 
applicable to non-pavement surfaces. ROVs are intended for off-highway 
use and any pavement use is product misuse, they assert.
    Response: Both the ANSI/ROHVA and ANSI/OPEI standards specify 
dynamic testing on a paved surface. This indicates that ROHVA and OPEI 
agree that testing of ROVs on pavement is appropriate because pavement 
has a uniform high-friction surface. Tests conducted on pavement show 
how the vehicle responds at lateral accelerations that range from low 
lateral accelerations (associated with low friction surfaces like sand) 
up to the highest lateral acceleration that can be generated by 
friction at the vehicle's tires. This provides a complete picture of 
how the vehicle handles on all level surfaces. The amount of friction 
at the tires, and thus, the lateral accelerations generated, varies on 
non-paved surfaces. However, the vehicle's handling at each lateral 
acceleration does not change when the driving surface changes.
    2. Comment: Comments from CEI state that CEI has performed various 
tests and analyses on ROVs that demonstrate that ROVs that exhibit 
oversteer are not unstable.
    Response: The Commission disagrees with the statement that ROVs 
that exhibit oversteer are stable. Vehicles that exhibit sub-limit 
oversteer have a unique and undesirable characteristic, marked by a 
sudden increase in lateral acceleration during a turn. This dynamic 
instability is called critical speed and is described by Thomas D. 
Gillespie in the Fundamentals of Vehicle Dynamics as the speed ``above 
which the vehicle will be unstable.'' \38\ Gillespie further explains 
that an oversteer vehicle ``becomes

[[Page 68989]]

directionally unstable at and above the critical speed'' because the 
lateral acceleration gain approaches infinity.
---------------------------------------------------------------------------

    \38\ Gillespie, T. (1992). Fundamentals of Vehicle Dynamics. 
Society of Automotive Engineers, Inc. p. 204-205.
---------------------------------------------------------------------------

    CEI states that their tests demonstrate that ROVs that exhibit 
oversteer are not unstable. However, testing performed by SEA shows 
that oversteering ROVs can exhibit a sudden increase in lateral 
acceleration resulting in a roll over. Plots from SIS tests illustrate 
this sudden increase in lateral acceleration, which is found only in 
vehicles that exhibit sub-limit oversteer (see Figure 15). Vehicle A is 
an ROV that transitions to oversteer; Vehicle H is the same model ROV, 
but a later model year in which the oversteer has been corrected to 
understeer.
[GRAPHIC] [TIFF OMITTED] TP19NO14.014

    When Vehicle A reached its dynamically unstable condition, the 
lateral acceleration suddenly increased from 0.50 g to 0.69 g 
(difference of 0.19 g) in less than 1 second, and the vehicle rolled 
over. (Outriggers on the vehicle prevented full rollover of the 
vehicle.) In contrast, Vehicle H never reached a dynamically unstable 
condition because the condition does not develop in understeering 
vehicles. The increase in Vehicle H's lateral acceleration remains 
linear, and the lateral acceleration increase from 0.50 g to 0.69 g 
(same difference of 0.19 g) occurs in 5.5 seconds. A driver in Vehicle 
H has more margin to correct the steering to prevent rollover than a 
driver in Vehicle A because Vehicle H remains in understeer during the 
turn, while Vehicle A transitions to oversteer and becomes dynamically 
unstable.
    SEA test results indicate that ROVs that exhibited sub-limit 
oversteer also exhibited a sudden increase in lateral acceleration that 
caused the vehicle to roll over. An ROV that exhibits this sudden 
increase in lateral acceleration is directionally unstable and 
uncontrollable.\39\ Tests conducted by SEA provide strong evidence that 
sub-limit oversteer in ROVs is an unstable condition that can lead to a 
rollover incident, especially given the low rollover resistance of 
ROVs.
---------------------------------------------------------------------------

    \39\ Bundorf, R. T. (1967). The Influence of Vehicle Design 
Parameters on Characteristic Speed and Understeer. SAE 670078; 
Segel, L. (1957). Research in the Fundamentals of Automobile Control 
and Stability. SAE 570044.
---------------------------------------------------------------------------

    3. Comment: Comments from CEI and the Companies state that all 
vehicles, whether they understeer or oversteer, can be driven to limit 
conditions and can spin or plough. Any vehicle can exhibit ``limit 
oversteer'' through manipulation by the driver.
    Response: The Commission does not dispute that operator input and 
road conditions can affect limit oversteer or understeer in a vehicle. 
The vehicle handling requirements proposed by the Commission specify 
that vehicles exhibit sub-limit understeer. The Commission believes 
that sub-limit oversteer is an unstable condition that can lead to a 
rollover incident. Ten sample ROVs were tested by SEA; five of the 10 
vehicles exhibited a desirable sub-limit understeer condition, and five 
exhibited a transition to undesirable sub-limit oversteer condition. 
CPSC's evaluation indicates that ROVs can be designed to understeer 
with minimal cost and without diminishing the utility or recreational 
value of this class of vehicle.
    4. Comment: Comments from the Companies state that oversteer is 
desirable for path-following capability. Specifically, vehicles in 
oversteer will generally follow the path and allow directional control 
of the vehicle. High rear tire slip angles and tire longitudinal slip 
are needed for traction on off-highway surfaces, such as loose soil.
    Response: The Commission is not aware of any studies that define 
``path-following capability'' and its relation to the sub-limit 
understeer or oversteer design of the vehicle. Of the 10 sample ROVs 
tested by SEA, five vehicles exhibited a desirable sub-limit understeer 
condition. The Commission is not aware of any reports of the steering 
of sub-limit understeering vehicles causing loss of control or 
preventing the driver from navigating off-road terrain.
    A significant body of research has been developed over many years 
regarding the science of vehicle dynamic handling and control. The 
Commission has reviewed technical papers regarding vehicle handling 
research and finds no agreement with the statement that ``a vehicle in 
an oversteer condition will generally follow the path and allow 
directional control of the vehicle to be maintained longer.'' In fact, 
the Commission's research finds universal characterization of sub-limit 
oversteer as directionally unstable, highly undesirable, and 
dynamically unstable at or above the

[[Page 68990]]

critical speed.\40\ The Commission's review of 80 years of automotive 
research did not find support for the suggestion that sub-limit 
oversteer provides superior precision in handling and control.
---------------------------------------------------------------------------

    \40\ Olley, M. (1934). Independent Wheel Suspension--Its Whys 
and Wherefores. SAE 340080.; Stonex, K. A. (1941). Car Control 
Factors and Their Measurement. SAE 410092.; Segel, L. (1957). 
Research in the Fundamentals of Automobile Control and Stability. 
SAE 570044.; Bergman, W. (1965). The Basic Nature of Vehicle 
Understeer--Oversteer. SAE 650085.; Bundorf, R. T. and Leffert, R. 
L. (1976). The Cornering Compliance Concept for Description of 
Vehicle Directional Control Properties. SAE 760713.; and Milliken, 
William F., Jr., et al. (1976). The Static Directional Stability and 
Control of the Automobile. SAE 760712.
---------------------------------------------------------------------------

    Likewise, limit oversteer is described by the Companies as the 
result of the driver ``operating the vehicle in a turn at a speed 
beyond what is safe and reasonable for that turn or applying excessive 
power in a turn.'' A vehicle in limit oversteer is essentially sliding 
with the rear of the vehicle rotating about the yaw axis. A vehicle in 
a slide is susceptible to a tripped rollover. ROVs have low rollover 
resistance and are at high risk of a violent, tripped rollover. 
Autonomous vehicle testing by SEA has duplicated these limit oversteer 
conditions and found that tripped rollovers can create in excess of 2 g 
to 3 g of instantaneous lateral acceleration, which produces a violent 
rollover event. CPSC's evaluation indicates that eliminating sub-limit 
oversteer will reduce unintentional transitions to limit oversteer.
    The Commission does not agree that producing power oversteer by 
spinning the rear wheels is a necessity for negotiating low-friction, 
off-highway surfaces. Drifting or power oversteering is a risky 
practice that presents tripped rollover hazards and does not improve 
the vehicle's controllability. However, the practice of power 
oversteering is the result of driver choices that are not under the 
control of the manufacturer or the CPSC, and will not be significantly 
affected by the elimination of sub-limit oversteer.
    5. Comment: Comments from the Companies state that requiring ROVs 
to exhibit understeer characteristics could create unintended and 
adverse risk, such as gross loss of mobility. These commenters assert 
that CPSC would be trading one set of purported safety issues for 
another, equally challenging set of safety issues, and running against 
100 years of experience in off-highway vehicle design and driving 
practice, which suggests that for off-highway conditions, limit 
oversteer is at least sometimes, if not most often, preferable to limit 
understeer.
    Response: ROVs that exhibit sub-limit understeering are currently 
in the U.S. market in substantial numbers. The Commission is not aware 
of any reports of the steering of sub-limit understeering vehicles 
causing loss of control or preventing the driver from navigating off-
road terrain. The CPSC is not aware of any reports of sub-limit 
understeering vehicles that exhibit the unintended consequences 
described by the Companies.
    The Commission believes that sub-limit oversteer is an unstable 
condition that can lead to a rollover incident. Based on the Yamaha 
Rhino repair program and the SEA test results indicating that half of 
the sample ROVs tested already exhibit sub-limit understeer, the CPSC 
believes that ROVs can be designed to understeer with minimum cost and 
without diminishing the utility or recreational value of this class of 
vehicle.
    6. Comment: Comments from CEI, the Companies, and the Committee 
state that no correlation can be shown between understeer/oversteer and 
ROV crashes or rollovers.
    Response: From a design and engineering perspective, the physics of 
vehicle rollover inherently support the fact that increasing a 
vehicle's resistance to rollover will make the vehicle more stable. In 
addition, eliminating a vehicle characteristic that exhibits a sudden 
increase in lateral acceleration during a turn will reduce the risk of 
rollover. The constant radius tests and SIS tests conducted by SEA 
provide strong evidence that sub-limit oversteer is an unstable 
condition that can lead to a rollover incident.
    Of the 428 ROV-related incidents reviewed by the CPSC, 291 (68 
percent) involved lateral rollover of the vehicle, and more than half 
of these (52 percent) occurred while the vehicle was turning. Of the 
147 fatal incidents that involved rollover, 26 (18 percent) occurred on 
a paved surface. A vehicle exhibiting oversteer is most susceptible to 
rollover in a turn where the undesirable sudden increase in lateral 
acceleration can cause rollover to occur quickly, especially on paved 
surfaces, where an ROV can exhibit an untripped rollover.
    The Commission believes that improving the rollover resistance and 
vehicle steering characteristics of ROVs is a practical strategy for 
reducing the occurrence of ROV rollover events.
Occupant Protection
    1. Comment: Comments from CEI, the Companies, and the Committee 
state that seat belt use is critically important. Increasing seat belt 
use is the most productive and effective way to reduce ROV-related 
injuries and deaths because seat belt use is so low among those injured 
in ROV incidents. A major challenge is clearly how to get occupants to 
use the seat belt properly.
    Response: The Commission agrees that the use of seat belts is 
important in restraining occupants in the event of a rollover or other 
accident. Results of the Commission's testing of belted and unbelted 
occupants in simulated ROV rollover events indicate that seat belt use 
is required to retain occupants within the vehicle. Without seat belt 
use, occupants experience partial to full ejection from the vehicle. 
This scenario has been identified as an injury hazard in the CPSC's 
review of ROV-related incidents. Of those incidents that involved 
occupant ejection, many occupants suffered crushing injuries caused by 
the vehicle.
    After reviewing the literature regarding automotive seat belts, the 
Commission believes that an 8-second reminder light, as required in 
ANSI/ROHVA 1-2011 and ANSI/OPEI B71.9-2012, is not adequate to increase 
meaningfully seat belt use rates in ROVs because the system is not 
intrusive enough to motivate drivers and passengers to wear their seat 
belts. Results from past studies on automotive seat belt reminders 
conclude that visual reminders are ineffective. Numerous studies 
conclude further that effective reminder systems have to be intrusive 
enough to motivate users to buckle their seat belts. The more intrusive 
reminders are more effective at changing user behavior, as long as the 
reminder is not so intrusive that users bypass the system.
    Based on literature and results from the Westat study, the 
Commission believes that a seat belt speed limiting system that 
restricts the maximum speed of the vehicle to 15 mph, if the driver 
seat and any occupied front seats are not buckled, is the most 
effective method to increase meaningfully seat belt use rates in ROVs. 
The system is transparent to users at speeds of 15 mph and below, and 
the system consistently motivates occupants to buckle their seat belts 
to achieve speeds above 15 mph.
    2. Comment: Comments from CEI state that four-point and five-point 
seat belts are not appropriate for ROVs. In contrast, several 
individual comments state that five-point seat belts should be required 
on ROVs.
    Response: The Commission identified lack of seat belt use as an 
injury hazard in the CPSC's review of ROV-related incidents. The 
majority of safety restraints in the ROV incidents were

[[Page 68991]]

three-point restraints, and to some extent, two-point seat belts. 
Although four-point seat belts might be superior to three-point seat 
belts in retaining occupants in a vehicle, three-point seat belts have 
been shown to be effective in reducing the risk of death and serious 
injury in automotive applications. The Commission believes that it is 
unlikely that users who already do not use three-point seat belts will 
use the more cumbersome four-point and five-point seat belts.
    A more robust seat belt reminder system than the current voluntary 
standard requirement for a visual reminder light is necessary to 
motivate users to wear their seat belts because automotive studies of 
seat belt reminders indicate that visual reminders do not increase seat 
belt use. Dynamic rollover tests of ROVs indicate that a three-point 
seat belt, in conjunction with a passive shoulder restraint, is 
effective in restraining an occupant inside the protective zone of the 
vehicle's ROPS during a quarter-turn rollover.
    3. Comment: Comments from CEI state that occupant protection 
requirements should be based on meaningful tests.
    Response: The Commission agrees that ROV occupant protection 
performance evaluations should be based on actual ROV rollovers or 
simulations of real-world rollovers. Occupant protection performance 
requirements for ROVs in the voluntary standard developed by ROHVA 
(ANSI/ROHVA 1-2011) and the voluntary standard developed by OPEI (ANSI/
OPEI B71.9-2012) are not supported by data from rollover tests.
    The SEA roll simulator is the most accurate simulation of an ROV 
rollover event because it has been validated by measurements taken 
during actual ROV rollovers. Rollover tests indicate that a seat belt, 
used in conjunction with a passive shoulder barrier, is effective at 
restraining occupants within the protective zone of the vehicle's ROPS 
during quarter-turn rollover events.
ROV Incident Analysis
    1. Comment: Comments from CEI state that ROV rollover incidents are 
caused by a small minority of drivers who intentionally drive at the 
limits of the vehicle and the driver's abilities, and intentionally 
drive in extreme environments.
    Response: Of the 224 reported ROV incidents that involved at least 
one fatality, 147 incidents involved lateral rollover of the vehicle. 
Of the 147 lateral rollover fatalities, it is reported that the ROV was 
on flat terrain in 56 incidents (38 percent) and on a gentle incline in 
18 incidents (12 percent). Of the 224 fatal ROV incidents, the vehicle 
speed is unknown in 164 incidents (73 percent); 32 incidents (14 
percent) occurred at speeds of 20 miles per hour (mph) or less; and 28 
incidents (13 percent) occurred at speeds more than 20 mph. (Vehicle 
speeds were reported (i.e., not measured by instrumentation); so these 
speeds can be used qualitatively only and not as accurate values of 
speed at which incidents occurred.) Of the 224 fatal ROV incidents, the 
age of the driver was less than 16 years old in 61 incidents (27 
percent). Of the 231 fatalities, 77 victims (33 percent) were children 
less than 16 years of age.
    A review of the incident data shows no indication that the majority 
of rollover incidents are caused by drivers who ``purposely push the 
vehicle to and beyond its limits by engaging in stunts, racing, and 
intentional use of extreme environments.'' An analysis of the reported 
ROV incidents indicates that many of the details of the circumstances 
of the event, such as vehicle speed or terrain slope, are not known. In 
cases in which details of the event are known, roughly 50 percent of 
the fatal lateral rollover incidents occurred on flat or gentle slope 
terrain; and 14 percent occurred at speeds below 20 miles per hour. 
Twenty-seven percent of the drivers in fatal rollover incidents are 
children under 16 years of age; and 33 percent of all ROV-related 
fatalities are children under 16 years of age.
    2. Comment: Comments from the Companies state that the CPSC failed 
to use data from the NEISS in its analysis of ROV hazards. The comments 
suggest further that analysis of the NEISS data on utility-terrain 
vehicles (UTVs) indicate that UTVs, and therefore, ROVs, have a low 
hospitalization rate.
    Response: The joint comment's conclusions based on the commenters' 
analyses of the NEISS UTV data are not technically sound because the 
NEISS results do not specifically identify ROVs. NEISS has a product 
code for UTVs and several product codes for ATVs, but there is no 
separate product code for ROVs. ATVs have a straddle seat for the 
operator and handlebars for steering. UTVs have bucket or bench seats 
for the operator/passengers, a steering wheel for steering, and UTVs 
may or may not have a ROPS. ROVs are a subset of UTVs and are 
distinguished by having a ROPS, seat belts, and a maximum speed above 
30 mph. However, many official entities, news media, and consumers 
refer to ROVs as ATVs. Injuries associated with ROVs are usually 
assigned to either an ATV product category or to the UTV product 
category in NEISS. At a minimum, ROVs can be thought of as a subset of 
UTVs and/or ATVs, and cannot be identified on a consistent basis 
through the NEISS case records because NEISS requires knowledge of the 
make/model of the vehicle (which is not coded in the NEISS for any 
product). Occasionally, the NEISS narrative contains make/model 
identification, but this cannot be used to identify ROVs accurately and 
consistently.
    CPSC conducted a special study in 2010, in which all cases coded as 
ATVs or UTVs were selected for telephone interviews to gather 
information about the product involved. Sixteen of the 668 completed 
surveys had responses that identified the vehicle as an ROV. Staff's 
analysis shows that many ROVs are coded as ATVs; many UTVs are also 
coded as ATVs; and identification of ROVs and UTVs is difficult because 
the NEISS narratives often do not include enough information to 
identify the product. The miscoding rate for UTVs and ROVs is high, and 
most likely, the miscoding is due to consumer-reported information in 
the emergency department.
    The CPSC added the UTV product code 5044 to the NEISS in 2005. In 
the years 2005 to 2008 (the years cited in the joint comment document), 
the UTV product code had mostly out-of-scope records, with a large 
number of utility trailers and similar records. After these out-of-
scope records are removed, the only viable estimate is obtained by 
aggregating the cases across 2005 to 2008, to get an estimated 1,300 
emergency department-treated injuries related to UTVs (see Tab K, Table 
1). This estimate is considerably less than the estimate reported by 
Heiden in the joint comment. This estimate also does not include the 
UTV-related injuries that were miscoded as ATVs in the ATV product 
codes.
    As the years have passed and the UTV product code is being used 
more as intended, a completely different picture is seen for UTVs. From 
2009 to 2012, there are an estimated 6,200 emergency department-
treated, UTV-related injuries (which can be attributed to an increase 
in the number of UTV-related injuries, a larger portion of injuries 
being identified in NEISS as UTVs, or a combination of all of these and 
other factors not identified). Of these estimated 6,200 injuries, only 
80.2 percent are treated and released. The proportion of treated and 
released injuries for UTVs is significantly below the proportion of 
treated and released for all consumer products (92.0 percent of 
estimated consumer product-related, emergency department-treated 
injuries

[[Page 68992]]

were treated and released from 2009 to 2012). This illustrates a hazard 
of more severe injuries associated with UTVs.
    In conclusion, data are insufficient to support the argument that 
UTV injuries are not as severe as those associated with other products. 
As more data have become available in recent years, it appears that 
about 80 percent of the injuries associated with UTVs have been treated 
and released as compared to about 92 percent of the injuries associated 
with all consumer products.
    3. Comment: The Companies provided their own analysis of ROV-
related reports that were used in the CPSC's ANPR analysis. In 
particular, the Companies criticize Commission staff's analysis because 
asserting that staff's analysis did not include factors related to 
incident conditions and user behavior.
    Response: Commission staff's analysis of incidents for the ANPR was 
a preliminary review of reported incidents to understand the overall 
hazard patterns. For the NPR, Commission staff conducted an extensive, 
multidisciplinary review of 428 reported ROV-related incidents 
resulting in at least one death or injury. The results of this study 
are summarized in two reports in the NPR briefing package, along with 
analyses of victim characteristics, hazard patterns, environmental 
characteristics, and make and model characteristics. (The approach 
taken in the comments from the Companies, to remove reports from the 
analysis because there is unknown information, is not the Commission's 
approach in analyzing ROV-related incidents.) Unknowns from all reports 
are included with the knowns to ensure that the full picture is seen 
because every report will have at least one piece of unknown 
information, and every report will have at least one piece of known 
information. The unknowns are reported in all tables, if unknowns were 
recorded for the variables used.
    The analysis of IDIs summarized in the comments from the Companies 
does not define ``excessive speed,'' ``dangerous maneuver,'' or ``sharp 
turn.'' In fact, in other places in the comments, the companies 
mention: ``There is also no evidence suggesting that speed is an 
important factor in preventing accidents.'' The companies also state: 
``Tight steering turn capability is an important feature in certain 
ROVs, particularly those for trail use, because of the need to respond 
quickly to avoid obstacles and trail-edge drop-offs, and otherwise 
navigate in these off-highway terrains'' Thus, there is ambiguity in 
what the definitions could mean in the analysis of the IDIs (When is 
the vehicle at an excessive speed? When is a turn too sharp? When is a 
maneuver dangerous?). The Commission's approach to analyzing the 428 
incidents summarized in the reports available in the NPR briefing 
package is to consider the sequence of events, the vehicle, the driver, 
any passenger, and environment characteristics across all incidents. 
All definitions are set and used consistently by the multidisciplinary 
review team to understand the hazard patterns and incident 
characteristics across all incidents, not to set responsibility in one 
place or another.
    4. Comment: Comments from CEI state that the CPSC should begin to 
address human factors that pertain to risk-taking behavior of the small 
minority of ROV users who operate the vehicles at their limits without 
crash-worthiness concerns. In particular, CEI proposes that the CPSC 
focus primarily on changing consumer behavior to wearing seat belts, 
wearing helmets, and refraining from driving ROVs irresponsibly.
    Response: The Commission agrees that human factors and behavior 
affect the risk of death and injury for ROV users. However, the CPSC 
believes that establishing minimum requirements for ROVs can also 
reduce the hazards associated with ROVs. As explained in this preamble, 
the ANSI/ROHVA voluntary standard does not adequately addresses the 
risk of injury and death associated with lateral rollovers of ROVs 
because the standards do not have robust lateral stability 
requirements, do not have vehicle handling requirement to ensure 
understeer, and do not have robust occupant restraint requirements to 
protect occupants from vehicle rollover.
    An analysis of the reported ROV incidents indicates that many of 
the details of an event, such as vehicle speed or terrain slope, are 
not known. Where details of the event are known, roughly 50 percent of 
the fatal lateral rollover incidents occurred on flat or gentle slope 
terrain, and 14 percent occurred at speeds below 20 miles per hour. 
Twenty-seven percent of the drivers in fatal rollover incidents are 
children under 16 years of age; and 33 percent of all ROV-related 
fatalities are children under 16 years of age. There is no indication 
that the majority of rollover incidents are caused by drivers who 
intentionally drive under extreme conditions.
    Regarding seat belt use, results from past studies on automotive 
seat belt reminders conclude that visual seat belt reminders are 
ineffective. Numerous studies further conclude that effective reminder 
systems have to be intrusive enough to motivate users to buckle their 
seat belts. The more intrusive reminders are more effective at changing 
user behavior, as long as the reminder is not so intrusive that users 
bypass the system.
    The Commission believes that a seat belt speed-limiting system that 
restricts the maximum speed of the vehicle to 15 mph if the driver seat 
and any occupied front seats are not buckled is the most effective 
method to increase meaningfully seat belt use rates in ROVs. The system 
is transparent to users at speeds of 15 mph and below, and the system 
consistently motivates occupants to buckle their seat belts to achieve 
speeds above 15 mph.

IX. Description of the Proposed Rule

A. Scope, Purpose, and Compliance Dates--Sec.  1422.1

    The proposed standard would apply to ``recreational off-highway 
vehicles'' (ROVs), as defined, which would limit the scope to vehicles 
with a maximum speed greater than 30 mph. The proposed standard would 
include requirements relating to lateral acceleration, vehicle 
handling, and occupant protection. The requirements are intended to 
reduce or eliminate an unreasonable risk of injury associated with 
ROVs. The proposed standard would specifically exclude ``golf cars,'' 
``all-terrain vehicles,'' ``fun karts,'' ``go karts,'' and ``light 
utility vehicles,'' as defined by the relevant voluntary standards. The 
Commission proposes two compliance dates: ROVs would be required to 
comply with the lateral stability and vehicle handling requirements 
(Sec. Sec.  1422.3 and 1422.4) 180 days after publication of the final 
rule in the Federal Register. ROVs would be required to comply with the 
occupant protection requirements (Sec.  1422.5) 12 months after 
publication of the final rule in the Federal Register. The Commission 
recognizes that some ROV manufacturers will need to redesign and test 
new prototype vehicles to meet the occupant protection requirements. 
This design and test process is similar to the process that 
manufacturers use when introducing new model year vehicles. As 
described more fully in Section X, staff estimates that it will take 
approximately 9 person-months per ROV model to design, test, implement, 
and begin manufacturing vehicles to meet the occupant protection 
performance requirements. Therefore, the Commission believes that 12 
months is a reasonable time period for manufacturers to comply with all 
of new mandatory requirements.

[[Page 68993]]

B. Definitions--Sec.  1422.2

    The proposed standard would provide that the definitions in section 
3 of the Consumer Product Safety Act (15 U.S.C. 2051) apply. In 
addition, the proposed standard would include the following 
definitions:
     ``Recreational off-highway vehicle''--a motorized vehicle 
designed for off-highway use with the following features: Four or more 
wheels with pneumatic tires; bench or bucket seating for two or more 
occupants; automotive-type controls for steering, throttle, and 
braking; rollover protective structures (ROPS); occupant restraint; and 
maximum speed capability greater than 30 mph.
     ``two-wheel lift''--point at which the inside wheels of a 
turning vehicle lift off the ground, or when the uphill wheels of a 
vehicle on a tilt table lift off the table. Two-wheel lift is a 
precursor to a rollover event. We use the term ``two-wheel lift'' 
interchangeably with ``tip-up.''
     ``threshold lateral acceleration''--minimum lateral 
acceleration of the vehicle at two-wheel lift.

C. Requirements for Dynamic Lateral Stability--Sec.  1422.3

1. Proposed Performance Requirement
a. Description of Requirement
    The proposed rule would require that all ROVs meet a minimum 
requirement for lateral stability. The dynamic lateral stability 
requirement would set a minimum value for the lateral acceleration at 
rollover of 0.70 g, as determined by a 30 mph drop-throttle J-turn 
test. The 30 mph drop-throttle J-turn test uses a programmable steering 
controller to turn the test vehicle traveling at 30 mph at prescribed 
steering angles and rates to determine the minimum steering angle at 
which two-wheel lift is observed. These are the conditions and 
procedures that were used in testing with SEA. Under the proposed 
requirements, the data collected during these tests are analyzed to 
compute and verify the lateral acceleration at rollover for the 
vehicle. The greater the lateral acceleration value, the greater is the 
resistance of the ROV to tip or roll over.
b. Rationale
    The J-turn test is the most appropriate method to measure the 
rollover resistance of ROVs because the J-turn test has been evaluated 
by NHTSA as the most objective and repeatable method for vehicles with 
low rollover resistance. As discussed previously, static metrics, such 
as SSF and TTR, cannot be used to evaluate accurately ROV rollover 
resistance because static tests are unable to account fully for the 
dynamic tire deflections and suspension compliance exhibited by ROVs 
during a J-turn maneuver. The Commission also verified that the J-turn 
test is objective and repeatable for ROVs by conducting numerous J-turn 
tests on several ROVs.
    As explained above, testing conducted by CPSC staff and SEA 
supports the proposed requirement that ROVs demonstrate a minimum 
threshold lateral acceleration at rollover of 0.70 g or greater in a J-
turn. Results of J-turn tests performed on a sample of 10 ROVs 
available in the U.S. market indicate that six of the 10 ROVs tested 
measured threshold lateral accelerations below 0.70 g (values ranged 
from 0.625 g to 0.690 g). The Commission believes that minor changes to 
vehicle suspension and/or track width spacing, similar to the changes 
in the Yamaha Rhino repair program, can increase the threshold lateral 
acceleration of these vehicles to 0.70 g or greater. The Yamaha repair 
program improved the rollover resistance of the Yamaha Rhino from 0.670 
g (unrepaired Yamaha Rhino) to 0.705 g (repaired Yamaha Rhino).
    Based on CPSC's evaluation of ROV testing and the decrease in 
injuries and deaths associated with Yamaha Rhino vehicles after the 
repair program was implemented, the Commission believes that improving 
the rollover resistance of all ROVs can reduce injuries and deaths 
associated with ROV rollover events.
2. Proposed Requirements for Hang Tag
a. Description of Requirement
    The Commission is proposing a requirement that ROV manufacturers 
provide technical information for consumers on a hangtag at the point 
of purchase.
    As discussed previously, the Commission is proposing a requirement 
that ROVs meet a minimum lateral acceleration of 0.70 g at rollover, as 
identified by J-turn testing. The Commission proposes requiring a 
hangtag on each ROV that would state the actual measured lateral 
acceleration at rollover (as identified by the J-turn testing) of each 
ROV model. The Commission believes that the hang tag will allow 
consumers to make informed decisions on the comparative lateral 
stability of ROVs when making a purchase and will provide a competitive 
incentive for manufacturers to improve the rollover resistance of ROVs.
    The proposed rule specifies the content and format for the hang 
tag, and includes an example hang tag. Under the proposal, the hang tag 
must conform in content, form, and sequence as specified in the 
proposed rule.
    The Commission proposes the following ROV hangtag requirements:
     Content. Every ROV shall be offered for sale with a 
hangtag that graphically illustrates and textually states the lateral 
acceleration threshold at rollover for that ROV model. The hangtag 
shall be attached to the ROV and may be removed only by the first 
purchaser.
     Size. Every hangtag shall be at least 15.24 cm (6 inches) 
wide by 10.16 cm (4 inches) tall.
     Attachment. Every hangtag shall be attached to the ROV and 
be conspicuous to a person sitting in the driver's seat; and the 
hangtag shall be removable only with deliberate effort.
     Format. The hang tag shall provide all of the elements 
shown in the example hangtag (see Figure 16).
b. Rationale
    Section 27(e) of the CPSA authorizes the Commission to require, by 
rule, that manufacturers of consumer products provide to the Commission 
performance and technical data related to performance and safety as may 
be required to carry out the purposes of the CPSA, and to give 
notification of such performance and technical data at the time of 
original purchase to prospective purchasers and to the first purchaser 
of the product. 15 U.S.C. 2076(e)). Section 2 of the CPSA provides that 
one purpose of the CPSA is to ``assist consumers in evaluating the 
comparative safety of consumer products.'' 15 U.S.C. 2051(b)(2).
    Other federal government agencies currently require on-product 
labels with information to help consumers in making purchasing 
decisions. For example, NHTSA requires automobiles to come with 
comparative information on vehicles regarding rollover resistance. 49 
CFR 575.105. NHTSA believes that consumer information on the rollover 
risk of passenger cars would influence consumers to purchase vehicles 
with a lower rollover risk and inspire manufacturers to produce 
vehicles with a lower rollover risk.\41\ A subsequent study of SSF 
trends in automobiles found that SSF values increased for all vehicles 
after 2001, particularly SUVs, which tended to have the worst SSF 
values in the earlier years.\42\
---------------------------------------------------------------------------

    \41\ Walz, M. C. (2005). Trends in the Static Stability Factor 
of Passenger Cars, Light Trucks, and Vans. DOT HS 809 868. Retrieved 
from https://www.nhtsa.gov/cars/rules/regrev/evaluate/809868/pages/.
    \42\ Walz, M.C. (2005). Trends in the Static Stability Factor of 
Passenger Cars, Light Trucks, and Vans. DOT HS 809868. Retrieved 
from https://www.nhtsa.gov/cars/rules/regrev/evaluate/809868/pages/.

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

[[Page 68994]]

    EnergyGuide labels, required on most appliances, are another 
example of federally-mandated labels to assist consumers in making 
purchase decisions. 16 CFR part 305. Detailed operating cost and energy 
consumption information on these labels allows consumers to compare 
competing models and identify higher efficiency products. The 
EnergyGuide label design was developed based on extensive consumer 
research and following a two-year rulemaking process.
    Like NHTSA rollover resistance information and EnergyGuide labels, 
the proposed ROV hang tags are intended to provide important 
information to consumers at the time of purchase. Providing the value 
of each ROV model vehicle's threshold lateral acceleration to consumers 
will assist consumers with evaluating the comparative safety of the 
vehicles in terms of resistance to rollover. Requiring that ROV lateral 
acceleration test results be stated on a hangtag may motivate 
manufacturers to increase the performance of their ROV to achieve a 
higher reportable lateral acceleration, similar to incentives created 
as a result of NHTSA's NCAP program.
    The proposed hangtag is based, in part, on the point-of-purchase 
hangtag requirements for ATVs. ATVs must have hangtags that include 
general warning information regarding operation and operator and 
passenger requirements, as well as behavior that is warned against. 
Most ROV manufacturers are also manufacturers of ATVs. Accordingly, ROV 
manufacturers are likely to be familiar with the hangtag requirements 
for ATVs. The ANSI/SVIA 1-2010 voluntary standard that applies to ATVs 
requires ATVs to be sold with a hangtag that is to be removed only by 
the purchaser and requires ATV hangtags to be 6-inches tall x 4-inches 
wide. Because ROV manufacturers are likely to be familiar with the 
hangtag requirements for ATVs, the Commission is proposing the same 
size requirements for ROV hang tags.
    The hang tag graph draws its format from well-recognized principles 
in effective warnings. When presenting graphical information, it is 
important to include labels so that the data can be understood. Graphs 
should have a unique title, and the axes should be fully labeled with 
the units of measurement. Graphs should also be distinguished from the 
text, by adding white space, or enclosing the graphs in a box.\43\
---------------------------------------------------------------------------

    \43\ Markel, M. (2001). Technical Communication. Boston, MA: 
Bedford/St. Martin's.
[GRAPHIC] [TIFF OMITTED] TP19NO14.015


[[Page 68995]]


    (1) The ROV icon helps identify the product. The icon is presented 
at a slight angle to help consumers readily identify the label as 
addressing ROV rollover characteristics. Research has shown that 
pictorial symbols and icons make warnings more noticeable and easier to 
detect than warnings without such symbols and icons.\45\
---------------------------------------------------------------------------

    \44\ Hang tag not shown to scale.
    \45\ Wogalter, M., Dejoy, D., and Laughery, K. (1999). Warnings 
and Risk Communication. Philadelphia, PA: Taylor & Francis, Inc.
---------------------------------------------------------------------------

    (2) Graph label, ``Better,'' indicates that the higher the value 
(as shading increases to the right), the higher the ROV's resistance to 
rolling over during a turn on a flat surface.
    (3) The Manufacturer, Model, Model number, Model year help the 
consumer identify the exact ROV described by the label. Likewise, the 
EnergyGuide label provides information on the manufacturer, model, and 
size of the product so that consumers can identify exactly what 
appliance the label describes.\46\ The Commission is proposing a 
similar identification of the ROV model on the hangtag so that 
consumers can compare values among different model ROVs.
---------------------------------------------------------------------------

    \46\ Guide to EnergyGuide label retrieved at https://www.consumer.ftc.gov/articles/0072-shopping-home-appliances-use-energyguide-label.
---------------------------------------------------------------------------

    (4) Textual information. Technical communication that includes 
graphs should also include text to paraphrase the importance of the 
graphic and explain how to interpret the information presented.\47\ 
Additionally, including a graphic before introducing text may serve as 
a valuable reference for consumers, by maintaining attention and 
encouraging further reading.\48\ The textual informational in the 
hangtag provides consumers with more definition of the values given in 
the graph.
---------------------------------------------------------------------------

    \47\ Markel, M. 2001.
    \48\ Smith, T.P. (2003). Developing consumer product 
instructions. Washington, DC: U.S. Consumer Product Safety 
Commission.
---------------------------------------------------------------------------

    (5) Linear scale, and anchor showing minimally acceptable value on 
the scale. Currently, the EnergyGuide label uses a linear scale with 
the lowest and highest operating costs for similar models so that 
consumers can compare products; the yearly operating cost for the 
specific model is identified on the linear scale.\49\ The Commission is 
proposing a linear scale format for the ROV hangtag, as well. The text 
identifies the minimally accepted lateral acceleration at rollover as 
being 0.7 g. When providing this on the scale, people are able to 
determine visually how a specific model compares to the minimal value.
---------------------------------------------------------------------------

    \49\ FTC. Retrieved from: https://www.consumer.ftc.gov/articles/0072-shopping-home-appliances-use-energyguide-label.
---------------------------------------------------------------------------

    (6) Scale starts at 0.65 g to allow a shaded bar for those ROVs 
meeting only the minimally acceptable lateral acceleration value.

D. Vehicle Handling--Sec.  1422.4

1. Description of Requirement
    The proposed rule would require that all ROVs meet a vehicle 
handling requirement, which requires that ROVs exhibit understeer 
characteristics. The understeer requirement would mandate that ROVs 
exhibit understeer characteristics in the sublimit range of the turn 
circle test. The test for vehicle handling or understeer performance 
involves driving the vehicle around a 100-foot radius circle at 
increasing speeds, with the driver making every effort to maintain 
compliance of the vehicle path relative to the circle. SEA testing was 
based on a 100-foot radius circle. Data collected during these tests 
are analyzed to determine whether the vehicle understeers through the 
required range. The proposed rule would require that all ROVs exhibit 
understeer for values of ground plane lateral acceleration from 0.10 to 
0.50 g.
 2. Rationale
    The CPSC believes that the constant radius test is the most 
appropriate method to measure an ROV's steering gradient because SAE 
J266, Surface Vehicle Recommended Practice, Steady-State Directional 
Control Test Procedures for Passenger Cars and Light Trucks, 
establishes the constant radius test as a method to measure understeer/
oversteer in passenger cars. The test procedures are also applicable to 
ROVs because ROVs are similar to cars, have four steerable wheels and a 
suspension system, and thus, ROVs obey the same principles of motion as 
automobiles.
    The Commission believes that the appropriate lateral acceleration 
range to measure steering gradient is from 0.10 g to 0.50 g because SEA 
test results indicate that spurious data occur at the beginning and end 
of a constant radius test conducted up to vehicle rollover. Data 
collected in the range of 0.10 g to 0.50 g of lateral acceleration 
provide the most accurate plots of the vehicle's steering 
characteristic.\50\
---------------------------------------------------------------------------

    \50\ Heydinger, G. (2011) Vehicle Characteristics Measurements 
of Recreational Off-Highway Vehicles. Retrieved from https://www.cpsc.gov/PageFiles/96037/rov.pdf. Page 18.
---------------------------------------------------------------------------

    Tests conducted by SEA show that ROVs in sub-limit oversteer 
transition to a condition where the lateral acceleration increases 
suddenly and exponentially. Based on testing and relevant literature, 
the CPSC believes that this condition can lead to untripped ROV 
rollovers or may cause ROVs to slide into limit oversteer and 
experience tripped rollover. Ensuring sub-limit understeer eliminates 
the potential for sudden and exponential increase in lateral 
acceleration that can cause ROV rollovers.
    The decrease in Rhino-related incidents after the repair program 
was initiated and the low number of vehicle rollover incidents 
associated with repaired Rhino vehicles are evidence that increasing 
the lateral stability of an ROV and correcting oversteer 
characteristics to understeer reduces the occurrence of ROV rollover on 
level terrain. In particular, the Commission believes the elimination 
of runaway lateral acceleration associated with oversteer contributed 
to a decrease in Rhino-related rollover incidents.
    As mentioned previously, ROVs can be designed to understeer in sub-
limit operation with minimum cost and without diminishing the utility 
or recreational value of this class of vehicle. Half of the vehicles 
CPSC tested already exhibit sub-limit understeer condition for the full 
range of the test, and this includes both utility and recreational 
model ROVs.

E. Occupant Retention System--Sec.  1422.5

    The proposed rule includes two requirements that are intended to 
keep the occupant within the vehicle or the ROPs. First, each ROV would 
be required to have a means to restrict occupant egress and excursion 
in the shoulder/hip zone defined by the proposed rule. This requirement 
could be met by a fixed barrier structure or structure on the ROV or by 
a barrier or structure that can be put into place by the occupant using 
one hand in one operation, such as a door. Second, the proposed rule 
would require that the speed of an ROV be limited to a maximum of 15 
mph, unless the seat belts for both the driver and any front seat 
passengers are fastened. The purpose of these requirements is to 
prevent deaths and injury incidents, especially incidents that involve 
full or partial ejection of the rider from the vehicle.
1. Speed Limitation
a. Requirement
    The Commission proposes a performance requirement that limits the 
maximum speed that an ROV can attain to 15 mph or less when tested with 
unbuckled front seat belts during the maximum speed test. Section 5 of 
ANSI/

[[Page 68996]]

ROHVA 1-2011, ``Maximum Speed,'' establishes test protocols to measure 
maximum speed on level ground. Because ROV manufacturers are already 
familiar with these test procedures and the proposed test would add 
elements to a test procedure manufacturers already conduct to meet the 
voluntary standard, the CPSC believes that the maximum speed test from 
ANSI/ROHVA 1-2011 is the most appropriate method to measure the limited 
speed of an ROV.
b. Rationale
i. Importance of Seat Belts
    As discussed in section V of this preamble, results of the CPSC's 
exploratory testing of belted and unbelted occupants in simulated ROV 
rollover events indicate that seat belt use is required to retain 
occupants within the vehicle. This conclusion corresponds with the 
incident data for ROV rollovers, in which 91 percent of the fatal 
victims who were partially or fully ejected from the vehicle were not 
wearing seat belts. Of the incidents that involved occupant ejection, 
many occupants were injured when struck by the vehicle after ejection. 
The Commission believes that many of the ROV occupant ejection deaths 
and injuries can be eliminated if occupants wear seat belts.
    Studies have shown that automobile seat belt reminders do not 
increase seat belt use, unless the reminders are aggressive enough to 
motivate users to buckle seat belts without alienating the user into 
bypassing or rejecting the system. Based on the Commission's testing 
and literature review and the low seat belt use rates in ROV-related 
incidents, the Commission believes that a seat belt speed limiting 
system that restricts the maximum speed of the vehicle to 15 mph if any 
occupied front seats are not buckled, is the most effective method to 
increase seat belt use rates in ROVs.
ii. Likely Acceptance of Speed-Limitation Technology
    The Commission believes that in-vehicle technology that limits the 
speed of the ROV if the front occupied seats are not buckled will be 
accepted by ROV users because the technology does not interfere with 
the operation of the ROV below the threshold speed, and users will be 
motivated to wear seat belts if they wish to exceed the threshold 
speed. This conclusion is based on automotive studies that show drivers 
accepted a system that reduced vehicle function (i.e., requiring more 
effort to depress the accelerator pedal) after a threshold speed, if 
the driver's seat belt was not buckled. The system did not interfere 
with the operation of the vehicle below the threshold speed, and 
drivers were willing to buckle their seat belts to access unhindered 
speed capability of the vehicle.
    The Commission also believes that speed-limitation technology will 
be accepted by ROV users because the technology is already included on 
the BRP Can-Am Commander and Can-Am Maverick model ROVs, and the 
manufacturer with the largest ROV market share, Polaris, announced that 
it will introduce the technology on model year 2015 Ranger and RZR 
ROVs.
    The Commission's literature review concludes that intrusive 
reminders are effective at changing user behavior, as long as the 
reminder is not so intrusive that users bypass the system. Limitation 
of vehicle speed is the intrusive reminder for ROV users to buckle 
their seat belt; therefore, the Commission believes that the threshold 
speed for a seat belt speed-limitation system should be as high as 
possible to gain user acceptance (and reduce bypass of the system), but 
low enough to allow relatively safe operation of the vehicle.
iii. Choice of 15 MPH
    The Commission believes 15 mph is the appropriate speed threshold 
for a seat belt speed-limitation system. Based on information about 
ROVs and vehicles similar to ROVs, the Commission concludes that ROVs 
can be operated relatively safely at 15 mph. For example:
     ANSI/NGCMA Z130.1-2004, American National Standard for 
Golf Carts--Safety and Performance Specifications, specifies the 
maximum speed for golf carts at 15 mph. This standard establishes 15 
mph as the maximum acceptable speed for unbelted drivers and passengers 
(golf carts do not have seat belts or ROPS) in vehicles that are often 
driven in off-road conditions.
     SAE J2258, Surface Vehicle Standard for Light Utility 
Vehicles, specifies a speed of 15 mph as acceptable for a vehicle, with 
a lateral stability of at least 25 degrees on a tilt table test, 
without seat belts or ROPS. This standard also establishes 15 mph as 
the maximum acceptable speed for unbelted drivers and passengers in 
vehicles that are driven in off-road conditions.
     Polaris Ranger and RZR model year 2015 ROVs will be 
equipped with a seat belt speed limiter that limits the vehicle speed 
to 15 mph if the driver's seat belt is not buckled. The decision by the 
largest manufacturer of ROVs establishes 15 mph as the maximum 
acceptable speed for unbelted ROV drivers.
    Additionally, the principles of physics support this conclusion. 
The fundamental relationship between speed and lateral acceleration is:

A = V\2\/R where A = lateral acceleration
V = velocity
R = radius of turn

    The minimum proposed lateral acceleration threshold at rollover for 
ROVs is 0.70 g, and the typical turn radius of an ROV is 16 feet.\51\ 
Therefore, without any additional effects of tire friction, the speed 
at which rollover would occur during a turn on level ground is 13 mph. 
(The CPSC recognizes that on a slope, the lateral acceleration due to 
gravity can cause ROV rollover at speeds below 15 mph. However, the 
CPSC believes that it is appropriate to use level ground as a 
baseline.) In reality, friction at the tires would increase the speed 
at which rollover occurs to above 13 mph.
---------------------------------------------------------------------------

    \51\ Turn radius values retrieved at: https://www.atv.com/features/choosing-a-work-vehicle-atv-vs-utv-2120.html and https://www.utvunderground.com/2014-kawasaki-teryx-4-le-6346.html.
---------------------------------------------------------------------------

iv. User Acceptance of 15 mph
    Based on CPSC's study and the experience of some ROVs that have 
speed limitations, the Commission believes that ROV users are likely to 
accept a 15 mph threshold speed limitation. The following reasons 
support this conclusion:
     Results of Westat's Phase 1 focus group study of ROV users 
indicate that ROV users value easy ingress and egress from an ROV and 
generally drive around 15 mph to 30 mph during typical use of the ROV. 
Users had mixed reactions to a speed threshold of 10 mph and were more 
accepting of a speed-limitation technology if the threshold speed was 
15 mph.
     There are many situations in which an ROV is used at slow 
speeds, such as mowing or plowing, carrying tools to jobsites, and 
checking property. The Commission believes that a speed-limitation 
threshold of 15 mph allows the most latitude for ROV users to perform 
utility tasks where seat belt use is often undesired.
     The Commission believes that ROV user acceptance of a seat 
belt speed-limitation system will be higher at 15 mph than the speed 
threshold of 9 mph on the Commander ROV. Although BRP continues to sell 
the Can-Am Commander and Can-Am Maverick ROVs with speed limitations 
set at around 10 mph, focus group responses indicate that many ROV 
users believe that 10 mph is too low a speed limit to

[[Page 68997]]

be acceptable, and therefore, these users will bypass the system. The 
15 mph threshold is 50 percent higher than a 10 mph threshold, and 
staff believes that the difference in the speed threshold will increase 
user acceptance of the system. Polaris's decision to include seat belt 
speed limiters with a 15 mph threshold speed in model year 2015 Ranger 
and RZR ROVs supports the Commission's belief that user acceptance of a 
speed-limitation system will be higher at 15 mph than 10 mph.
2. Shoulder Probe Test
a. Requirement
    CPSC is proposing a performance requirement that ROVs pass a probe 
test at a defined area near the ROV occupants' shoulder. The probe test 
is the most appropriate method to measure the occupant protection 
performance in the shoulder area of the ROV because various forms of 
the probe test are already used in the voluntary standard for ROVs and 
ATVs to determine occupant protection performance.
    The test applies a probe with a force of 163 lbs., to a defined 
area of the vehicle's ROPS near the ROV occupants' shoulder. The 
vertical and forward locations for the point of application of the 
probe are based upon anthropometric data. The probe dimensions are 
based on the upper arm of a 5th percentile adult female, and the 
dimensions of a 5th percentile adult female represent the smallest size 
occupant that may be driving or riding an ROV. The 163 lb. force 
application represents a 50th percentile adult male occupant pushing 
against the barrier during a rollover event. The probe is applied for 
10 seconds and the vehicle structure must absorb the force without 
bending more than 1 inch.
b. Rationale
    After exploring several methods to test occupant protection 
performance of ROVs during a rollover event, CPSC believes the SEA roll 
simulator is the most accurate simulation of a rollover because the 
roll simulator is able to reproduce the lateral acceleration and roll 
rate experienced by ROVs in rollover events. SEA conducted simulations 
of tripped and untripped rollovers on ROVs with belted and unbelted ATD 
occupants. CPSC's analysis of SEA's test results indicate that the best 
occupant retention performance results, where occupants remain within 
the protective zone of the vehicle's ROPS, occurred when a seat belt is 
used in conjunction with a passive shoulder barrier restraint.

F. Prohibited Stockpiling--Sec.  1422.6

    The proposed rule contains anti-stockpiling provisions to prohibit 
excessive production or importation of noncomplying ROVs during the 
period between the final rule's publication and its effective date. 
Anti-stockpiling provisions typically exist to prevent the production 
or importation of significant numbers--significantly beyond typical 
rates--of noncomplying products that can be sold after the effective 
date of a safety standard, which could present an unreasonable risk of 
injury to consumers. In order to balance the protection of consumers 
and the burden to manufacturers and importers of compliance with the 
effective date of a rule, a production limit is typically set at some 
minimal percentage above a single year's production rate as selected by 
the manufacturer or importer. This allows the manufacturer or importer 
to select the date most conductive to compliance, even if production or 
importation occurs at an unusually robust pace during the selected 
period.
    The prohibited stockpiling provision herein limits the production 
or importation of noncomplying products to 10% of the amount produced 
or imported in any 365-day period designated, at the option of each 
manufacturer or importer, beginning on or after October 1, 2009, and 
ending on or before the date of promulgation of the rule.

G. Findings--Sec.  1422.7

    In accordance with the requirements of the CPSA, we are proposing 
to make the findings stated in section 9 of the CPSA. The proposed 
findings are discussed in section XVI of this preamble.

X. Preliminary Regulatory Analysis

    The Commission is proposing to issue a rule under sections 7 and 9 
of the CPSA. The CPSA requires that the Commission prepare a 
preliminary regulatory analysis and that the preliminary regulatory 
analysis be published with the text of the proposed rule. 15 U.S.C. 
2058(c). The following discussion is extracted from staff's memorandum, 
``Draft Proposed Rule Establishing Safety Standard for Recreational 
Off-Road Vehicles: Preliminary Regulatory Analysis.''

A. Introduction

    The CPSC is issuing a proposed rule for ROVs. This rulemaking 
proceeding was initiated by an ANPR published in the Federal Register 
on October 28, 2009. The proposed rule includes: (1) Lateral stability 
and vehicle handling requirements that specify a minimum level of 
rollover resistance for ROVs and requires that ROVs exhibit sublimit 
understeer characteristics, and (2) occupant retention requirements 
that would limit the maximum speed of an ROV to no more than 15 miles 
per hour (mph), unless the seat belts of both the driver and front 
passengers, if any, are fastened; and in addition, would require ROVs 
to have a passive means, such as a barrier or structure, to limit 
further the ejection of a belted occupant in the event of a rollover.
    Following is a preliminary regulatory analysis of the proposed 
rule, including a description of the potential costs and potential 
benefits. Each element of the proposed rule is discussed separately. 
For some elements, the benefits and costs cannot be quantified in 
monetary terms. Where this is the case, the potential costs and 
benefits are described and discussed conceptually.

B. Market Information

1. Manufacturers and Market Shares
    The number of manufacturers marketing ROVs in the United States has 
increased substantially in recent years. The first utility vehicle that 
exceeded 30 mph, thus putting the utility vehicle in the ROV category, 
was introduced in the late 1990s. No other manufacturer offered an ROV 
until 2003. In 2013, there were 20 manufacturers known to CPSC to be 
supplying ROVs to the U.S. market. One manufacturer accounted for about 
60 percent of the ROVs sold in the United States in 2013. Another seven 
manufacturers, including one based in China, accounted for about 36 
percent of the ROVs sold in the same year. None of these seven 
manufacturers accounted for more than 10 percent of the market. The 
rest of the market was divided among about 12 other manufacturers, most 
of which were based in China or Taiwan.\52\ Commission staff's analysis 
attempted to exclude vehicles that had mostly industrial or commercial 
applications and were not likely to be purchased by consumers. The 
Commission has identified more than 150 individual ROV models from 
among these manufacturers. However, this count includes some models 
that appear to be very similar to other models produced by the same 
manufacturer but sold through different distributors in the United 
States.
---------------------------------------------------------------------------

    \52\ Market share is based upon Commission analysis of sales 
data provided by Power Products Marketing, Eden Prairie, MN (2014).
---------------------------------------------------------------------------

    About 92 percent of ROVs sold in in the United States are 
manufactured in North America. About 7 percent of the ROVs sold in the 
United States are

[[Page 68998]]

manufactured in China (by nine different manufacturers). Less than 1 
percent of ROVs are produced in other countries other than the United 
States or China.\53\
---------------------------------------------------------------------------

    \53\ This information is based upon a Commission analysis of 
sales data provided by Power Products Marketing, Eden Prairie, MN 
(2012).
---------------------------------------------------------------------------

    Seven recreational vehicle manufacturers, which together account 
for more than 90 percent of the ROV market, established ROHVA. The 
stated purpose of ROHVA is ``to promote the safe and responsible use of 
recreational off-highway vehicles (ROVs) manufactured or distributed in 
North America.'' ROHVA is accredited by the American National Standards 
Institute (ANSI) to develop voluntary standards for ROVs. ROHVA members 
have developed a voluntary standard (ANSI/ROHVA 1-2011) that sets some 
mechanical and performance requirements for ROVs. Some ROV 
manufacturers that emphasize the utility applications of their vehicles 
have worked with the Outdoor Power Equipment Institute (OPEI) to 
develop another ANSI voluntary standard that is applicable to ROVs 
(ANSI/OPEI B71.9-2012). This voluntary standard also sets mechanical 
and performance requirements for ROVs. The requirements of both 
voluntary standards are similar, but not identical.
2. Retail Prices
    The average manufacturer's suggested retail price (MSRP) of ROVs in 
2013 was approximately $13,100, with a range of about $3,600 to 
$20,100. The average MSRP for the eight largest manufacturers (in terms 
of market share) was about $13,300. The average MSRP of ROVs sold by 
the smaller, mostly Chinese manufacturers was about $7,900.\54\
---------------------------------------------------------------------------

    \54\ MSRPs for ROVs were reported by Power Products Marketing, 
Eden Prairie, MN (2014).
---------------------------------------------------------------------------

    The retail prices of ROVs tend to be somewhat higher than the 
retail prices of other recreational and utility vehicles. The MSRPs of 
ROVs are about 10 percent higher, on average, than the MSRPs of low-
speed utility vehicles. A comparison of MSRPs for the major 
manufacturers of ATVs and ROVs indicates that ROVs are priced about 10 
percent to 35 percent higher than ATVs offered by the same 
manufacturer.\55\ Another source indicates that the price of one ROV or 
other utility vehicle is about two-thirds the price of two ATVs.\56\ 
Go-karts usually retail for between $2,500 and $8,000.\57\
---------------------------------------------------------------------------

    \55\ This information is based upon a Commission analysis of 
data provided by Power Products Marketing, Eden Prairie, MN, (2014), 
and an examination of the suggested retail prices on several 
manufacturers' Internet sites.
    \56\ ``2009 Utility Vehicle Review,'' Southern Sporting Journal, 
October 2008, Vol. 14, Issue 5, pp. 58-70, accessed through: https://web.ebscohost.com on March 17. 2011.
    \57\ Tom Behrens, ``Kart Racing: Fast times out on the 
prairie,'' The Houston Chronicle, November 27, 2008, p. 4. (accessed 
from https://www.chron.com on January 17, 2014).
---------------------------------------------------------------------------

3. Sales and Number in Use
    Sales of ROVs have increased substantially since their 
introduction. In 1998, only one firm manufactured ROVs, and fewer than 
2,000 units were sold. By 2003, when a second major manufacturer 
entered the market, almost 20,000 ROVs were sold. The only dip in sales 
occurred around 2008, which coincided with the worst period of the 
credit crisis and a recession that also started about the same time. In 
2013, an estimated 234,000 ROVs were sold by 20 different 
manufacturers.\58\ The chart below shows ROV sales from 1998 through 
2013.
---------------------------------------------------------------------------

    \58\ This information is based upon a Commission analysis of 
sales data provided by Power Products Marketing, Eden Prairie, MN.
---------------------------------------------------------------------------

    The number of ROVs available for use has also increased 
substantially. Because ROVs are a relatively new product, we do not 
have specific information on the expected useful life of ROVs. However, 
using the same operability rates that CPSC uses for ATVs, we estimate 
that there were about 570,000 ROVs available for use in 2010.\59\ By 
the end of 2013, there were an estimated 1.2 million ROVs in use. (See 
Figure 17).
---------------------------------------------------------------------------

    \59\ CPSC Memorandum from Mark S. Levenson, Division of Hazard 
Analysis, to Susan Ahmed, Associate Executive Director, Directorate 
for Epidemiology, ``2001 ATV Operability Rate Analysis,'' U.S. 
Consumer Product Safety Commission, Bethesda Maryland (19 August 
2003). ``Operability rate'' refers to the probability that an ATV 
will remain in operation each year after the initial year of 
production.

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

[[Page 68999]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.016

    Most ROVs are sold through retail dealers. Generally, dealers that 
offer ROVs also offer other products, such as motorcycles, scooters, 
ATVs, and similar vehicles. ROVs are also sold through dealers that 
carry farm equipment or commercial turf management supplies.
    While sales of ROVs have increased over the last several years, 
sales of competing vehicles have leveled off, or declined. Low-speed 
utility vehicles have been on the market since the early 1980s. Their 
sales increased from about 50,000 vehicles in 1998, to about 150,000 
vehicles in 2007. In 2011, however, sales fell to about 110,000 
vehicles. A substantial portion of these sales were for commercial 
applications rather than consumer applications.\60\
---------------------------------------------------------------------------

    \60\ This information is based upon a Commission analysis of 
information provided by Power Products Marketing of Eden Prairie, 
MN.
---------------------------------------------------------------------------

    After several years of rapid growth, U.S. sales of ATVs peaked in 
2006, when more than 1.1 million ATVs were sold.\61\ Sales have 
declined substantially since then. In 2012, less than 320,000 ATVs were 
sold, including those intended for adults, as well as those intended 
for children under the age of 16 years.\62\
---------------------------------------------------------------------------

    \61\ Mathew Camp, ``Nontraditional Quad Sales Hit 465,000,'' 
Dealer News, April 28, 2008. Available at: https://www.dealernews.com/dealernews/article/nontraditional-quad-sales-hit-465000?page=0,0, accessed June 19, 2013.
    \62\ Estimates of ATV sales are based on information provided by 
the Specialty Vehicle Manufacturers Association and on confidential 
data purchased from Power Products Marketing of Minneapolis, MN.
---------------------------------------------------------------------------

    One factor that could account for part of the decline in ATV sales 
is that after many years of increasing sales, the market may be 
saturated. Consequently, a greater proportion of future sales will 
likely be replacement vehicles or vehicles sold due to population 
growth. Another factor could be the increase in sales of ROVs. Some 
riders find that ROVs offer a more comfortable or easier ride, and ROVs 
are more likely to appeal to people who prefer the bench or bucket 
seating on ROVs over the straddle seating of ATVs. It is also easier to 
carry passengers on ROVs. Most ATVs are not intended to carry 
passengers, and the side-by-side seating offered by ROVs appears to be 
preferred over the tandem seating on the few ATVs intended to carry 
passengers.\63\ A disadvantage of an ROV compared to an ATV is that 
many ROVs are too wide to travel on some trail systems intended for 
ATVs. However, some of the more narrow ROVs are capable of negotiating 
many ATV trails.\64\
---------------------------------------------------------------------------

    \63\ ``UTV Sales Flatten Out in 2008,'' Dealer News, August 
2009, p. 40(4). ``2009 Kawasaki Teryx 750 FI 4x4 Sport RUV Test Ride 
Review,'' article posted on: https://www.atvriders.com, accessed 20 
August 2009 and Tom Kaiser, ``Slowing sales: It's now a trend,'' 
Powersports Business, 12 February 2007, p. 44(1).
    \64\ Chris Vogtman, ``Ranger shifts into recreation mode,'' 
Powersports Business, 12 February 2007, p. 46(2).
---------------------------------------------------------------------------

    Of the several types of vehicles that could be substitutes for 
ROVs, go-karts appear to be the smallest market segment. After 
increasing sales for several years, go-kart sales peaked at about 
109,000 vehicles in 2004. Sales of go-karts have since declined 
significantly. In 2013, fewer than 20,000 units were sold. However, 
many of these are aimed at young riders or intended for use on tracks 
or other prepared surfaces and would not be reasonable substitutes for 
ROVs for some purposes.\65\ The decline in go-kart sales may be due to 
the influx of inexpensive ATVs imported from China, which may have led 
some consumers to purchase an ATV rather than a go-kart.\66\
---------------------------------------------------------------------------

    \65\ ``U.S. Go-Kart Market in Serious Decline,'' Dealer News, 
October, 2009, p. 38.
    \66\ (``Karts Feel the Chinese Crunch,'' Dealer News, November 
2007, p. 44(2).
---------------------------------------------------------------------------

C. Societal Costs of Deaths and Injuries Associated With ROVs

    The intent of the proposed rule is to reduce the risk of injury and 
death associated with incidents involving ROVs. Therefore, any benefits 
of the proposed rule could be measured as a

[[Page 69000]]

reduction in the societal costs of injuries and deaths associated with 
ROVs. This section discusses the societal costs of injuries and deaths.
1. ROV Injuries
a. Nonfatal Injuries
    To estimate the number of nonfatal injuries associated with ROVs 
that were treated in hospital emergency departments, CPSC undertook a 
special study to identify cases that involved ROVs that were reported 
through the National Electronic Injury Surveillance System (NEISS) from 
January 1, 2010 to August 31, 2010. NEISS is a stratified national 
probability sample of hospital emergency departments that allows the 
Commission to make national estimates of product-related injuries. The 
sample consists of about 100 of the approximately 5,400 U.S. hospitals 
that have at least six beds and provide 24-hour emergency service.\67\
---------------------------------------------------------------------------

    \67\ Schroeder T, Ault K. The NEISS Sample (Design and 
Implementation): 1999 to Present. Bethesda, MD: U.S. Consumer 
Product Safety Commission; 2001. Available at: https://www.cpsc.gov/neiss/2001d011-6b6.pdf.
---------------------------------------------------------------------------

    NEISS does not contain a separate product code for ROVs. Injuries 
associated with ROVs are usually assigned to either an ATV product code 
(NEISS product codes 3286-3287) or to the utility vehicle category 
(NEISS product code 5044). Therefore, the Commission reviewed all NEISS 
cases that were coded as involving an ATV or a UTV that occurred during 
the first 8 months of 2010 and attempted follow-up interviews with each 
victim (or a relative of the victim) to gather more information about 
the incidents and the vehicles involved. The Commission determined 
whether the vehicle involved was an ROV based on the make and model of 
the vehicle reported in the interviews. If the make and model of the 
vehicle was not reported, the case was not counted as an ROV. Out of 
2,018 NEISS cases involving an ATV or UTV during the study period, a 
total of 668 interviews were completed for a response rate of about 33 
percent. Sixteen of the completed interviews were determined to involve 
an ROV. To estimate the number of ROV-related injuries initially 
treated in an emergency department in 2010, the NEISS weights were 
adjusted to account for both non-response and the fact that the survey 
only covered incidents that occurred during the first 8 months of the 
year. Variances were calculated based on the adjusted weights. Based on 
this work, the Directorate for Epidemiology estimated that there were 
about 3,000 injuries (95 percent confidence interval of 1,100 to 4,900) 
involving ROVs in 2010 that were initially treated in hospital 
emergency departments.\68\
---------------------------------------------------------------------------

    \68\ Sarah Garland, Directorate for Hazard Analysis, ``NEISS 
Injury Estimates for Recreational Off-Highway Vehicles (ROVs),'' 
U.S. Consumer Product Safety Commission (September 2011).
---------------------------------------------------------------------------

    NEISS injury estimates are limited to injuries initially treated in 
hospital emergency departments. NEISS does not provide estimates of the 
number of medically attended injuries that were treated in other 
settings, such as physicians' offices, ambulatory care centers, or 
injury victims who bypassed the emergency departments and were directly 
admitted to a hospital. However, the Injury Cost Model (ICM), developed 
by CPSC for estimating the societal cost of injuries, uses empirical 
relationships between cases initially treated in hospital emergency 
departments and cases initially treated in other medical settings to 
estimate the number of medically attended injuries that were treated 
outside of a hospital emergency department.\69\ According to ICM 
estimates, based on the 16 NEISS cases that were identified in the 2010 
study, injuries treated in hospital emergency departments accounted for 
about 27 percent of all medically treated injuries involving ROVs. 
Using this percentage, the estimate of 3,000 emergency department-
treated injuries involving ROVs suggests that there were about 11,100 
medically treated injuries involving ROVs in 2010 (i.e., 3,000 injuries 
initially treated in emergency departments and 8,100 other medically 
attended injuries) or 194 medically attended injuries per 10,000 ROVs 
in use (11,100 / 570,000 x 10,000).\70\
---------------------------------------------------------------------------

    \69\ For a more complete discussion of the Injury Cost Model see 
Ted R. Miller, et al., The Consumer Product Safety Commission's 
Revised Injury Cost Model, (December 2000). Available at: https://www.cpsc.gov/PageFiles/100269/costmodept1.PDF. https://www.cpsc.gov/PageFiles/100304/costmodept2.PDF.
    \70\ Using the ICM estimates for all cases involving ATVs and 
UTVs, injuries that were initially treated in a hospital emergency 
department accounted for about 35 percent of all medically-attended 
injuries. If this estimated ratio, which is based on a larger 
sample, but that includes vehicles that are not ROVs, was used 
instead of the ratio based strictly on the 16 known ROV NEISS cases 
in 2010, the estimated number of medically-attended injuries would 
be 8,600.
---------------------------------------------------------------------------

b. Fatal Injuries
    In addition to the nonfatal injuries, there are fatal injuries 
involving ROVs each year. As of April 5, 2013, the Commission had 
identified 49 fatalities involving ROVs that occurred in 2010, or about 
0.9 deaths per 10,000 ROVs in use ((49 / 570,000) x 10,000). The actual 
number of deaths in 2010 could be higher because reporting is ongoing 
for 2010. Overall, CPSC has counted 335 ROV deaths that occurred from 
January 1, 2003 to April 5, 2013. There were no reported deaths in 
2003, when relatively few ROVs were in use. As of April 5, 2013, there 
had been 76 deaths reported to CPSC that occurred in 2012.\71\
---------------------------------------------------------------------------

    \71\ Memorandum from Sarah Garland, Division of Hazard Analysis, 
``Additional ROV-related incidents reported from January 1, 2012 
through April 5, 2013,'' U.S. Consumer Product Safety Commission, 
Bethesda, MD (8 April 2013).
---------------------------------------------------------------------------

2. Societal Cost of Injuries and Deaths Associated With ROVs
a. Societal Cost of Nonfatal Injuries
    The CPSC's ICM provides comprehensive estimates of the societal 
costs of nonfatal injuries. The ICM is fully integrated with NEISS and 
provides estimates of the societal costs of injuries reported through 
NEISS. The major aggregated components of the ICM include: Medical 
costs; work losses; and the intangible costs associated with lost 
quality of life or pain and suffering.\72\
---------------------------------------------------------------------------

    \72\ A detailed description of the cost components, and the 
general methodology and data sources used to develop the CPSC's 
Injury Cost Model, can be found in Miller et al. (2000), available 
at https://www.cpsc.gov//PageFiles/100269/costmodept1.PDF and https://www.cpsc.gov//PageFiles/100304/costmodept2.PDF.
---------------------------------------------------------------------------

    Medical costs include three categories of expenditure: (1) Medical 
and hospital costs associated with treating the injury victim during 
the initial recovery period and in the long run, the costs associated 
with corrective surgery, the treatment of chronic injuries, and 
rehabilitation services; (2) ancillary costs, such as costs for 
prescriptions, medical equipment, and ambulance transport; and (3) 
costs of health insurance claims processing. Cost estimates for these 
expenditure categories were derived from a number of national and state 
databases, including the National Healthcare Cost and Utilization 
Project--National Inpatient Sample and the Medical Expenditure Panel 
Survey, both sponsored by the Agency for Healthcare Research and 
Quality.
    Work loss estimates, based on information from the National Health 
Interview Survey and the U.S. Bureau of Labor Statistics, as well as a 
number of published wage studies, include: (1) The forgone earnings of 
parents and visitors, including lost wage work and household work, (2) 
imputed long term work losses of the victim that would be associated 
with permanent impairment, and (3) employer productivity losses, such 
as the costs incurred when employers spend time juggling schedules or 
training replacement workers. The earnings estimates were updated most 
recently with weekly earnings data from the Current

[[Page 69001]]

Population Survey conducted by the Bureau of the Census in conjunction 
with the Bureau of Labor Statistics.
    Intangible, or non-economic, costs of injury reflect the physical 
and emotional trauma of injury as well as the mental anguish of victims 
and caregivers. Intangible costs are difficult to quantify because they 
do not represent products or resources traded in the marketplace. 
Nevertheless, they typically represent the largest component of injury 
cost and need to be accounted for in any benefit-cost analysis 
involving health outcomes.\73\ The Injury Cost Model develops a 
monetary estimate of these intangible costs from jury awards for pain 
and suffering. While these awards can vary widely on a case-by-case 
basis, studies have shown them to be systematically related to a number 
of factors, including economic losses, the type and severity of injury, 
and the age of the victim.\74\ Estimates for the Injury Cost Model were 
derived from a regression analysis of about 2,000 jury awards in 
nonfatal product liability cases involving consumer products compiled 
by Jury Verdicts Research, Inc.
---------------------------------------------------------------------------

    \73\ Rice, D.P. & MacKenzie, E.J. (1989). Cost of injury in the 
United States: A report to Congress, Institute for Health and Aging. 
San Francisco, CA: University of California and The Johns Hopkins 
University.
    \74\ Viscusi, W.K. (1988). Pain and suffering in product 
liability cases: Systematic compensation or capricious awards? Int. 
Rev. Law Econ. 8, 203-220 and Rodgers, G.B. (1993). Estimating jury 
compensation for pain and suffering in product liability cases 
involving nonfatal personal injury. J. For. Econ. 6(3), 251-262.
---------------------------------------------------------------------------

    In addition to estimating the costs of injuries treated in U.S. 
hospital emergency departments and reported through NEISS, the Injury 
Cost Model uses empirical relationships between emergency department 
injuries and those treated in other settings (e.g., physicians' 
offices, clinics, ambulatory surgery centers, and direct hospital 
admissions) to estimate the number, types, and costs of injuries 
treated outside of hospital emergency departments. Thus, the ICM allows 
us to expand on NEISS by combining (1) the number and costs of 
emergency department injuries with (2) the number and costs of 
medically attended injuries treated in other settings to estimate the 
total number of medically attended injuries and their costs across all 
treatment levels.
    In this analysis, we use injury data from 2010, as a baseline from 
which to estimate the societal cost of injuries associated with ROVs. 
We use the year 2010 because 2010 is the year for which we have the 
most comprehensive estimates of both fatal and nonfatal injuries 
associated with ROVs. According to ICM, the average societal cost of a 
medically attended injury associated with ROVs in 2010 was $29,383 in 
2012 dollars. Based on this estimate, the total societal costs of the 
medically attended injuries involving ROVs in 2010 was about $326.2 
million in 2012 dollars (11,100 injuries x $29,383). About 75 percent 
of the cost was related to the pain and suffering. About 9 percent of 
the cost was related to medical treatment, and about 16 percent was 
related to work and productivity losses victim, caregivers, visitors, 
and employers. Less than 1 percent of the cost was associated with the 
costs of the legal and liability system.
    These cost estimates are based on a small sample of only 16 NEISS 
cases. This sample is too small to reflect the full range of injury 
patterns (i.e., the different combinations of injury diagnoses, body 
parts, and injury dispositions) and rider characteristics (i.e., age 
and sex) associated with ROV injuries. In fact, because the 16 NEISS 
cases did not include any case in which the victim required admission 
to a hospital, the cost estimates are probably low. Nevertheless, this 
estimate will be used in this analysis with the knowledge that the 
estimate's use probably leads to an underestimate of the societal costs 
associated with ROVs and underestimates of the potential benefits of 
the proposed rule intended to reduce the risk of injury associated with 
ROVs.\75\
---------------------------------------------------------------------------

    \75\ An alternative method for estimating the injury costs would 
be to assume that the patterns of injury associated with ROVs are 
similar to the injury patterns associated with all ATVs and UTVs. 
According to ICM estimates for all ATVs and UTVs (NEISS Product 
Codes 3285-3287 and 5044), injuries treated in hospital emergency 
departments accounted for about 35 percent of the medically attended 
injuries. This would suggest that the number of medically attended 
injuries involving an ROV was about 8,600. The average cost of a 
medically attended injury involving an ATV or UTV was $42,737. 
Therefore, the total societal cost of medically attended injuries 
would be $367.5 million.
---------------------------------------------------------------------------

b. Societal Cost of Fatal Injuries
    As discussed above, there were at least 49 fatal injuries involving 
ROVs in 2010. If we assign a cost of $8.4 million for each death, then 
the societal costs associated with these deaths would amount to about 
$411.6 million (49 deaths x $8.4 million). The estimate of $8.4 million 
is the estimate of $7.4 million (in 2006 dollars) developed by the U.S. 
Environmental Protection Agency (EPA) updated to 2012 dollars and is 
consistent with willingness-to-pay estimates of the value of a 
statistical life (VSL). According to OMB's 2013 Draft Report to 
Congress on the Benefits and Costs of Federal Regulations and Agency 
Compliance with the Unfunded Mandates Reform Act, willingness-to-pay-
estimates of the VSL generally vary from about $1.3 million to $12.2 
million in 2010 dollars. In 2012 dollars, the range would be $1.3 
million to 13.0 million.\76\
---------------------------------------------------------------------------

    \76\ The estimate of the VSL developed by the EPA is explained 
EPA's Guidelines for Preparing Economic Analysis, Appendix B: 
Mortality Risk Valuation Estimates (Environmental Protection Agency, 
2014) and is available at https://yosemite.epa.gov/ee/epa/eerm.nsf/
vwAN/EE-0568-50.pdf/$file/EE-0568-50.pdf. The OMB's 2013 Draft 
Report to Congress is available at: https://www.whitehouse.gov/sites/default/files/omb/inforeg/2013_cb/draft_2013_cost_benefit_report.pdf. Both reports were accessed on 
August 6, 2014.
---------------------------------------------------------------------------

c. Societal Cost of Injuries per ROV in Use
    Based on the previous discussion, the total estimated societal 
costs of deaths and injuries associated with ROVs were $737.8 million 
in 2010 (expressed in 2012 dollars). The estimate does not include the 
costs associated with any property damage, such as property damage to 
the ROVs involved or other property, such as another vehicle or object 
that might have been involved in an incident.
    Given the earlier estimate that about 570,000 ROVs were in use at 
the end of 2010, the estimated societal costs of deaths and medically 
attended injuries was about $1,294 per ROV in use ($737.8 million / 
570,000) in 2010. However, because the typical ROV is expected to be in 
use for 15 to 20 years, the expected societal cost of fatalities or 
deaths per ROV over the vehicle's useful life is the present value of 
the annual societal costs summed over the ROV's expected useful life. 
CPSC has not estimated the operability rates of ROVs as they age. 
However, CPSC has estimated the operability rates for ATVs as they age, 
based on the results of exposure surveys.\77\ ROVs and ATVs are similar 
vehicles in that they are both off-road recreational vehicles generally 
produced by the same manufacturers. If ROVs have the same operability 
rates as they age as ATVs, the present value of the societal cost of 
injuries over the expected useful life of an ROV (at a 3 percent 
discount rate) is $17,784.\78\
---------------------------------------------------------------------------

    \77\ CPSC Memorandum from Mark S. Levenson, Division of Hazard 
Analysis, to Susan Ahmed, Associate Executive Director, Directorate 
for Epidemiology, ``2001 ATV Operability Rate Analysis,'' U.S. 
Consumer Product Safety Commission, Bethesda MD (19 August 2003).
    \78\ The choice of discount rate is consistent with research 
suggesting that a real rate of 3 percent is an appropriate discount 
rate for interventions involving public health (see Gold, Marthe R, 
Joanna E. Siegel, Louise B. Russell and Milton C. Weinstein, 1996, 
Cost-Effectiveness in Health and Medicine, New York: Oxford 
University Press).

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

[[Page 69002]]

D. Requirements of the Proposed Rule: Costs and Benefits

    The proposed rule would establish a mandatory safety standard for 
ROVs. The requirements of the proposed rule can be divided into two 
general categories: (1) Lateral stability and vehicle handling 
requirements, and (2) occupant-retention requirements. Following is a 
discussion of the costs and benefits that are expected to be associated 
with the requirements of the proposed rule. As discussed earlier, we 
use 2010 as the base year for this analysis because it is the only year 
for which we have estimates of both fatal and nonfatal injuries 
associated with ROVs. However, where quantified, the costs and benefits 
are expressed in 2012 dollars.
    In general, the cost estimates were developed in consultation with 
the Directorate for Engineering Sciences (ES staff). Estimates are 
based on ES staff's interactions with manufacturers and knowledge 
related to ROV design and manufacturing process as well as direct 
experience with testing ROVs and similar products. In many cases, we 
relied on ES staff's expert judgment. Consequently, we note that these 
estimates are preliminary and welcome comments on their accuracy and 
the assumptions underlying their constructions. We are especially 
interested in data that would help us to refine our estimates to more 
accurately reflect the expected costs of the draft proposed rule as 
well as any alternative estimates that interested parties can provide.
1. Lateral Stability and Vehicle Handling Requirements
    The lateral stability and vehicle handling requirements of the 
proposed rule would require that all ROVs meet a minimum level of 
rollover resistance and that ROVs exhibit sub-limit understeer 
characteristics. The dynamic lateral stability requirement would set a 
minimum value for the lateral acceleration at roll-over of 0.70 g (unit 
of standard gravity), as determined by a 30 mph drop-throttle J-turn 
test. The greater the lateral acceleration value, the greater the 
resistance of the ROV is to tipping or rolling over. The understeer 
requirement would mandate that ROVs exhibit understeer characteristics 
in the sublimit range of the turn circle test described in the proposed 
rule.
    The proposed rule would also require manufacturers to place a 
hangtag on all new vehicles that provides the lateral acceleration at 
rollover value for the model and provides information to the consumer 
about how to interpret this value. The intent of the hangtag is to 
provide the potential consumer with information about the rollover 
propensity of the model to aid in the comparison of ROV models before 
purchase. The content and format of the hangtag are described in 
Section IX.C.2.
    The proposed rule describes the test procedures required to measure 
the dynamic rollover resistance and the understeering performance of 
the ROV, including the requirements for the test surface, the loading 
of test vehicles, and the instrumentation required for conducting the 
tests and for data-acquisition during the tests. The test for rollover 
resistance would use a 30 mph drop-throttle J-turn test. This test uses 
a programmable steering controller to turn the test vehicle traveling 
at 30 mph at prescribed steering angles and rates to determine the 
minimum steering angle at which two-wheel lift is observed. The data 
collected during these tests are analyzed to compute and verify the 
lateral acceleration at rollover for the vehicle.
    The test for vehicle handling or understeer performance involves 
driving the vehicle around a 100-foot radius circle at increasing 
speeds, with the driver making every effort to maintain compliance of 
the vehicle path relative to the circle. Data collected during the 
tests are analyzed to determine whether the vehicle understeers through 
the required range. The proposed rule would require that all ROVs 
exhibit understeer for values of ground plane lateral acceleration from 
0.10 to 0.50 g.
a. Cost of Lateral Stability and Vehicle Handling Requirements
    All manufacturers would have to conduct the tests prescribed in the 
proposed rule to determine whether their models meet the requirements 
and to obtain the information on dynamic lateral stability that must be 
reported to consumers on the hangtag. If any model fails to meet one or 
both of the requirements, the manufacturer would have to make 
adjustments or modifications to the design of the model. After the 
model has been modified, the manufacturer would have to conduct tests 
on the modified models to check that the model meets the requirements.
    There is substantial overlap in the conditions under which the 
tests for dynamic lateral stability and vehicle handling must be 
performed. The test surfaces are the same, and the vehicle condition, 
loading, and instrumentation required for both tests are virtually the 
same. The one difference is that the test for dynamic lateral stability 
also requires that the test vehicle be equipped with a programmable 
steering controller. Because there is substantial overlap in the 
conditions under which the tests must be conducted, manufacturers 
likely will conduct both sets of tests on the same day. This would save 
manufacturers the cost of loading and instrumenting the test vehicle 
twice and renting a test facility for more than one day.
    We estimate that the cost of conducting the dynamic lateral 
stability tests and the vehicle handling tests will be about $24,000 
per model.\79\ This includes the cost of conducting both sets of tests, 
measuring the center of gravity of the test vehicle, which is required 
for the dynamic lateral stability test, transporting the test vehicle 
to and from the test site, outfitting the test vehicles with the needed 
equipment and instruments, and the cost of renting the test facility. 
This estimate also assumes that both tests are being conducted on the 
same day and that the manufacturer only needs to rent the test facility 
for one day and pay for loading and instrumenting the test vehicles 
once.
---------------------------------------------------------------------------

    \79\ This estimate is based on the rates that CPSC has most 
recently paid a contractor for conducting these tests. For example, 
see contract CPSC-D-11-0003, which provides the following costs 
estimates: $3,000 for static measurement to determine center of 
gravity location, $19,000 to perform dynamic test, and $2,000 to 
ship vehicles. This amounts to approximately $24,000.
---------------------------------------------------------------------------

    If the model meets the requirements of both tests, the manufacturer 
would have no additional costs associated with these requirements. The 
tests would not have to be conducted again, unless the manufacturer 
makes changes to the model that could affect the vehicle's performance 
in these tests.
    If the model does not meet the requirements of one or both of the 
tests, the manufacturer will incur costs to adjust the vehicle's 
design. Engineers specializing in the design of utility and 
recreational vehicles are likely to have a good understanding of 
vehicle characteristics that influence vehicle stability and handling. 
Therefore, these engineers should be able to modify easily the design 
of a vehicle to meet the stability and handling requirements. The 
Yamaha Rhino repair program demonstrated that an ROV that did not meet 
the lateral stability and vehicle handling requirements was 
successfully modified to meet the requirements by increasing the track 
width and reducing the rear suspension stiffness (by removing the sway 
bar) of the ROV. Based on experience with automotive

[[Page 69003]]

manufacturing, ES staff believes that less than 1 or 2 person-months 
would be required to modify an ROV model that did not comply with the 
requirements. A high estimate would be that a manufacturer might 
require as many as 4 person-months (or about 700 hours) to modify. 
Assuming an hourly rate of $61.75, which is the estimated total hourly 
compensation for management, professional, and related workers, the 
cost to modify the design of an ROV model to meet the stability and 
handling requirements, using the high estimate, would be about $43,000.
    The Commission believes that most modifications that might be 
required to meet the lateral stability and vehicle handling 
requirements will have minimal, if any, impact on the production or 
manufacturing costs because the assembly of an ROV already includes 
installation of a wheel axle and installing a longer wheel axle or 
wheel spacer would not change the current assembly procedure; likewise, 
the assembly of an ROV already includes installation of sway bars and 
shock absorbers and installing different variations of these suspension 
components would not affect the current assembly procedure.
    Once an ROV model has been modified to comply with the 
requirements, the manufacturer will have to retest the vehicle to check 
that the model does comply with the requirements. Both the dynamic 
stability and vehicle handling tests will have to be conducted on the 
redesigned model, even if the original model failed only one of the 
tests. This is because the design changes could have impacted the ROVs 
ability to comply with either requirement. Therefore, the full cost of 
the proposed lateral stability and vehicle handling requirements could 
range from a low of about $24,000 for a model that already met the 
requirements, up to $91,000, for a scenario in which the model was 
tested, the manufacturer required 4 person-months to modify the 
vehicle, and the vehicle was retested to check that the modified 
vehicle complied with the requirements.\80\
---------------------------------------------------------------------------

    \80\ If the ROV already met the lateral stability and vehicle 
handling requirements, the low estimate of $24,000 could overstate 
the incremental cost of meeting the requirements if the manufacturer 
was already performing the tests prescribed in the proposed rule.
---------------------------------------------------------------------------

    Although the plausible range for the cost of the lateral stability 
and vehicle handling requirement is $24,000 to $91,000 per model, the 
Commission believes that the average cost per model will be toward the 
low end of this range because CPSC tested 10 ROVs that represented the 
recreational and utility oriented ROVs available in 2010, and found 
that four out of 10 ROVs met the lateral stability requirement and five 
out of 10 ROVs met the vehicle handling requirements. As discussed 
previously, for models that already meet the requirements, the 
manufacturer will incur no additional costs other than the cost of the 
testing. Based upon CPSC examination of models that do not meet the 
requirements, CPSC believes in most cases the manufacturers should be 
able to bring the model into compliance with the requirements by making 
simple changes to the track width, or to the suspension of the vehicle. 
These are relatively modest modifications that probably can be 
accomplished in less time than the high estimate of 4 months. However, 
the Commission welcomes comments on our underlying rationale for the 
estimates as well as the estimates themselves.
    It is frequently useful to compare the benefits and costs of a rule 
on a per-unit basis. Based on 2011 sales data, the average unit sales 
price per ROV model was about 1,800.\81\ ROVs are a relatively new 
product and the average number of years a ROV model will be produced 
before being redesigned is uncertain. It is often observed that 
automobile models are redesigned every 4 to 6 years. If a ROV model is 
produced for about 5 years before being redesigned, then the cost of 
testing the model for compliance with the dynamic lateral stability and 
vehicle handling requirements, and, if necessary, modifying the design 
of the vehicle to comply with the requirements and retesting the 
vehicle would apply to about 9,000 units. (The Commission welcomes 
comments on this assumption.) Therefore, the average per-unit cost of 
the proposed dynamic lateral stability and vehicle handling 
requirements would be about $3 per unit ($24,000 / 9,000), if the model 
already complies with the requirements. Using the high estimate of the 
time that it could take to modify a model that fails or one or both of 
the tests, the per-unit cost would be about $10 per unit ($91,000 / 
9,000).\82\
---------------------------------------------------------------------------

    \81\ In 2011, the average number of units sold per model was 
about 1,800. Depending on the particular model, the units sold 
ranged from less than 10 for some models, to more than 10,000 for 
others (based on an analysis by CPSC staff of a database obtained 
from Power Products Marketing of Eden Prairie, MN).
    \82\ These per-unit cost estimates are an attempt to estimate 
the average per-unit costs across all ROV models. The actual per-
unit cost for any ROV model would depend upon the sales volume for 
that model. If the sales were substantially more than 1,800 units 
annually, then the per-unit cost would be substantially lower than 
the estimate above. If sales were substantially less than 1,800 
units annually, then the per-unit cost of the proposed requirements 
would be substantially higher.
---------------------------------------------------------------------------

    The proposed rule requires that the manufacturer attach a hangtag 
on each new ROV that provides the ROV's lateral acceleration at 
rollover value, which can be used by the consumer to compare the 
rollover resistance of different ROVs. We estimate that the cost of the 
hangtag, including the designing and printing of the hangtag, and 
attaching the hang tag to the vehicle, will be less than $0.25 per 
vehicle. Our estimates are based on the following assumptions: (1) The 
cost of printing the hang tag and the wire for attaching the hang tag 
is about 8 cents per vehicle, (2) placing the hang tag on each vehicle 
will require about 20 seconds at an hourly rate of $26.11 \83\ and (3) 
designing and laying out the hang tag for each model will require about 
30 minutes at an hourly rate of $61.75.\84\ The estimate of 30 minutes 
for the hang tag design reflects that the proposed rule provides a 
sample of the required hang tag and guidance regarding the layout of 
the hang tag for manufacturers to follow. Also, if the manufacturer has 
multiple models, the same template could be used across models; the 
manufacturer would simply need to change the lateral acceleration 
number and model identification. In light of these considerations, CPSC 
believes that 30 minutes per model represents a reasonable estimate of 
the effort involved, but we welcome comments on this estimate, 
especially comments that will assist us in refining the estimate.
---------------------------------------------------------------------------

    \83\ U.S. Bureau of Labor Statistics, Table 9 (Employer Costs 
for Employee Compensation (ECEC), total compensation for production, 
transportation, and material moving for all workers in private 
industry), June 2012. U.S. Department of Labor. Accessed on January 
9, 2014. Available at: https://www.bls.gov/news.release/archives/ece0c_09112012.pdf
    \84\ U.S. Bureau of Labor Statistics, Table 9 (Employer Costs 
for Employee Compensation (ECEC), total compensation for all 
management, professional, and related for all workers in private 
industry), June 2012. U.S. Department of Labor. Accessed on January 
9, 2014. Available at: https://www.bls.gov/news.release/archives/ecec_09112012.pdf.
---------------------------------------------------------------------------

    According to several ROV manufacturers, some ROV users ``might 
prefer limit oversteer in the off-highway environment.'' This assertion 
appeared in a public comment on the ANPR for ROVs (Docket No. CPSC-
2009-0087), submitted jointly on behalf of Arctic Cat, Inc., Bombardier 
Recreational Products, Inc., Polaris Industries, Inc., and Yamaha Motor 
Corporation, USA. To the extent that the requirements in the proposed 
rule would reduce the ability of these users to reach limit

[[Page 69004]]

oversteer intentionally, the proposed rule could have some adverse 
impact on the utility or enjoyment that these users receive from ROVs. 
These impacts would probably be limited to a small number of 
recreational users who enjoy activities or stunts that involve power 
oversteering or limit oversteer.
    Although the impact on consumers who prefer limit oversteer cannot 
be quantified, the Commission expects that the impact will be low. Any 
impact would be limited to those consumers who wish to engage 
intentionally in activities involving the loss of traction or power 
oversteer. The practice of power oversteer, such as the speed at which 
a user takes a turn, results from driver choice. The proposed rule 
would not prevent ROVs from reaching limit oversteer under all 
conditions; nor would the rule prevent consumers from engaging in these 
activities. At most, the proposed rule might make reaching limit 
oversteer in an ROV to be somewhat more difficult for users to achieve.
b. Benefits of the Lateral Stability and Vehicle Handling Requirements
    The benefit of the dynamic lateral stability and vehicle handling 
or understeer requirements would be the reduction of injuries and 
deaths attributable to these requirements. The intent of the dynamic 
lateral stability requirement is to reduce rollover incidents that 
involve ROVs. A CPSC analysis of 428 ROV incidents showed that at least 
68 percent involved the vehicle rolling sideways. More than half of the 
overturning incidents (or 35 percent of the total incidents) occurred 
during a turn. There were other incidents (24 percent of the total 
incidents) in which the vehicle rolled sideways, but it is not known 
whether the incident occurred during a turn.\85\ The dynamic lateral 
stability requirement is intended to ensure that all ROVs on the market 
have at least a minimum level of resistance to rollover during turns, 
as determined by the test in the proposed rule. Additionally, by 
requiring through the use of hang tags that consumers be informed of 
the rollover resistance of ROV models, the proposed rule would make it 
easier for consumers to compare the rollover resistance of ROV models 
before making a purchase. Manufacturers might be encouraged to develop 
ROV models with greater resistance to rollover if consumers show a 
clear preference for ROVs with the higher values for lateral 
acceleration threshold at rollover when they purchase new ROVs. As a 
similar example, in 2001, NHTSA began including rollover resistance 
information in its new car assessment program (NCAP).\86\ NHTSA 
believed that consumer information on the rollover risk of passenger 
cars would influence consumers to purchase vehicles with a lower 
rollover risk and inspire manufacturers to produce vehicles with a 
lower rollover risk.\87\ A subsequent study of static stability factor 
(SSF) trends in automobiles found that SSF values increased for all 
vehicles after 2001, particularly SUVs, which tended to have the worst 
SSF values in the earlier years.\87\
---------------------------------------------------------------------------

    \85\ Sarah Garland, Ph.D., Analysis of Reported Incidents 
Involving Deaths or Injuries Associated with Recreational Off-
Highway Vehicles (ROVs), U.S. Consumer Product Safety Commission, 
Bethesda, MD (May 2012).
    \86\ 65 FR 34988 (June 1, 2000).
    \87\ Walz, M. C. (2005). Trends in the Static Stability Factor 
of Passenger Cars, Light Trucks, and Vans. DOT HS 809 868. Retrieved 
from https://www.nhtsa.gov/cars/rules/regrev/evaluate/809868/pages/.
---------------------------------------------------------------------------

    The understeer requirement is intended to reduce the likelihood of 
a driver losing control of an ROV during a turn, which can lead to the 
vehicle rollover, striking another vehicle, or striking a fixed object. 
Oversteer is an undesirable trait because it is a directionally 
unstable steering response that leads to dynamic instability and loss 
of control. For this reason, automobiles are designed to exhibit 
understeer characteristics up to the traction limits of the tires. Sub-
limit oversteer is also undesirable for off-highway vehicles due to the 
numerous trip hazards that exist in the off-highway environment and can 
cause the vehicles to roll over.
    Although the Commission believes that the dynamic lateral stability 
and vehicle handling requirements will reduce the number of deaths and 
injuries involving ROVs, it is not possible to quantify this benefit 
because we do not have sufficient data to estimate the injury rates of 
models that already meet the requirements and models that do not meet 
the requirements. Thus, we cannot estimate the potential effectiveness 
of the dynamic lateral stability and vehicle handling requirements in 
preventing injuries. However, these requirements are intended to reduce 
the risk of an ROV rolling sideways when making a turn. Because the 
estimated societal cost of deaths and injuries associated with ROVs is 
$17,784 over the useful life of an ROV, and because at least 35 percent 
of the injuries occurred when an ROV rolled sideways when making a 
turn, these requirements would address approximately $6,224 in societal 
costs per ROV ($17,784 x .35). Consequently, given that the estimated 
cost of the lateral stability and handling requirements is less than 
$10 per ROV, the requirements would have to prevent less than about 0.2 
percent of these incidents ($10 / $6,224) for the benefits of the 
requirements to exceed the costs.
2. Occupant Retention Requirements
    The occupant retention requirements of the proposed rule are 
intended to keep the occupant within the vehicle or within the rollover 
protective structure (ROPs). First, each ROV would be required to have 
a means to restrict occupant egress and excursion in the shoulder/hip 
zone, as defined by the proposed rule. This requirement could be met by 
a fixed barrier or structure on the ROV or by a barrier or structure 
that can be put into place by the occupant using one hand in one 
operation, such as a door. Second, the proposed rule would require that 
the speed of an ROV be limited to a maximum of 15 mph, unless the seat 
belts for both the driver and any front seat passengers are fastened. 
The purpose of these requirements is to prevent deaths and injuries, 
especially incidents involving full or partial ejection of the rider 
from the vehicle.
a. Costs of Occupant Retention Requirements
i. Means To Restrict Occupant Egress or Excursion
    Most ROVs already have some occupant protection barriers or 
structures. In some cases, these structures might already meet the 
requirements of the proposed rule. In other cases, they could be 
modified or repositioned to meet the requirements of the proposed rule. 
A simple barrier that would meet the requirements of the proposed rule 
could be fabricated out of a length of metal tubing that is bent and 
bolted or welded to the ROPs or other suitable structure of the vehicle 
in the shoulder/hip zone of the vehicle, as defined in the proposed 
rule. ES staff believes that any additional metal tubing required to 
form such a barrier could be obtained for a cost of about $2 per 
barrier. ES also believes that the additional time that would be 
required to bolt or weld the barrier to the vehicle would be less than 
1 minute. Assuming an hourly labor cost of $26.11, the labor time 
required would be less than $0.50. ES staff also believes that it would 
take manufacturers only a few hours to determine how an existing ROV 
model would need to be modified to comply with the requirement and to 
make the necessary drawings to implement the change. When spread over 
the

[[Page 69005]]

production of the model, this cost would only amount to a few cents per 
vehicle. Therefore, the estimated cost is expected to be less than $3 
per barrier.
    Based on a cost of less than $3 per barrier, the cost per vehicle 
would be less than $6 for ROVs that do not have rear seats and $12 for 
ROVs with rear seats. One exposure study found that about 20 percent of 
ROVs had a seating capacity of 4 or more, which indicates that these 
ROVs have rear seats. Therefore, if all ROV models required 
modification to meet the standard, the weighted average cost per ROV 
would be about $7 ($6 x 0.8 + $12 x 0.2). However, CPSC tested 10 ROVs 
that represented the recreational and utility oriented ROVs available 
in 2010, and found that four out 10 ROVs had a passive shoulder barrier 
that passed a probe test specified in ANSI/ROHVA 1-2011. Therefore, 
this estimate of the average cost is high because there would be no 
additional cost for models that already meet the proposed requirement. 
We welcome comments on these costs and the assumptions underlying their 
constructions. We are especially interested in data that would help us 
to refine our estimates to more accurately reflect the expected costs 
of this proposed requirement as well as any alternative estimates that 
interested parties can provide.
ii. Requirement To Limit Speed If the Driver's Seat Belt Is Not 
Fastened
    The requirement that the speed of the vehicle be limited if the 
driver's seat belt is unfastened does not mandate any specific 
technology. Therefore, manufacturers would have some flexibility in 
implementing this requirement. Nevertheless, based on staff's 
examination of and experience with speed-limiting technology, including 
examination of current ROV models with this feature, most systems to 
meet this requirement will probably include the following components:
    1. A seat belt use sensor in the seat belt latch, which detects 
when the seat belt is fastened;
    2. a means to limit the speed of the vehicle when the seat belt is 
not fastened;
    3. a means to provide a visual signal to the driver of the vehicle 
when the speed of the vehicle is limited because the seat belt is not 
fastened;
    4. wiring or other means for the sensor in the seat belt latch to 
send signals to the vehicle components used to limit the speed of the 
vehicle and provide feedback to the driver.
    Before implementing any changes to their vehicles to meet the 
requirement, manufacturers would have to analyze their options for 
meeting the requirement. This process would include developing 
prototypes of system designs, testing the prototypes, and refining the 
design of the systems based on this testing. Once the manufacturer has 
settled upon a system for meeting the requirement, the system will have 
to be incorporated into the manufacturing process of the vehicle. This 
will involve producing the engineering specifications and drawings of 
the system, parts, assemblies, and subassemblies that are required. 
Manufacturers will need to obtain the needed parts from their suppliers 
and incorporate the steps needed to install the system on the vehicles 
in the assembly line.
    ES staff believes that it will take about nine person-months per 
ROV model to design, test, implement, and begin manufacturing vehicles 
that meet the requirements. The total compensation for management, 
professional, and related occupations as of 2012, is about $61.75 per 
hour.\88\ Therefore, if designing and implementing a system to meet the 
requirement entails about nine person months (or 1,560 hours), the cost 
to the company would be about $100,000 per ROV model.\89\
---------------------------------------------------------------------------

    \88\ U.S. Bureau of Labor Statistics, Table 9 (Employer Costs 
for Employee Compensation (ECEC), total compensation for all 
management, professional, and related for all workers in private 
industry), June 2012. U.S. Department of Labor. Accessed on January 
9, 2014. Available at: https://www.bls.gov/news.release/archives/ecec_09112012.pdf.
    \89\ The estimate has been rounded to the nearest $10,000.
---------------------------------------------------------------------------

    Manufacturers would be expected to perform certification tests, 
following the procedure described in the proposed rule, at least once 
for each model the manufacturer produces, to ensure that the model, as 
manufactured, meets the rule's requirements. Additionally, 
manufacturers would be expected to perform the certification testing 
again if they make any changes to the design or components used in a 
vehicle that could impact the ROV's compliance with this requirement. 
We estimate that the cost of this testing would be about $4,000 per 
model. This estimate assumes that the testing will require three 
professional employees 4 hours to conduct the testing at $61.75 per 
hour, per person. Additionally, the rental of the test facility will 
cost $1,000; rental of the radar gun will cost $400; and transportation 
to the test facility will cost $1,400, and that the test vehicle can be 
sold after the testing is completed.
    In addition to the cost of developing and implementing the system, 
manufacturers will incur costs to acquire any parts required for the 
system and to install the parts on the vehicles. We estimate the cost 
of adding a seat belt-use sensor to detect when the seat belt is 
fastened to be about $7 per seat belt. This estimate is based on 
figures used by the National Highway Traffic Safety Administration 
(NHTSA) in its preliminary economic assessment of an advanced air bag 
rule.\90\ This is a widely used technology; virtually all passenger 
cars have such sensors in their driver side seat belt latches to signal 
the seat belt reminder system in the car. The sensors and seat belt 
latches that would be expected to be used to meet this requirement in 
ROVs are virtually the same as the sensors used in passenger cars.
---------------------------------------------------------------------------

    \90\ NHTSA estimated the cost of a seat belt use sensor to be $2 
to $5 in 1997 dollars. The cost has been adjusted to 2012 dollars 
using the CPI Inflation Calculator at: https://www.bls.gov/data/inflation_calculator.htm.
---------------------------------------------------------------------------

    There is more than one method manufacturers could use to limit the 
maximum speed of the vehicle when the driver's seat belt is unfastened. 
One method would be to use a device, such as a solenoid, that limits 
mechanically the throttle opening. Based on observed retail prices for 
solenoid valves used in automotive applications, the cost to 
manufacturers of such a solenoid should be no more than about $25 per 
vehicle. One retailer had 24 different solenoids available at retail 
prices ranging from about $24 to $102. We expect that a manufacturer 
would be able to obtain similar solenoids for substantially less than 
the retail price. Thus, using the low end of the observed retail prices 
suggests that manufacturers would probably be able to acquire 
acceptable solenoids for about $25 each.
    Manufacturers of ROVs equipped with electronic throttle control 
(ETC or ``throttle by wire'') would have at least one other option for 
limiting the maximum speed of the vehicle. Instead of using a 
mechanical means to limit the throttle opening, the engine control unit 
(ECU) of the vehicle, which controls the throttle, could be 
reprogrammed or ``mapped'' in a way that would limit the speed of the 
vehicle if the seat belt was not fastened. If the ECU can be used to 
limit the maximum speed of the ROV, the only cost would be the cost of 
reprogramming or mapping the ECU, which would be completed in the 
implementation stage of development, discussed previously. There would 
be no additional manufacturing costs involved.
    There would be at least two options for providing a visual signal 
to the driver that the speed of the vehicle is limited because seat 
belts are not

[[Page 69006]]

fastened. One option would be to use an LCD display. Most ROV models 
already have an LCD display in the dashboard that could be used for 
this purpose. If an LCD display is present, the only cost would be the 
cost of the programming required for the display to show this message. 
This cost would be included in the estimated cost of the research and 
development, and there would be no additional manufacturing cost.
    Another option for providing a visual signal to the driver that the 
speed of the vehicle is limited would be to use a lighted message or 
icon on the dashboard or control panel of the vehicle. Both voluntary 
standards already require a ``lighted seat belt reminder.'' To comply 
with this proposed requirement, the current visual reminder would have 
to be modified. For example, the wording or icons of the reminder would 
change, and the reminder would probably require a somewhat larger area 
on the dashboard or control panel. There could be some additional cost 
for an extra bulb or lamp to illuminate the larger area or icon. Based 
on its experience, ES staff believes that the cost of an additional 
bulb or lamp would be about $1 or less per vehicle.
    There will be some labor costs involved in installing the 
components needed to meet this requirement, including installing and 
connecting the wires. We expect that the components would be installed 
at the stage of assembly that would minimize the amount of labor 
required. If the amount of additional labor per vehicle was about 5 
minutes, and assuming a total labor compensation rate of $26.11 an 
hour,\91\ the labor cost is estimated to amount to approximately $2 per 
vehicle.
---------------------------------------------------------------------------

    \91\ U.S. Bureau of Labor Statistics, Table 9 (Employer Costs 
for Employee Compensation (ECEC), total compensation for production, 
transportation, and material moving for all workers in private 
industry), June 2012. U.S. Department of Labor. Accessed on January 
9, 2014. Available at: https://www.bls.gov/news.release/archives/ecec_09112012.pdf
---------------------------------------------------------------------------

    In addition to the certification testing discussed previously, most 
manufacturers would be expected to conduct some quality assurance 
testing on vehicles as the vehicles come off the assembly line. 
Virtually all manufacturers already perform some quality control or 
quality assurance tests on their vehicles. The tests are intended to 
ensure, among other things, that the vehicle starts properly, that the 
throttle and brakes function properly, and that any lights function 
properly. Testing of the system limiting the maximum speed when the 
driver's seat belt is not fastened would likely be incorporated into 
this testing to ensure that the system is working as intended. These 
tests could simply involve running the vehicle once with the seat belt 
unfastened to determine whether speed was limited and running the 
vehicle again with the seat belt fastened to determine whether the 
maximum speed was no longer limited. If this testing added an 
additional 10 minutes to the amount of time it takes to test each 
vehicle, the cost would be about $4 per vehicle, assuming a total 
hourly compensation rate of $26.11.
    The manufacturing costs that would be associated with meeting the 
seat belt reminder and speed limitation requirement of the proposed 
rule are summarized in Table 8. These costs include the cost of one 
seat belt-use sensor, the throttle or engine control, the visual 
feedback to the driver, and about 5 minutes of labor time and about 10 
minutes for testing.

     Table 8--Estimated Manufacturing Costs of Requirement, per ROV
------------------------------------------------------------------------
               Component                               Cost
------------------------------------------------------------------------
Seat Belt-Use Sensor...................  $7.
Throttle or Engine Control.............  $0 to $25.
Visual Signal to Driver................  $1.
Labor..................................  $2.
Quality Control Testing................  $4.
                                        --------------------------------
  Total................................  $14 to $39.
------------------------------------------------------------------------

    As discussed previously, we estimate the upfront research, design, 
and implementation costs to be about $100,000 per model, and the 
certification testing costs are estimated to be about $4,000 per model. 
Assuming, as before, that the average annual sales per model are 1,800 
units, and assuming that the typical model is produced for 5 years, 
then the research, design, and certification testing costs would 
average about $12 per vehicle. The average cost for models produced at 
lower volumes would be higher, and the average cost for models produced 
at higher-than-average volumes would be lower. Given the average cost 
of the design and development and the costs of the parts and 
manufacturing, we estimate that this requirement would cost between $26 
($14 + $12) and $51 ($39 + 12) per vehicle.
    Unquantifiable Costs to Users--The requirement could impose some 
unquantifiable costs on certain users who would prefer not to use seat 
belts. The cost to these users would be the time required to buckle and 
unbuckle their seat belts and any disutility cost, such as discomfort 
caused by wearing the seat belt. We cannot quantify these costs because 
we do not know how many ROV users choose not to wear their seat belts. 
Nor do we have the ability to quantify any discomfort or disutility 
that ROV users would experience from wearing seat belts. However, the 
proposed rule does not require that the seat belts be fastened, unless 
the vehicle is traveling 15 mph or faster. This requirement should 
serve to mitigate these costs because many people who would be 
inconvenienced or discomforted by the requirement, such as people using 
the vehicle for work or utility purposes, or people who must get on and 
off the vehicle frequently, are likely to be traveling at lower speeds.
iii. Requirement To Limit Speed If Seat Belts for Front Passengers Are 
Not Fastened
    The proposed rule would also require that the speed of the ROV be 
limited to no more than 15 mph if the seat belt of any front passenger, 
who is seated in a location intended by the manufacturer as a seat, is 
not fastened. Based on conversations with ES staff, designing a system 
that also limits the speed of the vehicle if the seat belt of a 
passenger is not fastened would require only minor adjustments to the 
system limiting the speed if the driver's seat belt is not fastened. 
The speed-limiting system uses sensor switches (seat belt latch sensors 
and/or occupant presence sensors) to determine if seat belts are in 
use, and the speed-limiting system controls the vehicle's speed based 
on whether the switch is activated or not. ES staff believes adding 
requirements for front passenger seat belt use will not add significant 
time to the research and design effort for a speed-limitation system 
because the system would only have to incorporate additional switches 
to the side of the system that determines whether vehicle speed should 
be limited.
    However, incorporating the front passenger seats into the 
requirement would require additional switches or sensors. A seat belt-
use sensor like the one used on the driver's side seat belt latch, 
would be required for each passenger seat belt. The cost of a seat 
belt-use sensor was estimated to be about $7. Additionally, there would 
likely be a sensor switch in each front passenger seat to detect the 
presence of a passenger. This switch could be similar to the seat 
switches in riding lawn mowers that shut off the engine if a rider is 
not detected. Similarly, in a ROV, if the presence of a passenger is 
not detected, the switch would not include the passenger seat belt 
sensor in circuit for determining whether the speed of the ROV should 
be limited. We

[[Page 69007]]

estimate that the cost of this switch is $13 per seat, based on the 
retail price of a replacement switch for the seat switch in a riding 
lawn mower.
    There will be labor costs involved in installing the components 
needed to meet this requirement. The components would probably be 
installed at the stage of assembly that would minimize the amount of 
labor required and would probably not require more than about 5 
minutes. Additionally, manufacturers will need to conduct tests of the 
system to ensure that the system functions as required. These tests 
could take an additional 5 minutes per vehicle. Assuming a total labor 
compensation rate of $26.11 an hour,\92\ the labor cost would probably 
amount to about $4 per vehicle. Therefore, the full cost of meeting 
this requirement would be about $24 per passenger seat ($7 for seat 
belt latch sensor + $13 for seat switch + $4 for labor). Therefore, the 
quantifiable cost of extending the seat belt/speed limitation 
requirement to include the front passenger seat belts would be $24 for 
ROVs with only two seating positions in the front, (i.e., the driver 
and right front passenger) and $48 for ROVs that have three seating 
positions in the front. According to a survey by Heiden Associates, 
about 9 percent of ROVs were reported to have a seating capacity of 
three.\93\ Therefore, the average cost of extending the seat belt/speed 
limitation requirement per ROV would be $26 ($24 + 0.09 x $24).
---------------------------------------------------------------------------

    \92\ U.S. Bureau of Labor Statistics, Table 9 (Employer Costs 
for Employee Compensation (ECEC), total compensation for production, 
transportation, and material moving for all workers in private 
industry), June 2012. U.S. Department of Labor. Accessed on January 
9, 2014. Available at: https://www.bls.gov/news.release/archives/ecec_09112012.pdf.
    \93\ Heiden Associates et al. provided results from a 2009 ROV 
Survey, which is included in Appendix 2 of Docket No. CPSC--2009-
0087).
---------------------------------------------------------------------------

    An additional cost that is unquantifiable but should be considered 
nevertheless, is the impact that the failure of a component of the 
system could have on consumers. The more components that a system has, 
or the more complicated that a system is, the more likely it is that 
there will be a failure of a component somewhere in the system. A 
system that limits the speed of an ROV if a front passenger's seat belt 
is unbuckled would consist of more components and the system would be 
more complicated than a system that only limited the speed of the 
vehicle if the driver's seat belt is unfastened. Failure in one or more 
of the components would impose some costs on the consumer, and this 
failure could possibly affect consumer acceptance of the requirement. 
For example, if the sensor in a passenger's seat belt failed to detect 
that the seat belt was latched, the speed of the vehicle could be 
limited, even though the seat belts were fastened. The consumer would 
incur the costs of repairing the vehicle and the loss in utility 
because the speed was limited until the repairs were made.
b. Benefits of the Occupant Retention Requirements
    The benefit of the occupant-retention requirement is the reduction 
in the societal cost of fatal and nonfatal injuries that could be 
attributable to the requirements. In passenger cars, NHTSA assumes that 
a belted driver has a 45 percent reduction in the risk of death.\94\ 
Research confirms the validity of that estimate.\95\ The effectiveness 
of seat belts in reducing the number or severity of nonfatal injuries 
is less certain than in the cases resulting in deaths. Nevertheless, 
there is evidence that the use of seat belts is associated with a 
reduction in injury severity. A study by Robert Rutledge and others 
found statistically significant decreases in the severity of injuries 
in belted patients versus unbelted patients admitted to trauma center 
hospitals in North Carolina for variables such as the trauma scores, 
the Glasgow coma scale, days on a ventilator, days in an intensive care 
unit, days in a hospital, and hospital charges.\96\ This study found, 
for example, that the mean stay in the hospital for belted patients was 
about 20 percent shorter than for unbelted patients: 10.5 days for 
belted patients as opposed to 13.2 days for unbelted patients. The 
hospital charges for belted patients were 31 percent less than the 
charges incurred by unbelted patients: $10,500 versus $15,250.\97\
---------------------------------------------------------------------------

    \94\ Charles J. Kahane, ``Fatality Reduction by Safety Belts for 
Front-Seat Occupants of Cars and Light Trucks: Updated and Expanded 
Estimates Based on 1986-99 FARS Data,'' U.S. Department of 
Transportation, Report No. DOT HS 809 199, (December 2000).
    \95\ ``Analysis of Reported Incidents Involving Deaths or 
Injuries Associated with Recreational Off-Highway Vehicles (ROVs),'' 
U.S. Consumer Product Safety Commission, Bethesda, MD (May 2012).
    \96\ Robert Rutledge, Allen Lalor, Dale Oller, et al., ``The 
Cost of Not Wearing Seat Belts: A Comparison of Outcome in 3396 
Patients,'' Annals of Surgery, Vol. 217, No. 2, 122-127 (1993).
    \97\ Note that the Rutledge study looked only at the difference 
in the severity of cases involving belted, as opposed to unbelted 
victims. It did not estimate the number of injuries that were 
actually prevented. It should also be noted that the Rutledge study 
focused only on patients that were hospitalized for at least one 
day. It might not be as applicable to patients who were treated and 
released without being admitted to a hospital.
---------------------------------------------------------------------------

    In this analysis, we assume that the effectiveness estimate that 
NHTSA uses for seat belts in automobiles is a reasonable approximation 
of the effectiveness of seat belts at reducing fatalities in ROVs. 
However, according to Kahane (2000), the effectiveness of seat belts 
was significantly higher in accidents involving rollover and other 
incidents where the potential for ejection was high.\98\ A significant 
portion of the fatal and nonfatal injuries associated with ROVs are 
associated with rollovers, which suggests that a higher effectiveness 
estimate could be warranted.
---------------------------------------------------------------------------

    \98\ In these incidents, the researchers found the effectiveness 
of seat belts was 74 percent in passenger cars and 80 percent in 
light trucks. Incidents involving overturning of the vehicle or the 
ejection of the victim are associated with a larger proportion of 
the fatal injuries involving ROVs. At least 65 percent of the 
fatalities were in incidents where the vehicle rolled sideways and 
at least 70 percent of those injured or killed were either fully or 
partially ejected.
---------------------------------------------------------------------------

    The work by Rutledge, et al., showed that mean hospital stays were 
about 20 percent less and hospital charges were 31 percent less for 
belted patients. This work provides some evidence that seat belts can 
reduce some components of the societal costs of nonfatal injuries by 20 
to 31 percent. In this analysis we use the low end of this range, 20 
percent, and assume that it applies to all components of the societal 
costs associated with nonfatal ROV injuries, including work losses and 
pain and suffering. The assumed 20 percent reduction in societal costs 
could come about because some injuries were prevented entirely or 
because the severity of some injuries was reduced.
    These assumptions are justified because the seat belts used in ROVs 
are the same type of seat belts used in automobiles. Additionally, the 
requirement that ROVs have a passive means to restrict the egress or 
excursion of an occupant in the event of a rollover would ensure that 
there would be some passive features on ROVs that will help to retain 
occupants within the protective structure of the ROV just as there are 
in automobiles. We welcome comment on the accuracy of these estimates 
and underlying assumptions and will consider alternative estimates or 
assumptions that commenters wish to provide.
    A separate estimate of the benefit of the requirement for a passive 
means to restrict occupant egress or excursion is not calculated. The 
primary benefit of this requirement is to ensure that ROVs have passive 
features that are more effective at retaining occupants within the 
protective zone of the vehicle in the event of a rollover. Therefore, 
the passive means to restrict occupant egress or excursion acts 
synergistically with the seat belt requirements to keep occupants 
within the protective zone of

[[Page 69008]]

the vehicle or ROPS, and in addition, provides justification for 
applying to the proposed rule for ROVs estimates from studies on the 
effectiveness of seat belts in automobiles.
i. Benefit of Limiting Speed If Driver's Seat Belt Is Not Fastened
    As noted previously, the benefit of the occupant-retention 
requirements would be the reduction in the societal costs of fatal and 
nonfatal injuries that would be expected. The incremental benefit of 
applying the requirement to limit the speed of the vehicle if the 
driver's seat belt is not fastened is discussed below. The incremental 
benefit of applying the same requirement to the front passengers is 
discussed separately.
Potential Reduction in Fatal Injuries
    Table 9 shows the 231 fatality cases that CPSC has reviewed 
according to the seating location of the victim and whether the victim 
was wearing a seat belt. Ignoring the cases in which the location of 
the victim or the seat belt use by the victim is unknown (and thereby, 
erring on the side of underestimating the benefits), the data show that 
about 40 percent (92 / 231) of the deaths happened to drivers who were 
not wearing seat belts. If the pattern of deaths in 2010 is presumed to 
match the overall pattern of the deaths reviewed by CPSC, then about 20 
of the reported 49 deaths associated with ROVs in 2010 \99\ would have 
been to drivers who did not have their seat belts fastened. (The actual 
pattern of deaths in any given year will likely be higher or lower than 
the overall or average pattern. In this analysis, we imposed the 
overall pattern to the reported fatalities in 2010, so that the results 
would be more representative of all reported ROV fatalities.)
---------------------------------------------------------------------------

    \99\ The collection of fatalities associated with ROVs in 2010 
was ongoing at the time this analysis was conducted. The actual 
number of deaths associated with ROVs in 2010 could be higher.

                          Table 9--ROV Fatalities by Victim Location and Seat Belt Use
                                               [2003 through 2011]
----------------------------------------------------------------------------------------------------------------
                                                                                 Seat belt use
                                                             ---------------------------------------------------
                          Location                                                       Unknown or
                                                                  Yes           No          N/A         Total
----------------------------------------------------------------------------------------------------------------
Driver......................................................           16           92           33          141
Right Front Passenger.......................................           10           33            6           49
Middle Front Passenger......................................            0            6            0            6
Rear Passenger..............................................            0            3            1            4
Unknown Location............................................            1            6            5           12
Cargo Area..................................................            1            8            1           10
Bystander or Other..........................................            0            3            6            9
                                                             ---------------------------------------------------
    Total...................................................           28          150           53          231
----------------------------------------------------------------------------------------------------------------
Source: CPSC Directorate for Epidemiology.

    The requirement limiting the maximum speed would apply only to 
incidents involving unbelted drivers that occurred at speeds of greater 
than 15 mph. Of the ROV incidents that the Commission has reviewed, the 
speed of the vehicle was reported for only 89 of the 428 incidents. 
Therefore, estimates based on this data need to be used cautiously. 
Nevertheless, for victims who are known to have been injured and for 
which both their the seat belt use and the speed of the vehicle are 
known, about 73 percent of the unbelted victims were traveling at 
speeds greater than 15 mph. (Victims who were involved in an ROV 
incident but were not injured, or whose injury status is not known, 
were not included in this analysis.) Consequently, if we assume that 73 
percent of the fatalities occurred to unbelted drivers who were 
traveling at speeds greater than 15 mph, then about 15 (20 x 0.73) of 
the fatalities in 2010 would have been addressed, although not 
necessarily prevented, by the proposed requirement.
    As discussed previously, in passenger cars, NHTSA assumes that a 
belted driver has a 45 percent reduction in the risk of death. If seat 
belts have the same effectiveness in reducing the risk of death in 
ROVs, the seat belt/speed limitation requirement would have reduced the 
number of fatal injuries to drivers of ROVs by about 7 (15 x 0.45) in 
2010, if all ROVs in use at the time had met this requirement.\100\ 
This represents an annual risk reduction of 0.0000123 deaths per ROV in 
use (7 / 570,000).
---------------------------------------------------------------------------

    \100\ Alternatively, the drivers could opt to leave their seat 
belts unfastened and accept the lower speed. Because the risk of 
having an accident is probably directly related to the speed of the 
vehicle, this option would also be expected to reduce the number of 
fatal injuries.
---------------------------------------------------------------------------

    As discussed previously, in this analysis, we assume a value of 
$8.4 million for each fatality averted. However, in this analysis, we 
assume that each fatal injury prevented by the use of seat belts still 
resulted in a serious, but nonfatal, injury. The average societal cost 
of a hospitalized injury involving all ATVs and UTVs in 2010 was about 
$350,000 in 2012 dollars. (Based on the ICM estimates of the cost of a 
hospitalized injury using NEISS Product Codes 3285, 3286, 3287, and 
5044.) Subtracting this from the assumed societal cost of $8.4 million 
per death results in a societal cost reduction of $8.05 million per 
death averted. Thus, a reduction in societal costs of fatal injuries of 
about $99 per ROV in use (0.0000123 x $8.05 million) per year could be 
attributable to the seat belt/speed limitation requirement.
Potential Reduction in Societal Cost of Nonfatal Injuries
    As discussed previously, for this analysis, we assumed that the 
seat belt/speed limitation requirement will reduce the societal cost of 
nonfatal ROV injuries by 20 percent. The assumed 20 percent reduction 
in societal costs could result because some injuries were prevented 
entirely, or because the severity of some injuries was reduced. The 
CPSC has investigated several hundred nonfatal injuries associated with 
ROVs. Table 10 summarizes the nonfatal injuries according to seating 
location and seat belt use. (Cases in which the occupant was not 
injured, or cases in which it is unknown whether the occupant was 
injured, were not included in this analysis.) Again, ignoring the cases 
in which the location of the victim or the seat belt use by the victim 
is unknown (and thereby, erring

[[Page 69009]]

on the side of underestimating the benefits), the data indicate that 
about 12 percent (46 / 388) of the nonfatal injuries happened to 
drivers who were not wearing seat belts. This suggests that 1,332 
(11,100 x 0.12) of the approximately 11,100 medically attended injuries 
in 2010 would have involved unbelted drivers. Assuming, as with the 
fatal injuries, that 73 percent were traveling at a speed greater than 
15 mph at the time of incident, 972 (1,332 x 0.73) of the injuries in 
2010 could have been addressed by the proposed seat belt/speed 
limitation requirement. These 972 injuries in 2010 represent an injury 
rate of about 0.00170526 (972 / 570,000) per ROV in use.

                      Table 10--Nonfatal ROV Injuries by Victim Location and Seat Belt Use
                                                 [2003 to 2011]
----------------------------------------------------------------------------------------------------------------
                                                                                 Seat belt use
                                                             ---------------------------------------------------
                     Location of victim                                                  Unknown or
                                                                  Yes           No          N/A         Total
----------------------------------------------------------------------------------------------------------------
Driver......................................................           23           46           51          120
Right Front Passenger.......................................           28           35            9           72
Middle Front Passenger......................................            0           14            1           15
Rear Passenger..............................................            2            3            0            5
Unknown Location............................................            8           21          128          157
Cargo Area..................................................            3           13            0           16
Bystander...................................................            0            0            3            3
                                                             ---------------------------------------------------
    Total...................................................           64          132          192         388
----------------------------------------------------------------------------------------------------------------
Source: CPSC Directorate for Epidemiology.

    Based on estimates from the CPSC's ICM, the average societal cost 
of the injuries addressed is estimated to be $29,383. Applying this 
cost estimate to the estimated injuries per ROV that could be addressed 
by the standard results in an annual societal cost of about $50 per ROV 
in use (0.00170526 x $29,383). If wearing seat belts could have reduced 
this cost by 20 percent (by reducing either the number or severity of 
injuries), the societal benefit, in terms of the reduced costs 
associated with nonfatal injuries, would be about $10 per ROV in use.
Total Benefit Over the Useful Life of an ROV
    The total benefit of the seat belt/speed limitation requirement per 
ROV would be the present value of the expected annual benefit per ROV 
in use, summed over the vehicle's expected useful life. Above, using 
2010 as the base year, we estimated that the annual benefit per ROV was 
about $99 in terms of reduced deaths and $10 in terms of reduced 
nonfatal injuries, for a total of $109 per ROV. Assuming that ROVs have 
the same operability rates as ATVs, the present value of the estimated 
benefit over the useful life of an ROV would be approximately $1,498 
per vehicle, at a 3 percent discount rate.
    The cost of the requirement to limit the speed of the vehicle if 
the driver's seat belt is not fastened was estimated to be between $26 
and $51 per vehicle. Additionally, the cost of the requirement for a 
means to restrict occupant egress and excursion via a passive method 
was estimated to be about $7 per vehicle. Therefore, the total cost 
would be between $33 and $58 per vehicle. The benefit of the 
requirement, estimated to be about $1,498 per vehicle, is substantially 
greater than the estimated cost of the requirement.
ii. Benefit of Limiting Speed If a Front Passenger's Seat Belt Is Not 
Fastened
    The potential incremental benefit of limiting the speed of an ROV 
if a front passenger's seat belt is not fastened can be calculated 
following the same procedure used to calculate the benefits of a 
requirement limiting the maximum speed when the driver's seat belt is 
not fastened. From the data presented in Table 9 (and ignoring the 
cases in which the seating location of the victim or the seat belt use 
is unknown), there were 33 victims seated in the right front passenger 
position, and six who were seated in the middle front passenger 
position were not using a seatbelt. However, some of the victims listed 
as a middle front seat passenger were not seated in places intended to 
be a seat. In some cases, the victim might have been seated on a 
console; in other cases, the victim might have been sharing the right 
front passenger seat and not a separate seat. Based on the information 
available about the incidents, we believe that only three of the six 
victims reported to be ``middle front passengers,'' were actually in 
positions intended by the manufacturer to be middle seats. Therefore, 
about 16 percent (36 / 231) of the fatal injuries involved front seat 
passengers who were not wearing seat belts.
    Applying this estimate to the fatalities in 2010 suggests that 
about 8 of the 49 fatalities happened to front passengers who were not 
wearing seat belts. Assuming that about 73 percent of the incidents 
involved vehicles traveling faster than 15 mph, about 6 of the 
fatalities would have been addressed, but not necessarily prevented, by 
the requirement. Assuming that seat belts reduce the risk of fatal 
injuries by 45 percent, about 3 fatalities might have been averted. 
This represents a risk reduction of 0.00000526 deaths per ROV in use (3 
/ 570,000). Assuming a societal benefit of $8.05 million for each death 
averted results in an estimated annual benefit of about $42 per ROV in 
use ($8.05 million x 0.00000526) in reduced fatal injuries.
    Similarly, the data show that 35 of the victims who suffered 
nonfatal injuries were seated in the right front passenger location, 
and 14 were seated in the middle front position. However, we believe 
that only 8 of the 14 were actually seated in a position intended by 
the manufacturer to be a seat. Therefore, 43 of the 388 victims (or 
about 11 percent of the total) with nonfatal injuries were front 
passengers who were not wearing seat belts. This suggests that 1,221 of 
the estimated 11,100 medically attended injuries in 2010 involved 
unbelted front passengers. Using the assumption that 73 percent of 
these incidents occurred at speeds greater than 15 mph, then about 891 
of the injuries might have been addressed by the requirement, or about 
0.00156315 injuries per ROV in use (891 / 570,000). Assuming that the 
average cost of a nonfatal injury involving ROVs is $29383, the 
estimated societal cost of these injuries is about $46 per ROV in use. 
If wearing seat belts could have

[[Page 69010]]

reduced the societal cost of the nonfatal injuries by 20 percent, then 
the benefits of the requirement would have been about $9 per ROV in 
use, per year.
    Combining the benefits of the reduction in the societal cost of 
deaths ($42 per ROV in use) and the societal cost of injuries ($9 per 
ROV in use) yields an estimated benefit of $51 per ROV in use. Assuming 
that ROVs have the same operability rates as ATVs over time, and 
assuming a discount rate of 3 percent, the estimated benefit would be 
$701 over the expected useful life of an ROV. This is greater than the 
expected cost of this potential requirement of $26 per vehicle.
iii. Impact of Any Correlation in Seat Belt Use Between Driver and 
Passengers
    The analysis above used a simplifying assumption that the use of 
seat belts by the passenger is independent of the use of seat belts by 
the driver. Therefore, we assumed that limiting the maximum speed of 
the ROV if the driver's seat belt was not fastened would have no impact 
on the seat belt use by any passenger. However, there is some evidence 
that the use of seat belts by passengers is correlated with the seat 
belt use of the driver. In the incidents examined by the Commission, of 
the 121 right front passengers with known seat belt usage, the driver 
and right passenger had the same seat belt use status most of the time 
(about 82 percent). In other words, most of the time, the driver's and 
right passenger's seat belts were either both fastened or both 
unfastened. This suggests that if the drivers were required to fasten 
his or her seat belt, at least some of the passengers would also fasten 
their seat belts.
    The implication that a correlation exists between seat belt use by 
drivers and by passengers indicates that the benefits of requiring the 
driver's seat belt to be fastened were underestimated and the benefits 
of extending the requirement to include the right front passenger are 
over estimated. For example, if 80 percent of the passengers who would 
not normally wear their seat belts were to wear their seat belts 
because the driver was required to wear his or her seat belt (for the 
ROV to exceed 15 mph), then 80 percent of the benefit, or $561 ($701 x 
0.80) attributed above to extending the speed limitation requirement to 
the front passengers would be attributed rightfully to the requirement 
that the driver's seat belt be fastened; and only 20 percent, or $140 
($701 x 0.20) would be attributable to the requirement that the front 
passengers' seat belts be fastened. In this example, the $140 in 
benefits attributed to extending the speed limitation requirement to 
include the front passenger's seat belts would still exceed the 
quantifiable cost of doing so, which was estimated to be $26.

E. Summary of the Costs and Benefits of the Proposed Rule

    As described previously, manufacturers would incur costs of 
$128,000 to $195,000 per model to test ROV models for compliance with 
the requirements of the proposed rule and to research, develop, and 
implement any needed changes to the models so that they would comply 
with the requirements. These costs would be incurred before the model 
is brought to market. To express these costs on a per-unit basis, we 
assumed that, on average, 1,800 units of a model were produced annually 
and that a typical model is produced for 5 years. These costs are 
summarized in Table 11.

                  Table 11--Summary of Certification Testing and Research and Development Costs
----------------------------------------------------------------------------------------------------------------
               Description                             Cost per model                      Cost per unit*
----------------------------------------------------------------------------------------------------------------
Lateral Stability and Vehicle Handling                                             .............................
 Requirements:
    Compliance Testing...................  $24,000...............................  $3
    Redesign of Noncomplying Models......  $43,000...............................  $5
    Retesting of Redesigned Models.......  $24,000...............................  $3
                                          ----------------------------------------------------------------------
        Total Costs for Lateral Stability  $24,000 to $91,000....................  $3 to $10
         and Vehicle Handling.
                                          ======================================================================
Occupant Retention Requirements:                                                   .............................
    Research, Design, Implementation.....  $100,000..............................  $11
    Certification Testing................  $4,000................................  <$1
                                          ----------------------------------------------------------------------
        Total R&D and Testing Costs for    $104,000..............................  $12
         Seat Belt Requirement.
                                          ======================================================================
            Total Certification Testing    $128,000 to $195,000..................  $14 to $22
             and Research and Development
             Costs.
----------------------------------------------------------------------------------------------------------------
* Per-unit costs are rounded to the nearest whole dollar. The sums might not equal the totals due to rounding.

    In addition to the testing, research, and development costs 
described above, manufacturers will incur some additional manufacturing 
costs for extra parts or labor required to manufacture ROVs that meet 
the requirements for the proposed rule. These costs are summarized in 
Table 12. As for the vehicle handling requirements, some modifications 
to vehicles that do not comply might increase manufacturing costs; 
other modifications could decrease manufacturing costs. Therefore, we 
have assumed, on average, that there will not be any additional 
manufacturing costs required to meet the vehicle handling requirements. 
However, most manufacturers will incur additional manufacturing costs 
to meet the occupant-retention requirements. These costs are expected 
to average between $47 and $72 per vehicle. Adding the estimated 
upfront testing, research, development, and implementation costs per 
unit from Table 11 brings the total cost of the proposed rule to an 
estimated $61 to $94 per vehicle.

[[Page 69011]]



            Table 12--Summary of Per-Unit Costs and Benefits
------------------------------------------------------------------------
             Description                         Value per unit
------------------------------------------------------------------------
                                  Costs
------------------------------------------------------------------------
Manufacturing Costs:
    Lateral Stability and Vehicle      $0
     Handling Requirements.
    Passive Occupant Retention         $7
     Requirement.
    Seat Belt/Speed Limitation         $14 to $39
     Requirement--Driver Seats.
    Seat Belt/Speed Limitation         $26
     Requirement--Front Passenger
     Seats.
                                      ----------------------------------
        Total Manufacturing Costs....  $47 to $72
Certification Testing and Research     $14 to $22
 and Development Costs (from Table 4).
                                      ----------------------------------
    Total Quantifiable Cost..........  $61 to $94
------------------------------------------------------------------------
                                Benefits
------------------------------------------------------------------------
Lateral Stability and Vehicle          (not quantifiable)
 Handling Requirements.
Occupant Retention Requirements......  $2,199
                                      ----------------------------------
    Total Quantifiable Benefits......  $2,199
------------------------------------------------------------------------
Net Quantifiable Benefits............  $2,105 to $2,138
------------------------------------------------------------------------

    We were able to estimate benefits for the occupant retention 
requirement. Applying this requirement to just the driver's seat belt 
would result in benefits of about $1,498 per unit. Applying the seat 
belt/speed limitation requirement to the front passenger seat belts 
could result in an additional benefit of $701 per unit. Therefore, the 
quantifiable benefits of the proposed rule would be $2,199 per unit. 
The benefit associated with the vehicle handling and lateral stability 
requirement could not be quantified. Therefore, the benefits of the 
proposed rule could exceed the $2,199 estimated above.
    The fact that the potential benefits of the lateral stability and 
vehicle handling requirements could not be quantified should not be 
interpreted to mean that they are low or insignificant. This only means 
that we have not developed the data necessary to quantify these 
benefits. The purpose of the occupant retention requirements is to 
reduce the severity of injuries, but this requirement is not expected 
to reduce the risk of an incident occurring. The lateral stability and 
vehicle handling requirement, on the other hand, is intended to reduce 
the risk of an incident occurring that involves an ROV, and therefore, 
prevent injuries from happening in the first place. At this time, 
however, we do not have a basis for estimating what would be the 
effectiveness of the lateral stability and vehicle handling 
requirements.
    Notably, to the extent that the lateral stability and vehicle 
handling requirements are effective in reducing the number of 
incidents, the incremental benefit of the occupant retention 
requirements also would be reduced. Additionally, if the lateral 
stability and vehicle handling requirements can reduce the number of 
accidents involving ROVs, there would be fewer resulting injuries whose 
severity would be reduced by the occupant retention requirements. 
However, the resulting decrease in the incremental benefit of the seat 
belt/speed limitation requirement would be less than the benefit 
attributable to the lateral stability and vehicle handling 
requirements. Again, this is largely because the benefit of preventing 
an injury from occurring in the first place is greater than the benefit 
of reducing the severity of harm of the injury.
    Although some assumptions used in this analysis would serve to 
reduce the estimated benefit of the draft proposed rule (e.g., ignoring 
incidents in which the use of seat belts was unknown), the analysis 
also assumes that all drivers and front seat passengers would opt to 
fasten their seat belts if the speed of the vehicle was limited; and 
the analysis also would assume that no driver or passenger would 
attempt to defeat the system, which could be accomplished simply by 
passing the belt behind the rider, or passing the belt behind the seat 
before latching the belt. To the extent that consumers attempt to 
defeat the seat belt/speed limitation system, the benefits are 
overestimated.
    The estimated costs and benefits of the rule on an annual basis can 
be calculated by multiplying the estimated benefits and costs per-unit 
by the number of ROVs sold in a given year. In 2013, 234,000 ROVs were 
sold. If the proposed rule had been in effect that year, the total 
quantifiable cost would have been between $14.3 million and $22.0 
million ($61 and $94 multiplied by 234,000 units, respectively). The 
total quantifiable benefits would have been at least $515 million 
($2,199 x 234,000). Of the benefits, about $453 million (or about 88 
percent) would have resulted from the reduction in fatal injuries, and 
about $62 million (or about 12 percent) of the benefits would have 
resulted from a reduction in the societal cost of nonfatal injuries. 
About $47 million of the reduction in the societal cost of nonfatal 
injuries would have been due to a reduction in pain and suffering.

F. Alternatives

    The Commission considered several alternatives to the requirements 
in the proposed rule. The alternatives considered included: (1) Not 
issuing a mandatory rule, but instead, relying on voluntary standards; 
(2) including the dynamic lateral stability requirement or the 
understeer requirement, but not both; (3) requiring a more intrusive 
audible or visual seatbelt reminder, instead of limiting the speed of 
the vehicle if the seatbelt is not fastened; (4) extending the 
seatbelt/speed limitation requirement to include rear seats; (5) 
requiring an ignition interlock if the seatbelts are not fastened 
instead of limiting the maximum speed; and (6) limiting the maximum 
speed to 10 mph, instead of 15 mph, if the seatbelts are not fastened. 
Each of these alternatives is discussed below. The discussion includes 
the reasons that the Commission did not include the alternative in the 
proposed rule as well as qualitative discussion of costs and benefits 
where possible.

[[Page 69012]]

1. No Mandatory Standard/Rely on Voluntary Standard
    If CPSC did not issue a mandatory standard, most manufacturers 
would comply with one of the two voluntary standards that apply to 
ROVs. However, neither voluntary standard requires that ROVs 
understeer, as required by the proposed rule. According to ES staff, 
drivers are more likely to lose control of vehicles that oversteer, 
which can lead to the vehicle rolling over or causing other types of 
accidents.
    Both voluntary standards have requirements that are intended to set 
standards for dynamic lateral stability. ANSI/ROHVA 1-2011 uses a turn-
circle test for dynamic lateral stability that is more similar to the 
test in the proposed rule (for whether the vehicle understeers) than it 
is to the test for dynamic lateral stability. The dynamic stability 
requirement in ANSI/OPEI B71.9-2012 uses a J-turn test, like the 
proposed rule, but measures different variables during the test and 
uses a different acceptance criterion. However, ES staff does not 
believe that the tests procedures in either standard have been 
validated properly to be deemed capable of providing useful information 
about the dynamic stability of the vehicle. Moreover, the voluntary 
standards would find some vehicles to be acceptable, even though their 
lateral acceleration at rollover is less than 0.70 g, which is the 
acceptance criterion in the proposed rule.
    Both voluntary standards require manufacturers to include a lighted 
seat-belt reminder that is visible to the driver and remains on for at 
least 8 seconds after the vehicle is started, unless the driver's 
seatbelt is fastened. However, virtually all ROVs on the market already 
include this feature; and therefore, relying only on the voluntary 
standards would not be expected to raise seatbelt use over current 
levels of use.
    The voluntary standards include requirements for retaining the 
occupant within the protective zone of the vehicle if a rollover 
occurs, including two options for restraining the occupants in the 
shoulder/hip area. However, testing performed by CPSC identified 
weaknesses in the performance-based tilt table test option that allows 
unacceptable occupant head ejection beyond the protective zone of the 
vehicle ROPs. CPSC testing indicated that a passive shoulder barrier 
could reduce the head excursion of a belted occupant during quarter-
turn rollover events. The Commission believes that this can be 
accomplished by a requirement for a passive barrier, based on the 
dimensions of the upper arm of a 5th percentile adult female, at a 
defined area near the ROV occupants' shoulder, as contained in the 
proposed rule.
    In summary, not mandating a standard would not impose any 
additional costs on manufacturers, but neither would it result in any 
additional benefits in terms of reduced deaths and injuries. Therefore, 
not issuing a mandatory standard was not proposed by the Commission.
2. Removing Either the Lateral Stability Requirement or the Handling 
Requirement
    The CPSC considered including a requirement for either dynamic 
stability or vehicle handling, but not both. However, the Commission 
believes that both of these characteristics need to be addressed. 
According to ES staff, a vehicle that meets both the dynamic stability 
requirement and the understeer requirement should be safer than a 
vehicle that meets only one of the requirements. Moreover, the cost of 
meeting just one requirement is not substantially lower than the cost 
of meeting both requirements. The cost of testing a vehicle for 
compliance with both the dynamic lateral stability requirement and the 
vehicle handling/understeer requirement was estimated to be about 
$24,000. However, the cost of testing for compliance with just the 
dynamic stability requirement would be about $20,000, or only about 17 
percent less than the cost of testing for compliance with both 
requirements. This is because the cost of renting and transporting the 
vehicle to the test site, instrumenting the vehicle for the tests, and 
making some initial static measurements are virtually the same for both 
requirements and would only have to be done once, if the tests for both 
requirements were conducted on the same day. Moreover, changes in the 
vehicle design that affect the lateral stability of the vehicle could 
also impact the handling of the vehicle. For these reasons, the 
proposed rule includes a dynamic stability requirement and a vehicle 
handling requirement.
3. Require Intrusive Seatbelt Reminder in Lieu of the Speed Limitation 
Requirements
    Instead of seatbelt/speed limitation requirements in the proposed 
rule, the Commission considered a requirement for ROVs to have loud or 
intrusive seatbelt reminders. Currently, most ROVs meet the voluntary 
standards that require an 8-second visual seatbelt reminder. Some more 
intrusive systems have been used on passenger cars. For example, the 
Ford ``BeltMinder'' system resumes warning the driver after about 65 
seconds if his or her seatbelt is not fastened and the car is traveling 
at more than 3 mph. The system flashes a warning light and sounds a 
chime for 6 seconds every 30 seconds for up to 5 minutes so long as the 
car is operating and the driver's seatbelt is not fastened. Honda 
developed a similar system in which the warning could last for longer 
than 9 minutes if the driver's seatbelt is not fastened. Studies of 
both systems found that a statistically significant increase in the use 
of seatbelts of 5 percent (from 71 to 76 percent) and 6 percent (from 
84 to 90 percent), respectively.\101\ However, these more intrusive 
seatbelt warning systems are unlikely to be as effective as the 
seatbelt speed limitation requirement in the proposed rule. The 
Commission believes that the requirement will cause most drivers and 
passengers who wish to exceed 15 mph to fasten their seatbelts. 
Research supports this position. One experiment used a haptic feedback 
system to increase the force the driver needed to exert to depress the 
gas pedal when the vehicle exceeded 25 mph if the seatbelt was not 
fastened. The system did not prevent the driver from exceeding 25 mph, 
but it increased the amount of force required to depress the gas pedal 
to maintain a speed greater than 25 mph. In this experiment all seven 
participants chose to fasten their seatbelts.\102\
---------------------------------------------------------------------------

    \101\ Caroleene Paul, ``Proposal for Seatbelt Speed Limiter On 
Recreational Off-Highway Vehicles (ROVs),'' CPSC Memorandum (2013).
    \102\ Ron Van Houten, Bryan Hilton, Richard Schulman, and Ian 
Reagan, ``Using Haptic Feedback to Increase Seatbelt Use of Service 
Vehicle Drivers,'' U.S. Department of Transportation, Report No. DOT 
HS 811 434 (January 2011).
---------------------------------------------------------------------------

    The more intrusive seatbelt reminder systems used on some passenger 
cars have been more limited in their effectiveness. The Honda system, 
for example, reduced the number of unbelted drivers by about 38 
percent; the Ford system reduced the number of unbelted drivers by only 
17 percent.\103\ Additionally, ROVs are open vehicles and the ambient 
noise is likely higher than in the enclosed passenger compartment of a 
car. It is likely that some ROV drivers would not hear the warning and 
be motivated to fasten their seatbelts unless the warning was 
substantially louder than the systems used in passenger cars.
---------------------------------------------------------------------------

    \103\ The Honda system increased seatbelt use from 84 percent to 
90 percent. Therefore, the percentage of unbelted drivers was 
reduced by about 38 percent, or 6 percent divided by 16 percent. The 
Ford system increased seatbelt use from 71 percent to 76 percent. 
Therefore, the percentage of unbelted drivers was reduced by about 
17 percent, or 5 percent divided by 29 percent.

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

[[Page 69013]]

    The cost to manufacturers of some forms of more intrusive seat belt 
reminders could be less than the cost of the speed limitation 
requirement in the draft proposed rule. However, the cost of the seat 
belt/speed limitation requirement was estimated to be less than $72 per 
ROV.\104\ If the experience with the Honda and Ford systems discussed 
above are relevant to ROVs, the benefits of a more intrusive seat belt 
reminder system could be less than 38 percent of the benefits estimated 
for the requirement in the draft proposed rule or less than $835 per 
ROV. Therefore, even if the cost of a more intrusive seat belt reminder 
system was close to $0, the net benefits would be less than the seat 
belt/speed limitation requirement in the draft proposed rule, which 
were estimated to be at least $2,105. Therefore, the alternative of a 
more intrusive seat belt reminder was not included in the proposed 
rule.
---------------------------------------------------------------------------

    \104\ This estimate is based on manufacturing cost estimates of 
$39 to apply the requirement to the driver's seat and $26 to apply 
the requirement to the front passenger's seat, plus $12 for 
research, development and certification testing.
---------------------------------------------------------------------------

4. Extending the Seatbelt/Speed Limitation Requirement To Include Rear 
Seats
    The Commission considered extending the seatbelt/speed limitation 
requirement to include the rear passenger seats, when present. 
According to one exposure survey, about 20 percent of the respondents 
reported that their ROVs had a seating capacity of at least four 
occupants, which indicates that the ROV had rear passenger seating 
locations.\105\
---------------------------------------------------------------------------

    \105\ Heiden Associates, Results from the 2008 ROV Exposure 
Survey (APPENDIX 2 to Joint Comments of Arctic Cat Inc., Bombardier 
Recreational Products Inc., Polaris Industries Inc., and Yamaha 
Motor Corporation, U.S.A regarding CPSC Advance Notice of Proposed 
Rulemaking-Standard for Recreational Off-Highway Vehicles: Docket 
No. CPSC--2009-0087), Alexandria Virginia (December 4, 2009).) This 
suggests that there were about 114,000 ROVs with rear passenger 
seats in 2010 (0.2 x 570,000).
---------------------------------------------------------------------------

    The cost of extending this requirement to include the rear 
passenger seats would be expected to be the same per seat as extending 
the requirement to include the right-front and middle-front passengers, 
or $24 per seat. Therefore, the cost of this requirement would be $48 
to $72 per ROV, depending upon whether the ROV had two or three rear 
seating locations.
    Three of the 231 fatalities (or 1.3 percent) involved a person in a 
rear seat who did not have their seatbelt fastened. Using the same 
assumptions used to calculate the benefits of the seatbelt/speed 
limitation for passengers in the front seats (i.e., that 73 percent 
occurred at speeds of 15 mph or greater and seatbelts would reduce the 
risk of death by 45 percent), extending the requirement to include the 
rear seats could have potentially reduced the number of fatalities in 
2010 by 0.2 or about one death every 5 years, all other things equal. 
Therefore, extending the seatbelt/speed limitation requirement to the 
rear passenger seats could reduce the annual risk of fatal injury by 
0.00000175 (0.2 / 114,000) per ROV in use. Assuming a societal benefit 
of $8.05 million per death averted results in an estimated annual 
benefit of about $14 per ROV in use ($8.05 million x 0.00000175) in 
terms of reduced fatal injuries.
    Three of the 388 nonfatal injuries (or 0.8 percent) involved 
passengers in rear seats who did not have their seatbelts fastened. 
This suggests that about 89 of the estimated 11,100 medically attended 
injuries in 2010 may have happened to unbelted rear passengers. Again, 
assuming that 73 percent of these occurred at speeds of 15 mph or 
faster, about 65 medically attended injuries might have been addressed 
by the seatbelt/speed limitation requirement if applied to the rear 
seating locations. This represents a risk of a nonfatal, medically 
attended injury of 0.0005702 (65 / 114,000) per ROV in use per year. 
The societal cost of this risk is $17, assuming an average nonfatal, 
medically attended injury cost of $29,383. If seatbelts could reduce 
the cost of these injuries by 20 percent, by reducing the number of 
injuries in their severity, the value of the reduction would be $3 per 
ROV in use per year.
    Combining the benefit of $14 for the reduction in fatal injuries 
and $3 for the reduced cost of nonfatal, medically attended injuries 
yields a combined benefit of $17 per ROV in use per year. The present 
value of this estimated benefit over the expected useful life of a ROV 
is $234. This is greater than the quantifiable cost of $48 to $72. 
However, these estimates of the costs and benefits are probably 
oversimplified the costs may have been understated and the benefits 
overstated. The Commission is hesitant to recommend this alternative 
for the several reasons.
    First, as discussed earlier, a system that includes all passenger 
seats would comprise more parts than a system that included only the 
front passenger seats. A failure in only one of the parts could result 
in significant cost to the users for repairs, lost time and utility of 
the vehicle while it is being repaired, or the inability of the vehicle 
to reach its potential speed. These failures could occur because a 
faulty seat belt latch sensor does not detect or signal that a seatbelt 
is latched or because a faulty seat switch incorrectly registers the 
presence of a passenger when a passenger is not present. This cost 
cannot be quantified. However, if such failures are possible, the costs 
of extending the seatbelt/speed limitation requirement to include the 
rear seats would be higher than the $48 to $72 estimated above.
    Second, as discussed previously, there is some correlation between 
the seatbelt use of the driver and other passengers on the ROV. If the 
driver and front passengers fasten their seatbelts, there is reason to 
believe that some rear passengers will also fasten their seatbelts. If 
so, the benefits of including the rear seat passengers could be 
overestimated above. Moreover, even if there was no correlation, 
including only the driver and front seat passengers would still achieve 
about 98 percent of the total potential benefits from the seatbelt/
speed limitation requirement.\106\
---------------------------------------------------------------------------

    \106\ The potential net benefit of the seatbelt/speed limitation 
requirement resulting from its application to the driver and front 
passengers was estimated to be $2,199 per ROV. The potential net 
benefit resulting from its application to the rear seats was 
estimated to be $234 per ROV with rear seats. However, only about 20 
percent of ROVs were assumed to have rear seats. Therefore, the 
weighted benefit over all ROVs of extending the seatbelt/speed 
limitation requirement to include the rear seats would be about $47 
per ROV ($234 x 0.2). The potential weighted benefit would be 
$2,246, of which about 2 percent ($47 / $2,246) would be 
attributable to extending the requirement to the rear seats.
---------------------------------------------------------------------------

5. Requiring an Ignition Interlock Instead of Limiting the Maximum 
Speed
    The Commission considered whether an ignition interlock requirement 
that did not allow the vehicle to be started unless the driver's 
seatbelt was buckled would be appropriate for ROVs. However, the 
history of ignition interlock systems to encourage seatbelt use on 
passenger cars suggests that consumer resistance to an ignition 
interlock system could be strong. In 1973, NHTSA proposed requiring an 
interlock system on passenger cars. However, public opposition to the 
proposed requirement led Congress to prohibit NHTSA from requiring an 
ignition interlock system.\107\ For this reason, the Commission is not 
proposing this alternative. Instead, the proposed rule would allow 
people to use ROVs at low speeds without requiring seat belts to be 
fastened.
---------------------------------------------------------------------------

    \107\ Caroleene Paul, ``Proposal for Seatbelt Speed Limiter on 
Recreational Off-Highway Vehicles (ROVs),'' CPSC Memorandum (2013). 
U.S. Consumer Product Safety Commission, Bethesda MD (2013).

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

[[Page 69014]]

6. Limiting the Maximum Speed to 10 mph if the Driver's Seatbelt Is Not 
Fastened
    The Commission considered limiting the maximum speed of the ROV to 
10 mph if the driver's seatbelt was not fastened, instead of 15 mph, as 
in the proposed rule. In making this determination, we weigh some 
potentially quantifiable factors against some unquantifiable factors. 
The expected benefits of limiting the maximum speed to 10 mph are 
higher than the expected benefits of limiting the maximum speed to 15 
mph. Based on the injuries reported to CPSC for which the speed was 
reported and the seatbelt use was known, about 15 percent of the people 
injured in ROV accidents who were not wearing seatbelts were traveling 
between 10 and 15 mph. Therefore, decreasing the maximum allowed speed 
of an ROV to 10 mph if the driver's or right front passenger's seatbelt 
is not fastened could increase the expected benefits of the requirement 
by up to 21 percent (0.15 / 0.73). There would be no difference between 
the two alternatives in terms of the quantified costs.
    Although the quantified benefits would be increased and the 
quantified costs would not be affected by this alternative, the 
Commission believes that the unquantifiable costs would be higher if 
the maximum speed allowed was set at 10 mph instead of 15 mph. 
Commission staff believes this could have a negative impact on consumer 
acceptance of the requirement. The unquantifiable costs include: The 
time, inconvenience, and discomfort to some users who would prefer not 
to wear seatbelts. These users could include: People using the ROVs for 
work or utility purposes, who might have to get on and off the ROV 
frequently, and who are likely to be traveling at lower rates of speed, 
but who occasionally could exceed 10 mph. Some of these users could be 
motivated to defeat the requirement (and this could be done easily), 
which could reduce the benefits of the proposed rule. Allowing ROVs to 
reach speeds of up to 15 mph without requiring the seatbelt to be 
fastened would mitigate some of the inconvenience or discomfort of the 
requirement to these users, and correspondingly, consumers would have 
less motivation to attempt to defeat the requirement.
    ROV manufacturers would have the option of setting the maximum 
speed that their models could reach without requiring the seatbelts to 
be fastened--so long as the maximum speed was no greater than 15 miles 
per hour. Therefore, manufacturers could set a maximum speed of less 
than 15 mph if they believed this was in their interest to do so. One 
ROV manufacturer has introduced ROV models that will not exceed 9.3 mph 
(15 km/hr.) unless the driver's seatbelt is fastened.

G. Conclusion

    We estimate the quantifiable benefits of the proposed rule to be 
about $2,199 per ROV, and we estimate the quantifiable costs to be 
about $61 to $94 per ROV. Therefore, the benefits would exceed the 
costs by a substantial margin. However, the only benefits that could be 
quantified would be the benefits associated with the seat belt/speed 
limitation requirement. The lateral stability and vehicle handling 
requirements would also be expected to reduce deaths and injuries and 
so result in additional benefits, but these were not quantifiable.
    There could be some unquantifiable costs associated with the rule. 
Some consumers might find the requirement to fasten their seat belts 
before the vehicle can exceed 15 mph to be inconvenient or 
uncomfortable. The 15 mph threshold as opposed to a 10 mph threshold 
was selected for the requirement to limit the number of consumers who 
would be inconvenienced by the requirement and might be motivated to 
defeat the system. Some consumers might prefer an ROV that oversteers 
under more conditions than the proposed rule would allow. However, the 
number of consumers who have a strong preference for oversteering 
vehicles is probably low.
    Several alternatives to requirements in the proposed rule were 
considered, including relying on voluntary standards or requiring more 
intrusive seat belt reminders (as opposed to the speed limitation 
requirement). However, the Commission determined that the benefits of 
the requirements in the proposed rule would probably exceed their 
costs, considering both the quantifiable and unquantifiable costs and 
benefits.

XI. Paperwork Reduction Act

    This proposed rule contains information collection requirements 
that are subject to public comment and review by OMB under the 
Paperwork Reduction Act of 1995 (44 U.S.C. 3501-3521). In this 
document, pursuant to 44 U.S.C. 3507(a)(1)(D), we set forth:
     A title for the collection of information;
     a summary of the collection of information;
     a brief description of the need for the information and 
the proposed use of the information;
     a description of the likely respondents and proposed 
frequency of response to the collection of information;
     an estimate of the burden that shall result from the 
collection of information; and
     notice that comments may be submitted to the OMB.
    Title: Safety Standard for Recreational Off-Highway Vehicles 
(ROVs).
    Number of Respondents: We have identified 20 manufacturers of ROVs.
    Number of Models: We estimate that there are about 130 different 
models of ROVs, or an average of 6.5 models per manufacturer. This 
estimate counts as a single model, all models of a manufacturer that do 
not appear to differ from each other in terms of performance, such as 
engine size, width, number of seats, weight, horsepower, capacity, and 
wheel size. In other words, if the models differed only in terms of 
accessory packages, or in the case of foreign manufacturers, differed 
only in the names of the domestic distributors, then they were counted 
as the same model.
    Number of Reports per Year: Manufacturers will have to place a hang 
tag on each ROV sold. In 2013, about 234,000 ROVs were sold, or about 
1,800 units per model. This would be a reasonable estimate of the 
number of responses per year. On average, each manufacturer would have 
about 11,700 responses per year.
    Burden Estimates per Model: The reporting burden of this 
requirement can be divided into two parts. The first is designing the 
hang tag for each model. The second is printing and physically 
attaching the hang tag to the ROV. These are discussed in more detail 
below.
    Designing the Hang tag: We estimate that it will take about 30 
minutes to design the hang tag for each model. The first year the rule 
is in effect, manufacturers will have to design the hang tag for each 
of their models. However, the same model might be in production for 
more than one year. If ROV models have a production life of about 5 
years before being redesigned, then the same hang tag might be useable 
for more than 1 year. Therefore, in year 1, on average, the burden on 
each manufacturer will be about 3.25 hours to design the hang tag (0.5 
hours per model x 6.5 models). In subsequent years, the burden on each 
manufacturer will be about 0.65 hours assuming that manufacturers will 
have to redesign the hang tag only when they redesign the

[[Page 69015]]

ROV and that ROVs are redesigned, on average, about every 5 years. 
Assuming this work will be performed by a professional employee, the 
cost per manufacturer will be $206 the first year and $41 in each 
subsequent year.\108\
---------------------------------------------------------------------------

    \108\ This estimate is based on the total compensation for 
management, professional, and related workers in private, goods 
producing industries, as reported by the Bureau of Labor Statistics 
(March 2014), available at https://www.bls.gov/ncs/. Please note, in 
the draft regulatory analysis, we are using 2010 as the base year 
with all values expressed in 2012 dollars. Therefore, these 
estimates might be slightly higher than estimated in the regulatory 
analysis.
---------------------------------------------------------------------------

    Printing and Placing the Hang tag on Each Vehicle: Based on 
estimates for printing obtained at: https://www.uprinting.com and 
estimates for the ties obtained from https://blanksusa.com, we estimate 
that the cost of the printed hang tag and wire for attaching the hang 
tag to the ROV will be about $0.08. Therefore, the total cost of 
materials for the average manufacturer with 6.5 models, producing 1,800 
units of each model, would be about $936 per year ($0.08 x 6.5 models x 
1,800 units).
    We estimate that it will take about 20 seconds to attach a hang tag 
to each vehicle. Assuming an annual production of 1,800 units of each 
model, on average, this comes to 10 hours per model or an average of 65 
hours per manufacturer or respondent, assuming an average of 6.5 models 
per manufacturer. Assuming a total compensation of $26.12 per hour, the 
cost would be $261 per model or $1,698 per manufacturer, assuming an 
average of 6.5 models per manufacturer.\109\
---------------------------------------------------------------------------

    \109\ Estimate is based on the total compensation for 
production, transportation, and material-moving workers, private, 
goods-producing industries, as reported by the Bureau of Labor 
Statistics (March 2014), available at: https://www.bls.gov/ncs/.
---------------------------------------------------------------------------

    Total Burden of the Hang tag Requirement: The total burden of the 
hang tag requirement the first year will consist of the following 
components:
    Designing the Hang tags: 65 hours (0.5 hours x 130 models). 
Assuming a total compensation rate of $63.36 per hour (professional and 
related workers), the cost would be $4,118.
    Placing the Hang tags on the Vehicles: 1,300 hours (234,000 
vehicles x 20 seconds). Assuming a total compensation rate of 26.12 per 
hour (production, transportation, and material moving workers), the 
total cost is $33,956.
    Total Compensation Cost: The total compensation cost for this 
requirement would be $38,074 in the first year. In subsequent years, 
the burden of designing the hang tag is estimated to be about one-fifth 
the burden in the initial year, or 13 hours, assuming that each ROV 
model either undergoes a significant design change or is replaced by a 
different model every 5 years. Therefore, the compensation cost of 
designing the hang tag in subsequent years would be about $824 ($4,118/
5). The total compensation cost in subsequent years would be $34,780.
    Total Material Cost: The cost of the printed hang tags and ties for 
attaching the hang tag to the vehicles is estimated to be about 8 cents 
each. Therefore, the total material cost would be $18,720 ($0.08 x 
234,000 units).
    Total Cost of Hang tag Requirement: Based on the above estimates, 
the total cost of the hang tag requirement in the initial year is 
estimated to be about $56,794. In subsequent years, the total cost 
would be slightly less, about $53,500.
    In compliance with the Paperwork Reduction Act of 1995 (44 U.S.C. 
3507(d)), we have submitted the information collection requirements of 
this rule to the OMB for review. Interested persons are requested to 
submit comments regarding information collection by December 19, 2014, 
to the Office of Information and Regulatory Affairs, OMB (see the 
ADDRESSES section at the beginning of this notice).
    Pursuant to 44 U.S.C. 3506(c)(2)(A), we invite comments on:
     Whether the collection of information is necessary for the 
proper performance of the CPSC's functions, including whether the 
information will have practical utility;
     the accuracy of the CPSC's estimate of the burden of the 
proposed collection of information, including the validity of the 
methodology and assumptions used;
     ways to enhance the quality, utility, and clarity of the 
information to be collected;
     ways to reduce the burden of the collection of information 
on respondents, including the use of automated collection techniques, 
when appropriate, and other forms of information technology; and
     the estimated burden hours associated with label 
modification, including any alternative estimates.

XII. Initial Regulatory Flexibility Analysis

    This section provides an analysis of the impact on small businesses 
of a proposed rule that would establish a mandatory safety standard for 
ROVs. Whenever an agency is required to publish a proposed rule, 
section 603 of the Regulatory Flexibility Act (5 U.S.C. 601-612) 
requires that the agency prepare an initial regulatory flexibility 
analysis (IRFA) that describes the impact that the rule would have on 
small businesses and other entities. An IRFA is not required if the 
head of an agency certifies that the proposed rule will not have a 
significant economic impact on a substantial number of small entities. 
5 U.S.C. 605. The IRFA must contain:
    (1) A description of why action by the agency is being considered;
    (2) a succinct statement of the objectives of, and legal basis for, 
the proposed rule;
    (3) a description of and, where feasible, an estimate of the number 
of small entities to which the proposed rule will apply;
    (4) a description of the projected reporting, recordkeeping and 
other compliance requirements of the proposed rule, including an 
estimate of the classes of small entities which will be subject to the 
requirement and the type of professional skills necessary for 
preparation of the report or record; and
    (5) an identification to the extent practicable, of all relevant 
Federal rules which may duplicate, overlap or conflict with the 
proposed rule.
    An IRFA must also contain a description of any significant 
alternatives that would accomplish the stated objectives of the 
applicable statutes and that would minimize any significant economic 
impact of the proposed rule on small entities. Alternatives could 
include: (1) Establishment of differing compliance or reporting 
requirements that take into account the resources available to small 
businesses; (2) clarification, consolidation, or simplification of 
compliance and reporting requirements for small entities; (3) use of 
performance rather than design standards; and (4) an exemption from 
coverage of the rule, or any part of the rule thereof, for small 
entities.

A. Reason for Agency Action

    ROVs were first introduced in the late 1990s. Sales of ROVs 
increased substantially over the next 15 years. The number of deaths 
associated with ROVs has substantially increased over the same period, 
from no reported deaths in 2003, to at least 76 reported deaths in 
2012. As explained in this preamble, some ROVs on the market have 
hazardous characteristics that could be addressed through a mandatory 
safety standard.

B. Objectives of and Legal Basis for the Rule

    The Commission proposes this rule to reduce the risk of death and 
injury associated with the use of ROVs. The rule is promulgated under 
the authority

[[Page 69016]]

of the Consumer Product Safety Act (CPSA).

C. Small Entities to Which the Rule Will Apply

    The proposed rule would apply to all manufacturers and importers of 
ROVs. Under criteria set by the U.S. Small Business Administration 
(SBA), manufacturers of ROVs are considered small businesses if they 
have fewer than 500 employees. We have identified one ROV manufacturer 
with fewer than 500 employees.
    Importers of ROVs could be wholesalers or retailers. Under the 
criteria set by the SBA, wholesalers of ROVs and other motor vehicles 
or powersport vehicles are considered small businesses if they have 
fewer than 100 employees; and retail dealers that import ROVs and other 
motor or powersport vehicle dealers are considered small if their 
annual sales volume is less than $30 million. We are aware of about 20 
firms in 2013 that import ROVs from foreign suppliers that would be 
considered small businesses.\110\ (There may be other small firms that 
manufacture or import ROVs of which we are not aware.)
---------------------------------------------------------------------------

    \110\ The Commission made these determinations using information 
from Dun & Bradstreet, Reference USAGov, company Web sites, and 
regional business publications.
---------------------------------------------------------------------------

D. Compliance, Reporting, and Record Keeping Requirements of Proposed 
Rule

    The proposed rule would establish a mandatory safety standard 
consisting of several performance requirements for ROVs sold in the 
United States. The proposed rule would also establish test procedures 
through which compliance with the performance requirements would be 
determined. The proposed rule includes: (1) Lateral stability and 
vehicle handling requirements that specify a minimum level of rollover 
resistance for ROVs and a requirement that ROVs exhibit sub-limit 
understeer characteristics; and (2) occupant retention requirements 
that would limit the maximum speed of an ROV to no more than 15 miles 
per hour (mph), unless the seat belts of the driver and front 
passengers are fastened, and would require ROVs to have a passive 
means, such as a barrier or structure, to limit the ejection of a 
belted occupant in the event of a rollover.
    Manufacturers would be required to test their ROV models to check 
that the models comply with the requirements of the proposed rule, and 
if necessary, modify their ROV models to comply. The costs of these 
requirements are discussed more fully in the preliminary regulatory 
analysis. Based on that analysis, we expect that the test for lateral 
stability and the test for vehicle handling will be conducted at the 
same time, and we estimate that the cost of this combined testing would 
be about $24,000 per model. In many cases, we expect that this testing 
will be performed by a third party engineering consulting or testing 
firm. If an ROV model must be modified to comply with the requirement 
and then retested, we estimate that the cost to manufacturers could 
reach $91,000 per model, including the cost of the initial testing, the 
cost of modifying design of the model, and the cost of retesting the 
model after the model has been modified. We estimate that the cost of 
implementing the occupant retention requirements will be about $104,000 
per model. This includes the cost to research, develop, implement, and 
test a system that will limit the speed of the ROV when the seat belts 
are not fastened, as well as an occupant protection barrier or 
structure. Therefore, the total cost of certification testing and 
research and design could range from about $128,000 to $195,000. (Costs 
are expressed in 2012 dollars.)
    In addition to the upfront testing and research and development 
costs, there will be some ongoing manufacturing costs associated with 
the proposed rule. These manufacturing costs include the cost of the 
parts required to meet any of the requirements of the proposed rule, 
such as seat belt use sensors and the necessary wiring and the cost of 
installing these parts on the vehicles during assembly. As estimated in 
the preliminary regulatory analysis, the ongoing manufacturing costs 
would be $47 to $72 per vehicle.
    The proposed rule includes a requirement that manufacturers report 
the lateral acceleration at rollover value of an ROV model to potential 
consumers through the use of a hang tag attached to the ROV. 
Manufacturers would obtain the rollover resistance value when they 
conduct the lateral stability and vehicle handling tests to determine 
compliance with both requirements. The required format of the hangtag 
is described in the proposed rule. We estimate that it will cost 
manufacturers less than $0.25 per vehicle to print the hangtags with 
the rollover resistance values and to attach the hangtags to the 
vehicles.

E. Federal Rules That May Duplicate, Overlap, or Conflict With the 
Proposed Rule

    In accordance with Section 14 of the Consumer Product Safety Act 
(CPSA), manufacturers would have to issue a general conformity 
certificate (GCC) for each ROV model, certifying that the model 
complies with the proposed rule. According to Section 14 of CPSA, GCCs 
must be based on a test of each product or a reasonable testing 
program; and GCCs must be provided to all distributors or retailers of 
the product. The manufacturer would have to comply with 16 CFR part 
1110 concerning the content of the GCC, retention of the associated 
records, and any other applicable requirement.

F. Potential Impact on Small Entities

    One purpose of the regulatory flexibility analysis is to evaluate 
the impact of a regulatory action and determine whether the impact is 
economically significant. Although the SBA allows considerable 
flexibility in determining ``economically significant,'' CPSC staff 
typically uses one percent of gross revenue as the threshold for 
determining ``economic significance.'' When we cannot demonstrate that 
the impact is lower than one percent of gross revenue, we prepare a 
regulatory flexibility analysis.\111\
---------------------------------------------------------------------------

    \111\ The one percent of gross revenue threshold is cited as 
example criteria by the SBA and is commonly used by agencies in 
determining economic significance (see U.S. Small Business 
Administration, Office of Advocacy. A Guide for Government Agencies: 
How to Comply with the Regulatory Flexibility Act and Implementing 
the President's Small Business Agenda and Executive Order 13272. May 
2012, pp. 18-20. https://www.sba.gov/sites/default/files/rfaguide_0512_0.pdf).
---------------------------------------------------------------------------

1. Impact on Small Manufacturers
    The sole, small ROV manufacturer may need to devote some resources 
to bringing its ROV models into compliance with the proposed rule. This 
is a relatively new manufacturer of ROVs and other utility vehicles. We 
do not have information on the extent to which the models offered by 
this manufacturer would meet the requirements of the proposed rule or 
the extent to which this particular manufacturer would be impacted by 
the proposed rule.
2. Impact on Small Importers
    CPSC is aware of about 20 firms that import ROVs from foreign 
suppliers that would be considered small businesses. As explained more 
fully below, a small importer could be adversely impacted by the 
proposed rule if its foreign supplier does not provide testing reports 
or a GCC and the small importer must conduct the testing in support of 
a GCC. Additionally, a small importer could experience a significant 
impact if the foreign supplier withdraws from the U.S. market rather 
than conduct the necessary testing or modify the ROVs to comply with 
the proposed rule. If sales

[[Page 69017]]

of ROVs are a substantial source of the importer's business, and the 
importer cannot find an alternative supplier of ROVs, the impact could 
be significant. However, we do not expect a widespread exodus of 
foreign manufacturers from the U.S. market. The U.S. market for ROVs 
has been growing rapidly in recent years, and at least some foreign 
manufacturers will likely want to continue taking advantage of these 
business opportunities by maintaining a U.S. presence. In addition, 
most of these importers also import products other than ROVs, such as 
scooters, motorcycles, and other powersport equipment. Therefore, ROVs 
are not their sole source of revenue. Importers may be able to reduce 
any impact on their revenue by increasing imports and sales of these 
other products.
    Small importers will be responsible for issuing a GCC certifying 
that their ROVs comply with the proposed rule if the rule becomes 
final. However, importers may issue GCCs based upon certifications 
provided by or testing performed by their suppliers. The impact on 
small importers should not be significant if their suppliers provide 
the certificates of conformity or testing reports on which the 
importers may rely to issue their own GCCs.
    If a small importer's supplier does not provide the GCC or testing 
reports, then the importer would have to test each model for 
conformity. Importers would likely contract with an engineering 
consulting or testing firm to conduct the certification tests. As 
discussed in the regulatory analysis, the certification testing could 
cost more than $28,000 per model ($24,000 for the lateral stability and 
vehicle handling requirements and $4,000 for the seat belt/speed 
limitation requirement). This would exceed 1 percent of the revenue for 
about one-half of the small importers, assuming that they continue to 
import the same mix of products as in the pre-regulatory environment.

G. Conclusion

    We do not know how many, if any, foreign suppliers might exit the 
market rather than comply with the proposed rule. Nor do we know the 
number of foreign suppliers that may not be willing to provide small 
importers with testing reports or GCCs. A small importer could 
experience a significant impact if the importer has to conduct testing 
in support of a GCC. We expect that most importers, however, will rely 
upon certifications or testing performed by their suppliers. Thus, 
although uncertainty exists, the proposed rule will not likely have a 
significant direct impact on a substantial number of small firms.

H. Alternatives for Reducing the Adverse Impact on Small Businesses

    The Commission welcomes comments on this IRFA. Small businesses 
that believe they will be affected by the proposed rule are especially 
encouraged to submit comments. The comments should be specific and 
describe the potential impact, magnitude, and alternatives that could 
reduce the impact of the proposed rule on small businesses.
    Several alternatives to the proposed rule were considered, some of 
which could reduce the potential impact on some small firms. These 
include: (1) Not issuing a mandatory standard; (2) dropping the lateral 
stability requirement or the vehicle handling requirement; (3) 
requiring a more intrusive seat belt reminder instead of the speed 
limitation requirement; and (4) requiring an ignition interlock if a 
seat belt is not fastened, instead of limiting the maximum speed. For 
the reasons discussed below, the CPSC did not include these 
alternatives in the proposed rule.
1. Not Issuing a Mandatory Standard
    If CPSC did not issue a mandatory standard, most manufacturers 
would comply with one of the two voluntary standards that apply to ROVs 
and there would be no impact on the small manufacturer or small 
importers. However, neither voluntary standard requires that ROVs 
understeer, as required by the proposed rule. According to ES staff, 
drivers are more likely to lose control of vehicles that oversteer, 
which can lead to the vehicle rolling over or to other types of 
accidents. Additionally, although both voluntary standards have 
requirements for dynamic lateral stability or rollover resistance, ES 
staff does not believe that the test procedures in these standards have 
been properly validated as being capable of providing useful 
information about the dynamic stability of the vehicle.
    The voluntary standards require that manufacturers include a 
lighted seat-belt reminder that is visible to the driver and remains on 
for at least 8 seconds after the vehicle is started, unless the 
driver's seat belt is fastened. However, virtually all ROVs on the 
market already include this feature; and therefore, relying only on the 
voluntary standards would not be expected to raise seat belt use over 
its current level. Moreover, the preliminary regulatory analysis showed 
that the projected benefits of the seat belt/speed limitation 
requirement would be substantially greater than the costs.
    Finally, the Commission believes that the occupant retention 
barrier in the current ROVs could be improved at a modest cost per ROV. 
For these reasons, the Commission believes that relying on compliance 
with voluntary standards is not satisfactory and is adopting the 
requirements in the proposed rule.
2. Dropping the Lateral Stability Requirement or the Understeer 
Requirement
    The Commission considered including a performance requirement for 
either lateral stability or vehicle handling, but not both. As 
mentioned previously, the vehicle handling requirement is designed to 
allow ROVs to understeer. However, the Commission believes that both of 
these characteristics need to be addressed. According to ES staff, a 
vehicle that meets both the lateral stability requirement and the 
understeer requirement should be safer than a vehicle that meets only 
one of the requirements. Moreover, the cost of meeting just one 
requirement is not substantially lower than the cost of meeting both 
requirements. The cost of testing a vehicle for compliance with both 
the dynamic lateral stability and vehicle handling requirements was 
estimated to be about $24,000. The cost of testing for compliance with 
the lateral stability requirement would be about $20,000, and the cost 
of testing for compliance with just the vehicle handling requirement 
would be about $17,000. Moreover, changes in the vehicle design that 
affect the lateral stability of the vehicle could also impact the 
handling of the vehicle. For these reasons, the proposed rule includes 
both the lateral stability and understeer requirements in the proposed 
rule.
3. Require ROVs To Have Loud or Intrusive Seat Belt Reminders in Lieu 
of the Speed Limitation Requirements
    Instead of seat belt/speed limitation requirements in the proposed 
rule, the Commission considered requiring ROVs to have loud or 
intrusive seat belt reminders. Most ROVs currently have a seat belt 
reminder in the form of a warning light that comes on for about 8 
seconds. Most do not include any audible warning. As discussed in the 
preliminary regulatory analysis, staff considered requiring a more 
intrusive seat belt reminder, such as a loud audible warning that would 
sound for a minute or more. Manufacturers would incur some costs to 
comply with a requirement for a more intrusive seat belt reminder. For 
example, the seat belt

[[Page 69018]]

use sensors (estimated to cost about $7 per seat) and sensor switches 
(estimated to cost about $13 per seat) would still be required. 
However, the research and development costs to design and implement a 
more intrusive seat belt reminder system would probably be less than 
the estimated cost to develop a system that limited the maximum speed 
of the vehicle.
    Some intrusive systems have been used on passenger cars and have 
been found to be effective in increasing seat belt use. One system 
reduced the number of unbelted drivers by 17 percent and another by 
about 38 percent.\112\ However, a more intrusive seat belt warning 
system is unlikely to be as effective as the seat belt/speed limitation 
requirement in the proposed rule. ROVs are open vehicles and the 
ambient noise is likely higher than in the enclosed passenger 
compartment of a car. It is likely that some ROV drivers would not hear 
the warning and be motivated to fasten their seat belts, unless the 
warning was substantially louder than the systems used in passenger 
cars. The Commission believes that the requirement will cause most 
drivers and passengers who want to exceed 15 mph to fasten their seat 
belts. Moreover, the analysis in the preliminary regulatory analysis 
showed that the societal benefits of the seat belt/speed limitation 
requirement in the proposed rule would exceed the costs by a 
substantial margin. Because CPSC does not believe that a more intrusive 
seat belt reminder would be effective in a ROV, and because Commission 
staff believes that the seat belt/speed limitation requirement would 
result in substantial net benefits, this alternative was not included 
in the proposed rule.
---------------------------------------------------------------------------

    \112\ Memorandum from Caroleene Paul, ``Proposal for Seat Belt 
Speed Limiter on Recreational Off-Highway Vehicles (ROVs),'' U.S. 
Consumer Product Safety Commission, Bethesda, MD 8 December 2013).
---------------------------------------------------------------------------

4. Requiring an Ignition Interlock Instead of Limiting the Maximum 
Speed
    CPSC considered whether an ignition interlock requirement that did 
not allow the vehicle to be started unless the driver's seat belt was 
buckled would be appropriate for ROVs. However, the history of ignition 
interlock systems as a way to encourage seat belt use on passenger cars 
suggests that consumer resistance to an ignition interlock system that 
prevents starting the vehicle could be strong. For this reason, CPSC 
rejects this alternative, and instead, proposes a rule that allows 
people to use ROVs at low speeds without having to fasten their seat 
belts. However, manufacturers who believe that the cost of an ignition 
interlock system will be substantially lower than a system that limits 
the maximum speed of the vehicle, and who do not believe that consumer 
rejection of an ignition interlock system will be a problem, can use an 
ignition interlock system to comply with the seat belt speed limitation 
requirement.

XIII. Environmental Considerations

    The Commission's regulations address whether we are required to 
prepare an environmental assessment or an environmental impact 
statement. If our rule has ``little or no potential for affecting the 
human environment,'' the rule will be categorically exempted from this 
requirement. 16 CFR 1021.5(c)(1). The proposed rule falls within the 
categorical exemption.

XIV. Executive Order 12988 (Preemption)

    As required by Executive Order 12988 (February 5, 1996), the CPSC 
states the preemptive effect of the proposed rule, as follows:
    The regulation for ROVs is proposed under authority of the CPSA. 15 
U.S.C. 2051-2089). Section 26 of the CPSA provides that ``whenever a 
consumer product safety standard under this Act is in effect and 
applies to a risk of injury associated with a consumer product, no 
State or political subdivision of a State shall have any authority 
either to establish or to continue in effect any provision of a safety 
standard or regulation which prescribes any requirements as the 
performance, composition, contents, design, finish, construction, 
packaging or labeling of such product which are designed to deal with 
the same risk of injury associated with such consumer product, unless 
such requirements are identical to the requirements of the Federal 
Standard''. 15 U.S.C. 2075(a). Upon application to the Commission, a 
state or local standard may be excepted from this preemptive effect if 
the state or local standard: (1) Provides a higher degree of protection 
from the risk of injury or illness than the CPSA standard, and (2) does 
not unduly burden interstate commerce. In addition, the federal 
government, or a state or local government, may establish and continue 
in effect a non-identical requirement that provides a higher degree of 
protection than the CPSA requirement for the hazardous substance for 
the federal, state or local government's use. 15 U.S.C. 2075(b).
    Thus, with the exceptions noted above, the ROV requirements 
proposed in today's Federal Register would preempt non-identical state 
or local requirements for ROVs designed to protect against the same 
risk of injury if the rule is issued in final.

XV. Certification

    Section 14(a) of the CPSA imposes the requirement that products 
subject to a consumer product safety rule under the CPSA, or to a 
similar rule, ban, standard or regulation under any other act enforced 
by the Commission, must be certified as complying with all applicable 
CPSC-enforced requirements. 15 U.S.C. 2063(a). A final rule on ROVs 
would subject ROVs to this certification requirement.

XVI. Effective Date

    The CPSA requires that consumer product safety rules take effect 
not later than 180 days from their promulgation unless the Commission 
finds there is good cause for a later date. 15 U.S.C. 2058(g)(1). The 
Commission proposes that this rule would take effect 180 days after 
publication of the final rule in the Federal Register and would have 
two compliance dates. ROVs would be required to comply with the lateral 
stability and vehicle handling requirements (Sec. Sec.  1411.3 and 
1422.4) 180 days after publication of a final rule in the Federal 
Register. ROVs would be required to comply with the occupant protection 
requirements (Sec.  1422.5) 12 months after publication of a final rule 
in the Federal Register. The requirements would apply to all ROVs 
manufactured or imported on or after the applicable date.
    CPSC believes ROV models that do not comply with the lateral 
stability and vehicle handling requirements can be modified, with 
changes to track width and suspension, in less than 4 person-months (a 
high estimate) and can be tested for compliance in one day. Therefore, 
CPSC believes 180 days is a reasonable time period for manufacturers to 
modify vehicles if necessary, conduct necessary tests, and analyze test 
results to ensure compliance with the lateral stability and vehicle 
handling requirements.
    The Commission is proposing the longer compliance date for the 
occupant protection requirements because we understand that some 
manufacturers will need to redesign and test new prototype vehicles to 
meet these requirements. This design and test process is similar to the 
process that manufacturers use when introducing new model year 
vehicles. We also estimate that it will take approximately 9 person-
months per ROV model to design, test, implement, and begin 
manufacturing vehicles to meet the occupant protection performance

[[Page 69019]]

requirements. Therefore, staff believes that 12 months from publication 
of a final rule would be sufficient time for ROVs to comply with all of 
the proposed requirements.

XVII. Proposed Findings

    The CPSA requires the Commission to make certain findings when 
issuing a consumer product safety standard. Specifically, the CPSA 
requires that the Commission consider and make findings about the 
degree and nature of the risk of injury; the number of consumer 
products subject to the rule; the need of the public for the rule and 
the probable effect on utility, cost, and availability of the product; 
and other means to achieve the objective of the rule, while minimizing 
the impact on competition, manufacturing, and commercial practices. The 
CPSA also requires that the rule must be reasonably necessary to 
eliminate or reduce an unreasonable risk of injury associated with the 
product and issuing the rule must be in the public interest. 15 U.S.C. 
2058(f)(3).
    In addition, the Commission must find that: (1) If an applicable 
voluntary standard has been adopted and implemented, that compliance 
with the voluntary standard is not likely to reduce adequately the risk 
of injury, or compliance with the voluntary standard is not likely to 
be substantial; (2) that benefits expected from the regulation bear a 
reasonable relationship to its costs; and (3) that the regulation 
imposes the least burdensome requirement that would prevent or 
adequately reduce the risk of injury. Id. These findings are discussed 
below.
    Degree and nature of the risk of injury. CPSC received 428 reports 
of ROV-related incidents from the Injury and Potential Injury Incident 
(IPII) and In-Depth Investigation (INDP) databases that occurred 
between January 1, 2003 and December 31, 2011, and were received by 
December 31, 2011. There were a total of 826 victims involved in the 
428 incidents. Among the 428 ROV-related incidents, there were a total 
of 231 reported fatalities and 388 reported injuries. Seventy-five of 
the 388 injuries (19 percent) could be classified as severe; that is, 
the victim has lasting repercussions from the injuries received in the 
incident, based on the information available. The remaining 207 victims 
were either not injured or their injury information was not known. Of 
the 428 ROV-related incidents, 76 involved drivers under 16 years of 
age (18 percent); 227 involved drivers 16 years of age or older (53 
percent); and 125 involved drivers of unknown age (29 percent).
    Using data reported through NEISS from January 1, 2010 to August 
31, 2010, the Commission conducted a special study to identify cases 
that involved ROVs that were reported through NEISS. Based on 
information obtained through the special study, the estimated number of 
emergency department-treated ROV-related injuries occurring in the 
United States between January 1, 2010 and August 31, 2010, is 2,200 
injuries. Extrapolating for the year 2010, the estimated number of 
emergency department-treated ROV-related injuries is 3,000, with a 
corresponding 95 percent confidence interval of 1,100 to 4,900.
    Number of consumer products subject to the rule. Sales of ROVs have 
increased substantially since their introduction. In 1998, only one 
firm manufactured ROVs, and fewer than 2,000 units were sold. By 2003, 
when a second major manufacturer entered the market, almost 20,000 ROVs 
were sold. The only dip in sales occurred around 2008, which coincided 
with the worst of the credit crisis and a recession that also started 
about the same time. In 2013, an estimated 234,000 ROVs were sold by 
about 20 different manufacturers.
    The number of ROVs available for use has also increased 
substantially. Because ROVs are a relatively new product, we do not 
have any specific information on the expected useful life of ROVs. 
However, using the same operability rates that CPSC uses for ATVs, we 
estimate that there were about 570,000 ROVs available for use in 2010. 
By the end of 2013, there were an estimated 1.2 million ROVs in use.
    The need of the public for ROVs and the effects of the rule on 
their utility, cost, and availability.
    Currently there are two varieties of ROVs: Utility and 
recreational. Early ROV models emphasized the utility aspects of the 
vehicles, but the recreational aspects of the vehicles have become very 
popular.
    Regarding the effects of the rule on ROVs utility, according to 
comments on the ANPR provided by several ROV manufacturers, some ROV 
users ``might prefer limit oversteer in the off-highway environment.'' 
To the extent that the requirements in the proposed rule would reduce 
the ability of these users to reach limit oversteer intentionally, the 
proposed rule could have some adverse impact on the utility or 
enjoyment that these users receive from ROVs. These impacts would 
probably be limited to a small number of recreational users who enjoy 
activities or stunts that involve power oversteering or limit 
oversteer.
    Although the impact on consumers who prefer limit oversteer cannot 
be quantified, the Commission expects that the impact will be low. Any 
impact would be limited to consumers who wish to engage intentionally 
in activities involving the loss of traction or power oversteer. The 
practice of power oversteer, such as the speed at which a user takes a 
turn, is the result of driver choice. The proposed rule would not 
prevent ROVs from reaching limit oversteer under all conditions; nor 
would the proposed rule prevent consumers from engaging in these 
activities. At most, the proposed rule might make it somewhat more 
difficult for users to reach limit oversteer in an ROV.
    The seat belt speed limiter requirement could have an effect on 
utility and impose some unquantifiable costs on some users who would 
prefer not to use seat belts. The cost to these users would be the time 
required to buckle and unbuckle their seat belts and any disutility 
cost, such as discomfort caused by wearing the seat belt. We cannot 
quantify these costs because we do not know how many ROV users choose 
not to wear their seat belts; nor do we have the ability to quantify 
any discomfort or disutility that they would experience from wearing 
seat belts. However, the proposed rule does not require that the seat 
belts be fastened unless the vehicle is traveling faster than 15 mph. 
This should serve to mitigate these costs because many people who would 
be inconvenienced or discomforted by the requirement, such as people 
using the vehicle for work or utility purposes, or who must frequently 
get on and off the vehicle, are likely to be traveling at lower speeds.
    The effect of the rule on cost and availability of ROVs is expected 
to be minimal. The average manufacturer's suggested retail prices 
(MSRP) of ROVs, weighted by units sold, was about $13,100 in 2013, with 
a range of about $3,600 to $20,100. The Commission estimates the per-
unit cost to ROVs of the rule to be $61 to $94. Because this per-unit 
cost resulting from the rule is a very small percentage of the overall 
retail price of an ROV, it is unlikely that the rule would have much of 
an effect on the cost or availability of ROVs.
    Other means to achieve the objective of the rule, while minimizing 
the impact on competition and manufacturing. The Commission does not 
believe the rule will have adverse impact on competition. The 
preliminary regulatory analysis estimates the per-unit cost to ROVs of 
the rule to be $61 to $94. The average manufacturer's suggested retail 
prices (MSRP) of ROVs, weighted by

[[Page 69020]]

units sold, was about $13,100 in 2013, with a range of about $3,600 to 
$20,100. The per-unit cost resulting from the rule is a very small 
percentage of the overall retail price of an ROV. With such a 
relatively low impact, it is unlikely that ROV companies would withdraw 
from the market or that the number of ROV models will be affected. 
Therefore, the preliminary regulatory analysis supports a finding that 
the proposed rule is unlikely to have an impact on competition.
    The Commission believes that some, but not all, ROV models already 
meet the rule's requirement that the speed of the vehicle be limited if 
the driver's seat belt is not fastened. Before implementing any changes 
to their vehicles to meet the requirement, manufacturers whose ROVs do 
not meet the seatbelt speed limiter requirement would have to analyze 
their options for meeting the requirement. This process would include 
developing prototypes of system designs, testing the prototypes, and 
refining the design of the systems based on this testing. Once the 
manufacturer has settled on a system for meeting the requirement, the 
system will have to be incorporated into the manufacturing process of 
the vehicle. This will involve producing the engineering specifications 
and drawings of the system, parts, assemblies, and subassemblies that 
are required. Manufacturers will need to obtain the needed parts from 
their suppliers and incorporate the steps needed to install the system 
on the vehicles in the assembly line. The Commission believes that 
manufacturers should be able to complete activities related to meeting 
the lateral stability and handling requirements within 180 days after 
publication of the final rule and activities related to meeting the 
occupant protection requirements within 12 months after publication of 
the final rule. The Commission's proposed effective date of 12 months 
for the occupant protection requirements may reduce the impact of the 
proposed requirements on manufacturing.
    Unreasonable risk. CPSC received 428 reports of ROV-related 
incidents from the Injury and Potential Injury Incident (IPII) and In-
Depth Investigation (INDP) databases that occurred between January 1, 
2003 and December 31, 2011, and were received by December 31, 2011. 
There were a total of 826 victims involved in the 428 incidents. Among 
the 428 ROV-related incidents, there were a total of 231 reported 
fatalities and 388 reported injuries. Seventy-five of the 388 injuries 
(19 percent) could be classified as severe; that is, the victim has 
lasting repercussions from the injuries received in the incident based 
on the information available.
    The estimated cost and benefits of the rule on an annual basis can 
be calculated by multiplying the estimated benefits and costs per unit 
by the number of ROVs sold in a given year. In 2013, 234,000 ROVs were 
sold. If the proposed rule had been in effect that year, the total 
quantifiable cost would have been between $14.3 million and $225.0 
million ($61 and $94 multiplied by 234,000 units, respectively). The 
total quantifiable benefits would have been at least $515 million 
($2,199 x 234,000). Of the benefits, about $453 million (or about 88 
percent) would have resulted from the reduction in fatal injuries, and 
about $62 million (or about 12 percent) of the benefits would have 
resulted from a reduction in the societal cost of nonfatal injuries. 
The reduction in the societal cost of nonfatal injuries, which amounts 
to about $47 million, would represent a reduction in pain and 
suffering. The Commission concludes preliminarily that ROVs pose an 
unreasonable risk of injury and finds that the proposed rule is 
reasonably necessary to reduce that unreasonable risk of injury.
    Public interest. This proposed rule is intended to address 
identified aspects of ROVs, ROV design, and ROV use, which are believed 
to contribute to ROV deaths and injuries, with a goal of reducing such 
incidents. The CPSC believes that adherence to the requirements of the 
proposed rule will reduce ROV deaths and injuries in the future; thus 
the rule is in the public interest. Specifically, the Commission 
believes that improving lateral stability (by increasing rollover 
resistance) and improving vehicle handling (by correcting oversteer to 
understeer) are the most effective approaches to reducing the 
occurrence of ROV rollover incidents. ROVs with higher lateral 
stability are less likely to roll over because more lateral force is 
necessary to cause rollover. ROVs exhibiting understeer during a turn 
are also less likely to roll over because lateral acceleration 
decreases as the path of the ROV makes a wider turn, and the vehicle is 
more stable if a sudden change in direction occurs.
    Furthermore, the Commission believes that when rollovers do occur, 
improving occupant protection performance (by increasing seat belt use) 
will mitigate injury severity. CPSC analysis of ROV incidents indicates 
that 91 percent of fatally ejected victims were not wearing a seat belt 
at the time of the incident. Increasing seat belt use, in conjunction 
with better shoulder retention performance, will significantly reduce 
injuries and deaths associated with an ROV rollover event.
    In summary, the Commission finds preliminarily that promulgating 
the proposed rule is in the public interest.
    Voluntary standards. The Commission is aware of two voluntary 
standards that are applicable to ROVs, ANSI/ROHVA 1, American National 
Standard for Recreational Off-Highway Vehicles, and ANSI/B71.9, 
American National Standard for Multipurpose Off-Highway Utility 
Vehicles. As described previously in detail in the preamble, the 
Commission believes that the current voluntary standard requirements do 
not adequately reduce the risk of injury or death associated with ROVs. 
Neither voluntary standard requires that ROVs understeer, as required 
by the proposed rule. Based on testing and experience with the Yamaha 
Rhino repair program, the Commission believes that drivers are more 
likely to lose control of vehicles that oversteer, which can lead to 
the vehicle rolling over or to other types of accidents.
    Both voluntary standards have requirements that are intended to set 
standards for dynamic lateral stability. ANSI/ROHVA 1-2011 uses a turn-
circle test for dynamic lateral stability. That is more similar to the 
test in the proposed rule for determining whether the vehicle 
understeers, than it is to the test for dynamic lateral stability. The 
dynamic stability requirement in ANSI/OPEI B71.9-2012 uses a J-turn 
test, like the proposed rule, but measures different variables during 
the test and uses a different acceptance criterion. The Commission does 
not believe that the tests procedures in either standard have been 
validated properly as being capable of providing useful information 
about the dynamic stability of the vehicle. Moreover, the voluntary 
standards would find some vehicles acceptable, even though their 
lateral acceleration at rollover is less than 0.70 g, which is the 
acceptance criterion in the proposed rule.
    Both voluntary standards require that manufacturers include a 
lighted seat-belt reminder that is visible to the driver and that 
remains on for at least 8 seconds after the vehicle is started, unless 
the driver's seatbelt is fastened. However, virtually all ROVs on the 
market already include this feature, and therefore, relying only on the 
voluntary standards would not be expected to raise seatbelt use over 
its current level.
    The voluntary standards include requirements for retaining the 
occupant within the protective zone of the vehicle in the event of a 
rollover, including two options for restraining the occupants in the 
shoulder/hip area. However, testing

[[Page 69021]]

performed by CPSC identified weaknesses in the performance-based tilt 
table test option that allows unacceptable occupant head ejection 
beyond the protective zone of the vehicle Rollover Protective Structure 
(ROPS). CPSC testing indicated that a passive shoulder barrier could 
reduce the head excursion of a belted occupant during quarter-turn 
rollover events. The Commission believes that this can be accomplished 
by a requirement for a passive barrier based on the dimensions of the 
upper arm of a 5th percentile adult female, at a defined area near the 
ROV occupants' shoulder, as contained in the proposed rule.
    Relationship of benefits to costs. The estimated costs and benefits 
of the rule on an annual basis can be calculated by multiplying the 
estimated benefits and costs per unit, by the number of ROVs sold in a 
given year. In 2013, 234,000 ROVs were sold. If the proposed rule had 
been in effect that year, the total quantifiable cost would have been 
between $14.3 million and $22.0 million ($61 and $94 multiplied by 
234,000 units, respectively). The total quantifiable benefits would 
have been at least $515 million ($2,199 x 234,000).
    On a per-unit basis, we estimate the total cost of the proposed 
rule to be $61 to $94 per vehicle. We estimate the total quantifiable 
benefits of the proposed rule to be $2,199 per unit. This results in 
net quantifiable benefits of $2,105 to $2,138 per unit. Quantifiable 
benefits of the proposed rule could exceed the estimated $1,329 per 
unit because the benefit associated with the vehicle handling and 
lateral stability requirement could not be quantified.
    Based on this analysis, the Commission finds preliminarily that the 
benefits expected from the rule bear a reasonable relationship to the 
anticipated costs of the rule.
    Least burdensome requirement. The Commission considered less-
burdensome alternatives to the proposed rule on ROVs, but we concluded 
that none of these alternatives would adequately reduce the risk of 
injury:
    (1) Not issuing a mandatory rule, but instead relying upon 
voluntary standards. If CPSC did not issue a mandatory standard, most 
manufacturers would comply with one of the two voluntary standards that 
apply to ROVs. As discussed previously, the Commission does not believe 
either voluntary standard adequately addresses the risk of injury and 
death associated with ROVs.
    (2) Including the dynamic lateral stability requirement or the 
understeer requirement, but not both. The Commission believes that both 
of these characteristics need to be addressed. A vehicle that meets 
both the dynamic stability requirement and the understeer requirement 
should be safer than a vehicle that meets only one of the requirements. 
Moreover, the cost of meeting just one requirement is not substantially 
lower than the cost of meeting both requirements. The cost of testing a 
vehicle for compliance with both the dynamic lateral stability and 
vehicle handling/understeer requirement was estimated to be about 
$24,000. However, the cost of testing for compliance with just the 
dynamic stability requirement would be about $20,000, or only about 17 
percent less than the cost of testing for compliance with both 
requirements. This is because the cost of renting and transporting the 
vehicle to the test site, instrumenting the vehicle for the tests, and 
making some initial static measurements are virtually the same for both 
requirements and would only have to be done once if the tests for both 
requirements were conducted on the same day. Moreover, changes in the 
vehicle design that affect the lateral stability of the vehicle could 
also impact the handling of the vehicle. For these reasons, the 
proposed rule includes both a dynamic stability and vehicle handling 
requirement.
    (3) Instead of seatbelt/speed limitation requirements in the 
proposed rule, the Commission considered a requirement for ROVs to have 
loud or intrusive seatbelt reminders. Currently, most ROVs meet the 
voluntary standards that require an 8-second visual seatbelt reminder. 
Some more intrusive systems have been used on passenger cars. For 
example, the Ford ``BeltMinder'' system resumes warning the driver 
after about 65 seconds if his or her seatbelt is not fastened and the 
car is traveling at more than 3 mph. The system flashes a warning light 
and sounds a chime for 6 seconds every 30 seconds for up to 5 minutes 
as long as the car is operating and the driver's seatbelt is not 
fastened. Honda developed a similar system in which the warning could 
last for longer than 9 minutes if the driver's seatbelt is not 
fastened. Studies of both systems found that a statistically 
significant increase in the use seatbelts of 5 percent (from 71 to 76 
percent) and 6 percent (from 84 to 90 percent), respectively.
    However, these more intrusive seatbelt warning systems are unlikely 
to be as effective as the seatbelt speed limitation requirement in the 
proposed rule. The Commission believes that the seatbelt speed 
limitation requirement will cause most drivers and passengers who 
desire to exceed 15 mph to fasten their seatbelts. Research supports 
this position. One experiment used a haptic feedback system to increase 
the force the driver needed to exert to depress the gas pedal when the 
vehicle exceeded 25 mph if the seatbelt was not fastened. The system 
did not prevent the driver from exceeding 25 mph, but the system 
increased the amount of force required to depress the gas pedal to 
maintain a speed greater than 25 mph. In this experiment, all seven 
participants chose to fasten their seatbelts. A follow-up study on the 
haptic feedback study focused on 20 young drivers ranging in age from 
18 to 21, and a feedback force set at 20 mph instead of 25 mph. The 
study results showed that the mean seat belt use increased from 54.7 
percent to 99.7 percent, and the few instances in which seat belts were 
not worn were on trips of 2 minutes long or less. Most significantly, 
participants rated the system as very acceptable and agreeable (9 out 
of a 10-point scale).
    The more intrusive seatbelt reminder systems used on some passenger 
cars have been more limited in their effectiveness. The Honda system, 
for example, reduced the number of unbelted drivers by about 38 
percent; the Ford system reduced the number of unbelted drivers by only 
17 percent. (The Honda system increased seatbelt use from 84 percent to 
90 percent. Therefore, the percentage of unbelted drivers was reduced 
by about 38 percent, or 6 percent divided by 16 percent. The Ford 
system increased seatbelt use from 71 percent to 76 percent. Therefore, 
the percentage of unbelted drivers was reduced by about 17 percent, or 
5 percent divided by 29 percent.) Additionally, ROVs are open vehicles 
and the ambient noise is likely higher than in the enclosed passenger 
compartment of a car. It is likely that some ROV drivers would not hear 
the warning, and therefore, they would be motivated to fasten their 
seatbelts, unless the warning was substantially louder than the systems 
used in passenger cars. Therefore, the Commission believes that the 
loud or intrusive seat belt reminders would not be as effective as the 
seat belt speed limiter requirement.
    For the reasons set forth above, the Commission finds preliminarily 
that the rule imposes the least burdensome requirement that prevents or 
adequately reduces the risk of injury for which promulgation of the 
rule is proposed.

XVIII. Request for Comments

    We invite all interested persons to submit comments on any aspect 
of the proposed rule. In particular, the Commission invites comments 
regarding the estimates used in the

[[Page 69022]]

preliminary regulatory analysis and the assumptions underlying these 
estimates. The Commission is especially interested in data that would 
help the Commission to refine its estimates to more accurately reflect 
the expected costs of the proposed rule as well as any alternate 
estimates that interested parties can provide. The Commission is also 
interested in comments addressing whether the proposed compliance dates 
of 180 days after the publication of the final rule to meet the lateral 
stability and vehicle handling requirements and 12 months after the 
publication of the final rule to meet the occupant protection 
requirements are appropriate. The Commission also seeks comments on the 
following:
     Additional key issues related to seatbelts for ROVs, 
including: available technology to prevent any hazards from the 
application of a passenger seatbelt requirement (such as sudden speed 
reductions if a passenger unbuckles); whether CPSC should extend the 
phase-in period for the seat-belt requirement; and any other relevant 
information related to the proposed seatbelt requirements.
     Whether CPSC should allow the use of doors or other 
mechanisms capable of meeting specified loading criteria to meet the 
shoulder restraint requirement.
     Whether there are further consistent and repeatable 
testing requirements that should be added to the proposed rule that 
would capture off-road conditions drivers experience in ROVs. If so, 
set forth the specifics of such further requirements.
     Whether CPSC should establish separate requirements for 
utility vehicles, including: definitions, scope, additional standards, 
and/or exemptions that would be suitable for requirements specific to 
utility vehicles.
The Commission seeks comment, data testing parameters and testing 
results concerning:
     Oversteer and understeer, dynamically unstable handling, 
and minimal path-following capabilities; and
     Whether there is a need for supplemental criteria in 
addition to specific lateral stability acceleration limits to avoid 
potential unintended consequences of a single criterion.

The public is invited to submit additional information about any other 
issues that stakeholders find relevant. Comments should be submitted in 
accordance with the instructions in the ADDRESSES section at the 
beginning of this notice.

XIV. Conclusion

    For the reasons stated in this preamble, the Commission proposes 
requirements for lateral stability, vehicle handing, and occupant 
protection to address an unreasonable risk of injury associated with 
ROVs.

List of Subjects in 16 CFR Part 1422

    Consumer protection, Imports, Information, Labeling, Recreation and 
Recreation areas, Incorporation by reference, Safety.

    For the reasons discussed in the preamble, the Commission proposes 
to amend Title 16 of the Code of Federal Regulations as follows:

0
1. Add part 1422 to read as follows:

PART 1422--SAFETY STANDARD FOR RECREATIONAL OFF-HIGHWAY VEHICLES

Sec.
1422.1 Scope, purpose and compliance dates.
1422.2 Definitions.
1422.3 Requirements for dynamic lateral stability.
1422.4 Requirements for vehicle handling.
1422.5 Requirements for occupant protection performance.
1422.6 Prohibited stockpiling.
1422.7 Findings.

    Authority:  15 U.S.C. 2056, 2058 and 2076.


Sec.  1422.1  Scope, purpose and compliance dates.

    (a) This part 1422, a consumer product safety standard, establishes 
requirements for recreational off-highway vehicles (ROVs), as defined 
in Sec.  1422.2(a). The standard includes requirements for dynamic 
lateral, vehicle handling, and occupant protection. These requirements 
are intended to reduce an unreasonable risk of injury and death 
associated with ROVs.
    (b) This standard does not apply to the following vehicles, as 
defined by the relevant voluntary standards:
    (1) Golf carts
    (2) All-terrain vehicles
    (3) Fun karts
    (4) Go karts
    (5) Light utility vehicles
    (c) Any ROV manufactured or imported on or after [date that is 180 
days after publication of a final rule] shall comply with the lateral 
stability requirements stated in Sec.  1422.3 and the vehicle handling 
requirements stated in Sec.  1422.4. Any ROV manufactured or imported 
on or after [date that is 12 months after publication of final rule] 
shall comply with the occupant protection requirements stated in Sec.  
1422.5.


Sec.  1422.2  Definitions.

    In addition to the definitions in section 3 of the Consumer Product 
Safety Act (15 U.S.C. 2051), the following definitions apply for 
purposes of this part 1422.
    (a) Recreational off-highway vehicle (ROV) means a motorized 
vehicle designed for off-highway use with the following features: Four 
or more wheels with pneumatic tires; bench or bucket seating for two or 
more people; automotive-type controls for steering, throttle, and 
braking; rollover protective structure (ROPS); occupant restraint; and 
maximum speed capability greater than 30 mph.
    (b) Two-wheel lift means the point at which the inside wheels of a 
turning vehicle lift off the ground, or when the uphill wheels of a 
vehicle on a tilt table lift off the table. Two-wheel lift is a 
precursor to a rollover event. We use this term interchangeably with 
the term ``tip-up.''
    (c) Threshold lateral acceleration means the minimum lateral 
acceleration of the vehicle at two-wheel lift.


Sec.  1422.3  Requirements for dynamic lateral stability.

    (a) General. The Recreational Off-Highway Vehicle (ROV) requirement 
for lateral stability is based on the average threshold lateral 
acceleration at rollover, as determined by a 30 mph dropped throttle J-
turn test. This threshold lateral acceleration is measured parallel to 
the ground plane at the center of gravity (CG) of the loaded test 
vehicle and occurs at the minimum steering wheel angle required to 
cause the vehicle to roll over in a 30 mph dropped throttle J-turn test 
on a flat and level, high-friction surface. Rollover is achieved when 
all of the wheels of the ROV that are on the inside of the turn lift 
off the ground. For convenience, this condition is referred to as two-
wheel lift, regardless of the number of wheels on the ROV. Testing 
shall be conducted on a randomly selected representative production 
vehicle.
    (b) Test surface. Tests shall be conducted on a smooth, dry, 
uniform, paved surface constructed of asphalt or concrete. The surface 
area used for dynamic testing shall be kept free of debris and 
substances that may affect test results during vehicle testing.
    (1) Friction. Surface used for dynamic testing shall have a peak 
braking coefficient greater than or equal to 0.90 and a sliding skid 
coefficient greater than or equal to 0.80 when measured in accordance 
with ASTM E 1337, Standard Test Method for Determining Longitudinal 
Peak Braking Coefficient of Paved Surfaces Using Standard

[[Page 69023]]

Reference Tire, approved December 1, 2012, and ASTM E274, Standard Test 
Method for Skid Resistance of Paved Surfaces Using a Full-Scale Tire, 
approved January 2011, respectively. The Director of the Federal 
Register approves these incorporations by reference in accordance with 
5 U.S.C. 552(a) and 1 CFR part 51. You may obtain a copy from ASTM 
International, 100 Bar Harbor Drive, P.O. Box 0700, West Conshohocken, 
PA 19428; https://www.astm.org/cpsc.htm. You may inspect a copy at the 
Office of the Secretary, U.S. Consumer Product Safety Commission, Room 
820, 4330 East West Highway, Bethesda, MD 20814, telephone 301-504-
7923, or at the National Archives and Records Administration (NARA). 
For information on the availability of this material at NARA, call 202-
741-6030, or go to: https://www.archives.gov/federal_register/code_of_federalregulations/ibr_locations.html.
    (2) Slope. The test surface shall have a slope equal to or less 
than 1 degree (1.7% grade).
    (3) Ambient conditions. The ambient temperature shall be between 0 
degrees Celsius (32 [ordm] Fahrenheit) and 38 [ordm]C (100 [ordm]F). 
The maximum wind speed shall be no greater than 16 mph (7 m/s).
    (c) Test conditions. (1) Vehicle condition. An ROV used for dynamic 
testing shall be configured in the following manner:
    (i) The test vehicle shall be a representative production vehicle. 
The ROV shall be in standard condition. Adjustable seats shall be 
located in the most rearward position.
    (ii) The ROV shall be operated in two-wheel drive mode, with 
selectable differential in its most-open setting. The tires shall be 
the manufacturer's original-equipment tires intended for normal retail 
sale to consumers. The tires shall be new when starting the tests, then 
broken-in by conducting a minimum total of ten J-turns with five in the 
right-turning direction and five in the left-turning direction. The J-
turns conducted for tire break-in shall be conducted at 30 mph and 
steering angles sufficient to cause two-wheel lift.
    (iii) Springs or shocks that have adjustable spring or damping 
rates shall be set to the manufacturer's recommended settings for 
delivery.
    (iv) Tires shall be inflated to the ROV manufacturer's recommended 
settings for normal operation for the load condition specified in 
paragraph (c)(vi) of this section. If more than one pressure is 
specified, the lowest value shall be used.
    (v) All vehicle operating fluids shall be at the manufacturer's 
recommended level, and the fuel tank shall be full to its rated 
capacity.
    (vi) The ROV shall be loaded, such that the combined weight of the 
test operator, test equipment, and ballast, if any, shall equal 430 
lbs.  11 lbs. (195 kg  5 kg).
    (vii) The center of gravity (CG) of the equipped test vehicle shall 
be no more than 0.5 inch below (and within 1.0 inch in the x-axis and 
y-axis directions) the CG of the vehicle as it is sold at retail and 
loaded according to paragraph (c)(vi) of this section.
    (2) Vehicle test equipment. (i) Safety equipment. Test vehicles 
shall be equipped with outriggers on both sides of the vehicle. The 
outriggers shall be designed to minimally affect the loaded vehicle's 
center of gravity location, shall permit the vehicle to experience two-
wheel lift during dynamic testing, and shall be capable of preventing a 
full vehicle rollover.
    (ii) Steering controller. The test vehicle shall be equipped with a 
programmable steering controller (PSC), capable of responding to 
vehicle speed, with a minimum steering angle input rate of 500 degrees 
per second, and accurate within + 0.25 degree. The steering wheel 
setting for 0.0 degrees of steering angle is defined as the setting 
which controls the properly aligned vehicle to travel in a straight 
path on a level surface. The PSC shall be operated in absolute steering 
mode, where the amount of steering used for each test shall be measured 
relative to the PSC reading when the vehicle steering is at zero 
degrees.
    (iii) Vehicle instrumentation. The vehicle shall be instrumented to 
record lateral acceleration, vertical acceleration, longitudinal 
acceleration, forward speed, steering wheel angle, steering wheel angle 
rate, vehicle roll angle, roll angle rate, pitch angle rate, and yaw 
angle rate. See Table 1 for instrumentation specifications. Ground 
plane lateral acceleration shall be calculated by correcting the body-
fixed acceleration for roll angle. A roll motion inertia measurement 
sensor that provides direct output of ground plane lateral acceleration 
at the vehicle CG may also be used in lieu of manual correction to 
obtain ground plane lateral acceleration. Roll angle may be calculated 
from roll rate data.

  Table 1--Instrumentation Specification For J-Turn and Constant Radius
                             Testing of ROVs
------------------------------------------------------------------------
                 Parameter                            Accuracy
------------------------------------------------------------------------
Vehicle Speed.............................   0.10 mph
Acceleration (x, y, and z directions )....   0.003 g
Steering Wheel Angle......................   0.25 deg.
Steering Wheel Angle Rate.................   0.5 deg./sec.
Pitch, Roll, and Yaw Rates................   0.10 deg./sec.
Roll Angle*...............................   0.20 deg.
------------------------------------------------------------------------
* For constant radius testing, roll angle must be measured directly or
  roll rate accuracy must be  0.01 deg./sec.

    (d) Test procedure. (1) 3.3.1. Set the vehicle drive train in its 
most-open setting. For example, two-wheel drive shall be used instead 
of four-wheel drive, and a lockable differential, if so equipped, shall 
be in its unlocked, or ``open,'' setting.
    (2) Drive the vehicle in a straight path to define zero degree 
(0.0) steer angle.
    (3) Program the PSC to input a 90-degree turn to the right at a 
minimum of 500 degrees per second as soon as the vehicle slows to 30 
mph. Program the PSC to hold steering angles for a minimum of 4 seconds 
before returning to zero steer angle. The steering rate when returning 
to zero may be less than 500 degrees per second.
    (4) Conduct a 30 mph dropped throttle J-turn.
    (i) Accelerate the vehicle in a straight line to a speed greater 
than 30 mph.
    (ii) As the vehicle approaches the desired test location, engage 
the PSC and fully release the throttle.
    (iii) The PSC shall input the programmed steering angle when the 
vehicle decelerates to 30 mph. Verify that the instrumentation recorded 
all of the data during this J-turn event.
    (5) Conduct additional J-turns, increasing the steer angle in 10-
degree increments, as required, until a two-wheel lift event is 
visually observed.
    (6) Conduct additional J-turns, decreasing the steering angle in 5-
degree increments to find the lowest steering angle that will produce 
two-wheel lift. Additional adjustments, up or down, in 1-degree 
increments may be used.
    (7) Repeat the process of conducting J-turns to determine minimum 
steer angle to produce two-wheel lift in left turn direction.
    (8) Start the data acquisition system.
    (9) Conduct J-turn test trials in the left and right directions 
using the minimum steering angles determined in paragraphs (d)(6) and 
(d)(7) of this section to verify that the steering angle

[[Page 69024]]

produces two-wheel lift in both directions.
    (10) Conduct five J-turn test trials with two-wheel lift in the 
left and right turn directions in one direction heading on the test 
surface (10 total trials). On the same test track, but in the opposite 
heading on the test surface, conduct five more J-turn test trials with 
two-wheel lift in the left and right turn directions (10 total trials). 
A minimum data set will consist of 20 total J-turn test trials with 
half of the tests conducted in one direction on the test surface and 
half of the tests conducted in the opposite direction. Review all data 
parameters for each trial to verify that the tests were executed 
correctly. Any trials that do not produce two-wheel lift should be 
diagnosed for cause. If cause is identified, discard the data and 
repeat the trial to replace the data. If no cause can be identified, 
repeat actions stated in paragraphs (d)(5) through (d)(7) of this 
section to ensure that the correct steering angle has been determined. 
Additional J-turn tests may be added to the minimum data set in groups 
of four, with one test for each left/right turn direction and one test 
for each direction heading on the test surface.
    (11) Determine value of threshold lateral acceleration at rollover.
    (i) Data recorded as required in paragraph (d)(10) of this section 
shall be digitally low-pass filtered to 2.0 hertz, using a phaseless, 
eighth-order, Butterworth filter to eliminate noise artifacts on the 
data.
    (ii) Plot the data for ground plane lateral acceleration corrected 
to the test vehicle CG location, steering wheel angle, and roll angle 
recorded for each trial conducted under paragraph (d)(10) of this 
section.
    (iii) Find and record the peak ground plane lateral acceleration 
occurring between the time of the PSC input and the time of two-wheel 
lift.
    (iv) If a body-fixed acceleration sensor is used, correct the 
lateral acceleration data for roll angle, using the equation:

Ay ground = Ay cos [Phi]-Az sin [Phi]
([Phi] = vehicle body roll angle)

    (v) Calculate the threshold lateral acceleration at rollover value, 
which is the average of the peak values for ground plane lateral 
acceleration for all of the trials conducted under paragraph (d)(10) of 
this section that produced two-wheel lift.
    (e) Performance requirements. The minimum value for the threshold 
lateral acceleration at rollover shall be 0.70 g or greater.
    (f) Consumer information requirements. The manufacturer shall 
provide a hang tag with every ROV that is visible to the driver and 
provides the value of the threshold lateral acceleration at rollover of 
that model vehicle. The label must conform in content, form, and 
sequence to the hang tag shown in Figure 1.
    (1) Size. Every hang tag shall be at least 6 inches (152 mm) wide x 
4 inches (102 mm) tall.
    (2) Content. Every hang tag shall contain the following:
    (i) Value of the threshold lateral acceleration at rollover of that 
model vehicle displayed on a progressive scale.
    (ii) The statement--``Compare with other vehicles before you buy.''
    (iii) The statement--``The value above is a measure of this 
vehicle's resistance to rolling over on a flat surface. Vehicles with 
higher numbers are more stable.''
    (iv) The statement--``Other vehicles may have a higher rollover 
resistance; compare before you buy.''
    (v) The statement--``Rollover cannot be completely eliminated for 
any vehicle.''
    (vi) The statement--``Lateral acceleration is measured during a J-
turn test; minimally accepted value is 0.7 g.''
    (vii) The manufacturer's name and vehicle model, e.g., XYZ 
corporation, Model x, ####.
    (3) Format. The hang tag shall be formatted as shown in Figure 1.
    (4) Attachment. Every hang tag shall be attached to the ROV and 
conspicuous to the seated driver.

[[Page 69025]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.017

Sec.  1422.4  Requirements for vehicle handling.

    (a) General. The ROV requirement for vehicle handling shall be 
based on the vehicle's steering gradient, as measured by the constant 
radius test method described in SAE Surface Vehicle Recommended 
Practice J266, published January 1996. The Director of the Federal 
Register approves this incorporation by reference in accordance with 5 
U.S.C. 552(a) and 1 CFR part 51. You may obtain a copy from ASTM 
International, 100 Bar Harbor Drive, P.O. Box 0700, West Conshohocken, 
PA 19428; https://www.astm.org/cpsc.htm. You may inspect a copy at the 
Office of the Secretary, U.S. Consumer Product Safety Commission, Room 
820, 4330 East West Highway, Bethesda, MD 20814, telephone 301-504-
7923, or at the National Archives and Records Administration (NARA). 
For information on the availability of this material at NARA, call 202-
741-6030, or go to: https://www.archives.gov/federal_register/code_of_federalregulations/ibr_locations.html.
    (b) Test surface. Tests shall be conducted on a smooth, dry, 
uniform, paved surface constructed of asphalt or concrete. The surface 
area used for dynamic testing shall be kept free of debris and 
substances that may affect test results during vehicle testing.
    (1) Friction. Surface used for dynamic testing shall have a peak 
braking coefficient greater than or equal to 0.90 and a sliding skid 
coefficient greater than or equal to 0.80 when measured in accordance 
with ASTM E 1337 and ASTM E274, respectively.
    (2) Slope. The test surface shall have a slope equal to or less 
than 1 degree (1.7% grade).
    (3) Ambient conditions. The ambient temperature shall be between 0 
degrees Celsius (32 [ordm] Fahrenheit) and 38 [ordm]C (100 [ordm]F). 
The maximum wind speed shall be no greater than 16 mph (7 m/s).
    (c) Test conditions.--(1) Vehicle condition. A vehicle used for 
dynamic testing shall be configured in the following manner. (i) The 
test vehicle shall be a representative production vehicle. The ROV 
shall be in standard condition. Adjustable seats shall be located in 
the most rearward position.
    (ii) The ROV shall be operated in two-wheel drive mode with 
selectable differential in its most-open setting. The tires shall be 
the manufacturer's original-equipment tires intended for normal retail 
sale to consumers. The tires shall be new when starting the tests, then 
broken-in by conducting a minimum total of ten J-turns with five in the 
right-turning direction and five in the left-turning direction. The J-
turns conducted for tire break-in shall be conducted at 30 mph and 
steering angles sufficient to cause two-wheel lift. Tires used for the 
full test protocol to establish the threshold lateral acceleration at 
rollover value for the test vehicle are acceptable for use in the 
handling performance test protocol.
    (iii) Springs or shocks that have adjustable spring or damping 
rates shall be set to the manufacturer's recommended settings for 
delivery.
    (iv) Tires shall be inflated to the ROV manufacturer's recommended 
settings for normal operation for the load condition specified in 
paragraph (c)(vi) of this section. If more than one pressure is 
specified, the lowest value shall be used.
    (v) All vehicle operational fluids shall be at the manufacturer's 
recommended level and the fuel tank shall be full to its rated 
capacity.
    (vi) The ROV shall be loaded, such that the combined weight of the 
test operator, test equipment, and ballast, if any, shall equal 430 
lbs.  11 lbs. (195 kg  5 kg).
    (vii) The center of gravity (CG) of the equipped test vehicle shall 
be no more than 0.5 inch below (and within 1.0

[[Page 69026]]

inch in the x-axis and y-axis directions) the CG of the vehicle as it 
is sold at retail and loaded according to paragraph (c)(vi) of this 
section.
    (2) Vehicle test equipment. Test vehicles shall be equipped with 
outriggers on both sides of the vehicle. The outriggers shall be 
designed to minimally affect the loaded vehicle's center of gravity 
location, shall permit the vehicle to experience two-wheel lift during 
dynamic testing, and shall be capable of preventing a full vehicle 
rollover.
    (ii) Vehicle instrumentation. The vehicle shall be instrumented to 
record lateral acceleration, vertical acceleration, longitudinal 
acceleration, forward speed, steering wheel angle, steering wheel angle 
rate, vehicle roll angle, roll angle rate, pitch angle rate, and yaw 
angle rate. See Table 1 in Sec.  1422.3(c) for instrumentation 
specifications. Ground plane lateral acceleration shall be calculated 
by correcting the body-fixed acceleration for roll angle. A roll motion 
inertia measurement sensor that provides direct output of ground plane 
lateral acceleration at the vehicle CG may also be used in lieu of 
manual correction to obtain ground plane lateral acceleration.
    (d) Test Procedure. (1) Handling performance testing shall be 
conducted using the constant radius test method described in SAE 
Surface Vehicle Recommended Practice J266. The minimum radius for 
constant-radius testing shall be 100 feet. In this test method, the 
instrumented and loaded vehicle is driven while centered on a 100-ft. 
radius circle marked on the test surface, with the driver making every 
effort to maintain the vehicle path relative to the circle. The vehicle 
is operated at a variety of increasing speeds, and data are recorded 
for those various speed conditions to obtain data to describe the 
vehicle handling behavior across the prescribed range of ground plane 
lateral accelerations. Data shall be recorded for the lateral 
acceleration range from 0.0 g to 0.5 g.
    (2) Start the data acquisition system.
    (3) Drive the vehicle on the circular path at the lowest possible 
speed. Data shall be recorded with the steering wheel position and 
throttle position fixed to record the approximate Ackermann angle.
    (4) Continue driving the vehicle to the next speed at which data 
will be taken. The vehicle speed shall be increased and data shall be 
taken until it is no longer possible for the driver to maintain 
directional control of the vehicle. Test shall be repeated at least 
three times so that results can be examined for repeatability and then 
averaged.
    (5) Data collection, method 1--discrete data points. In this data 
acquisition method, the driver maintains a constant speed while 
maintaining compliance with the circular path, and data points are 
recorded when a stable condition of speed and steering angle is 
achieved. After the desired data points are recorded for a given speed, 
the driver accelerates to the next desired speed setting, maintains 
constant speed and compliance with the path, and data points are 
recorded for the new speed setting. This process is repeated to cover 
the speed range from 0.0 mph to 28 mph, which will map the lateral 
acceleration range from near 0.0 g to 0.50 g. Increments of speed shall 
be 1 to 2 miles per hour, to allow for a complete definition of the 
understeer gradient. Data shall be taken at the lowest speed 
practicable to obtain an approximation of the vehicle's Ackermann 
steering angle.
    (6) Data collection, method 2--continuous data points In this data 
acquisition method, the driver maintains compliance with the circular 
path while slowly increasing vehicle speed; and data from the vehicle 
instrumentation is recorded continuously, so long as the vehicle 
remains centered on the intended radius. The rate of speed increase 
shall not exceed 0.93 mph per second. Initial speed shall be as low as 
is practicable, in order to obtain an approximation of the vehicle's 
Ackermann steering angle. The speed range shall be 0.0 mph to 28.0 mph, 
which will be sufficient to produce corrected lateral accelerations 
from near 0.0 g to 0.50 g.
    (7) Vehicle dimension coordinate system. The coordinate system 
described in SAE Surface Vehicle Recommended Practice J670, published 
in January 2008, shall be used. The Director of the Federal Register 
approves this incorporation by reference in accordance with 5 U.S.C. 
552(a) and 1 CFR part 51. You may obtain a copy from ASTM 
International, 100 Bar Harbor Drive, P.O. Box 0700, West Conshohocken, 
PA 19428; https://www.astm.org/cpsc.htm. You may inspect a copy at the 
Office of the Secretary, U.S. Consumer Product Safety Commission, Room 
820, 4330 East West Highway, Bethesda, MD 20814, telephone 301-504-
7923, or at the National Archives and Records Administration (NARA). 
For information on the availability of this material at NARA, call 202-
741-6030, or go to: https://www.archives.gov/federal_register/code_of_federalregulations/ibr_locations.html.
    (8) Data analysis. The lateral acceleration data shall be corrected 
for roll angle using the method described in Sec.  1422.3(11)(iv). To 
provide uniform and comparable data, the ground plane lateral 
acceleration shall also be corrected to reflect the value at the test 
vehicle's center of gravity. The data shall be digitally low-pass 
filtered to 1.0 Hz, using a phase-less, eighth-order, Butterworth 
filter, and plotted with ground plane lateral acceleration on the 
abscissa versus hand-wheel steering angle on the ordinate. A second-
order polynomial curve fit of the data shall be constructed in the 
range from 0.01 g to 0.5 g. The slope of the constructed plot 
determines the understeer gradient value in the units of degrees of 
hand-wheel steering angle per g of ground plane lateral acceleration 
(degrees/g). Using the coordinate system specified in paragraph (d)(7) 
of this section, positive values for understeer gradient are required 
for values of ground plane lateral acceleration values from 0.10 g to 
0.50 g.
    (e) Performance requirements. Using the coordinate system specified 
in section 1422.4(d)(7), values for the understeer gradient shall be 
positive for values of ground plane lateral acceleration values from 
0.10 g to 0.50 g. The ROV shall not exhibit negative understeer 
gradients (oversteer) in the lateral acceleration range specified.


Sec.  1422.5  Requirements for occupant protection performance.

    (a) General. The ROV requirement for occupant protection shall be 
based on the maximum vehicle speed limitation when the seat belt of any 
occupied front seat is not buckled, and on passive coverage of the 
occupant shoulder area as measured by a probe test.
    (b) Vehicle speed limitation. (1) Test surface. Tests shall be 
conducted on a smooth, dry, uniform, paved surface constructed of 
asphalt or concrete. The surface area used for dynamic testing shall be 
kept free of debris and substances that may affect test results during 
vehicle testing.
    (i) Friction. Surface shall have a peak braking coefficient greater 
than or equal to 0.90, and a sliding skid coefficient greater than or 
equal to 0.80, when measured in accordance with ASTM E 1337 and ASTM 
E274, respectively.
    (ii) Slope. The test surface shall have a slope equal to or less 
than 1 degree (1.7% grade).
    (2) Test condition 1. Test conditions shall be as follows:
    (i) The test vehicle shall be a representative production vehicle. 
The

[[Page 69027]]

ROV shall have a redundant restraint system in the driver's seat.
    (ii) ROV test weight shall be the vehicle curb weight plus the test 
operator, only. If the test operator weighs less than 215 lbs.  11 lbs. (98 kg  5 kg), then the difference in weight 
shall be added to the vehicle to reflect an operator weight of 215 lbs. 
 11 lbs. (98 kg  5 kg).
    (iii) Tires shall be inflated to the pressures recommended by the 
ROV manufacturer for the vehicle test weight.
    (iv) The driver's seat belt shall not be buckled; however, the 
driver shall be restrained by the redundant restraint system for test 
safety purposes.
    (3) Test condition 2. Test conditions shall be as follows:
    (i) The test vehicle shall be a representative production vehicle. 
in standard condition.
    (ii) ROV test weight shall be the vehicle curb weight, plus the 
test operator and a passenger surrogate that will activate the seat 
occupancy sensor. If the test operator weighs less than 215 lbs.  11 lbs. (98 kg  5 kg), then the difference in weight 
shall be added to the vehicle to reflect an operator weight of 215 lbs. 
 11 lbs. (98 kg  5 kg).
    (iii) Tires shall be inflated to the pressures recommended by the 
ROV manufacturer for the vehicle test weight.
    (iv) The driver's seat belt shall be buckled. The front passenger's 
seat belt(s) shall not be buckled.
    (4) Test procedure. Measure the maximum speed capability of the ROV 
under Test Condition 1, specified in paragraph (b)(2) of this section, 
and Test Condition 2, specified in paragraph (b)(3) of this section 
using a radar gun or equivalent method. The test operator shall 
accelerate the ROV until maximum speed is reached, and shall maintain 
maximum speed for at least 15 m (50 ft.). Speed measurement shall be 
made when the ROV has reached a stabilized maximum speed. A maximum 
speed capability test shall consist of a minimum of two measurement 
test runs conducted over the same track, one each in opposite 
direction. If more than two measurement runs are made, there shall be 
an equal number of runs in each direction. The maximum speed capability 
of the ROV shall be the arithmetic average of the measurements made.
    (5) Performance requirement. The maximum speed capability of a 
vehicle with an unbuckled seat belt of the driver or any occupied front 
passenger seat shall be 15 mph or less.
    (c) Passive coverage of shoulder area.
    (1) General test conditions.
    (i) Probes shall be allowed to rotate through a universal joint.
    (ii) Forces shall be quasi-statically applied and held for 10 
seconds.
    (2) Shoulder/Hip performance requirement. The vehicle structure or 
restraint system must absorb the force specified in Sec.  1422.5(c)(5) 
with less than 25 mm (1 inch) of permanent deflection along the 
horizontal lateral axis.
    (3) Location of applied force. Locate point R on the vehicle, as 
shown in Figure X of ANSI/ROHVA 1-2011, American National Standard for 
Recreational Off-Highway Vehicles, approved July 11, 2011. The Director 
of the Federal Register approves this incorporation by reference in 
accordance with 5 U.S.C. 552(a) and 1 CFR part 51. You may obtain a 
copy from ASTM International, 100 Bar Harbor Drive, P.O. Box 0700, West 
Conshohocken, PA 19428; https://www.astm.org/cpsc.htm. You may inspect a 
copy at the Office of the Secretary, U.S. Consumer Product Safety 
Commission, Room 820, 4330 East West Highway, Bethesda, MD 20814, 
telephone 301-504-7923, or at the National Archives and Records 
Administration (NARA). For information on the availability of this 
material at NARA, call 202-741-6030, or go to: https://www.archives.gov/federal_register/code_of_federalregulations/ibr_locations.html. All 
measurements for the point shall be taken with respect to the base of 
the seatback. The base of the seatback lies on the surface of the seat 
cushion along the centerline of the seating position and is measured 
without a simulated occupant weight on the seat. Point R is located 432 
mm (17 inches) along the seat back above the base of the seatback. The 
point is 152 mm (6 inches) forward of and perpendicular to the seatback 
surface as shown in the figure. For an adjustable seat, Point R is 
determined with the seat adjusted to the rear-most position. Point R2 
applies to an adjustable seat and is located in the same manner as 
Point R except that the seat is located in the forward-most position.
    (4) Barriers. Remove all occupant protection barriers that require 
action on the part of the consumer to be effective (i.e. remove nets). 
Passive barriers that do not require any consumer action are allowed to 
remain.
    (5) Shoulder/Hip test method. Apply a horizontal, outward force of 
725 N (163 lbf.). Apply the force through the upper arm probe shown in 
Figure 2. The upper arm probe shall be oriented so that Point Q on the 
probe is coincident with Point R for a vehicle with a fixed seat, or 
Point Q shall be coincident with any point between R and R2 for a 
vehicle with an adjustable seat. The probe's major axis shall be 
parallel to the seatback angle at a point 17 inches along the seat back 
above the base of the seatback.

[[Page 69028]]

[GRAPHIC] [TIFF OMITTED] TP19NO14.018

Sec.  1422.6  Prohibited stockpiling.

    (a) Stockpiling. Stockpiling means manufacturing or importing a 
product which is the subject of a consumer product safety rule between 
the date of issuance of the rule and its effective date at a rate that 
is significantly greater than the rate at which such product was 
produced or imported during a base period prescribed by the Consumer 
Product Safety Commission.
    (b) Base period. The base period for ROVs is, at the option of each 
manufacturer or importer, any period of 365 consecutive days beginning 
on or after October 1, 2009, and ending on or before [the date of 
promulgation of the rule].
    (c) Prohibited acts. Manufacturers and importers of ROVs shall not 
manufacture or import ROVs that do not comply with the requirements of 
this part between [the date of promulgation of the rule] and [the 
effective date of the rule] at a rate that exceeds 10 percent of the 
rate at which this product was produced or imported during the base 
period described in paragraph (b) of this section.


Sec.  1422.7  Findings.

    (a) General. In order to issue a consumer product safety standard 
under the Consumer Product Safety Act, the Commission must make certain 
findings and include them in the rule. 15 U.S.C. 2058(f)(3). These 
findings are discussed in this section.
    (b) Degree and nature of the risk of injury. (1) CPSC received 428 
reports of ROV-related incidents from the Injury and Potential Injury 
Incident (IPII) and In-Depth Investigation (INDP) databases that 
occurred between January 1, 2003 and December 31, 2011, and were 
received by December 31, 2011. There were a total of 826 victims 
involved in the 428 incidents. Within the 428 ROV-related incidents, 
there were a total of 231 reported fatalities and 388 reported 
injuries. Seventy-five of the 388 injuries (19 percent) could be 
classified as severe, that is, the victim has lasting repercussions 
from the injuries received in the incident, based on the information 
available. The remaining 207 victims were either not injured or their 
injury information was not known. Of the 428 ROV-related incidents, 76 
involved drivers under 16 years of age (18 percent); 227 involved 
drivers 16 years of age or older (53 percent); and 125 involved drivers 
of unknown age (29 percent).
    (2) Using data reported through the National Electronic Injury 
Surveillance System (NEISS) from January 1, 2010 to August 31, 2010, 
the Commission conducted a special study to identify cases that 
involved ROVs that were reported through NEISS. (NEISS is a stratified 
national probability sample of hospital emergency departments that 
allows the Commission to make national estimates of product-related 
injuries.) Based on information obtained through the special study, the 
estimated number of emergency department-treated ROV-related injuries 
occurring in the United States between January 1, 2010 and August 31, 
2010, is 2,200 injuries. Extrapolating for the year 2010, the estimated 
number of emergency department-treated ROV-related injuries is 3,000, 
with a corresponding 95 percent confidence interval of 1,100 to 4,900.
    (c) Number of consumer products subject to the rule. (1) Sales of 
ROVs have increased substantially since their introduction. In 1998, 
only one firm manufactured ROVs, and fewer than 2,000 units were sold. 
By 2003, when a second major manufacturer entered the market, almost 
20,000 ROVs were sold. The only dip in sales occurred around 2008, 
which coincided with the worst of the credit crisis and recession that 
also started about the same time. In 2013, an estimated 234,000 ROVs 
were sold by about 20 different manufacturers. (This information is 
based upon a Commission analysis of sales data provided by Power 
Products Marketing, Eden Prairie, MN.)
    (2) The number of ROVs available for use has also increased 
substantially. Because ROVs are a relatively new product, we do not 
have any specific information on the expected useful life of ROVs. 
However, using the same operability rates that CPSC uses for ATVs, we 
estimate that there were about 570,000 ROVs available for use in 2010. 
By the end of 2013, there were an estimated 1.2 million ROVs in use.
    (d) The need of the public for ROVs and the effects of the rule on 
their utility, cost, and availability. (1) Currently there are two 
varieties of ROVs: Utility and recreational. Early ROV models 
emphasized the utility aspects of the vehicles, but the recreational 
aspects of the vehicles have become very popular.
    (2) In terms of the effects of the rule on ROVs utility, according 
to several ROV manufacturers, some ROV users ``might prefer limit 
oversteer in the off-highway environment.'' (This assertion was 
contained in a public comment on

[[Page 69029]]

the ANPR for ROVs (Docket No. CPSC-2009-0087) submitted jointly on 
behalf of Arctic Cat, Inc., Bombardier Recreational Products, Inc., 
Polaris Industries, Inc., and Yamaha Motor Corporation, USA.) To the 
extent that the requirements in the proposed rule would reduce the 
ability of these users to intentionally reach limit oversteer, the 
proposed rule could have some adverse impact on the utility or 
enjoyment that these users receive from ROVs. These impacts would 
probably be limited to a small number of recreational users who enjoy 
activities or stunts that involve power oversteering or limit 
oversteer.
    (3) Although the impact on consumers who prefer limit oversteer 
cannot be quantified, the Commission expects that it will be low. Any 
impact would be limited to those consumers who wish to intentionally 
engage in activities involving the loss of traction or power oversteer. 
The practice of power oversteer is the result of driver choices, such 
as the speed at which a user takes a turn. The proposed rule would not 
prevent ROVs from reaching limit oversteer under all conditions; nor 
would the rule prevent consumers from engaging in these activities. At 
most, the proposed rule might make it somewhat more difficult for users 
to reach limit oversteer in an ROV. Moreover, consumers who have a high 
preference for vehicles that oversteer would be able to make 
aftermarket modifications, such as adjustments to the suspension of the 
vehicle, or using different wheels or tires to increase the potential 
for oversteering.
    (4) The seat belt speed limiter requirement could have a negative 
effect on utility and impose some unquantifiable costs on some users 
who would prefer not to use seat belts. The cost to these users would 
be the time required to buckle and unbuckle their seat belts and any 
disutility cost, such as discomfort caused by wearing the seat belt. We 
cannot quantify these costs because we do not know how many ROV users 
choose not to wear their seat belts, nor do we have the ability to 
quantify any discomfort or disutility that they would experience from 
wearing seat belts. However, the proposed rule does not require that 
the seat belts be fastened unless the vehicle is traveling 15 mph or 
faster. This should serve to mitigate these costs because many people 
who would be inconvenienced or discomforted by the requirement, such as 
people using the vehicle for work or utility purposes or who must 
frequently get on and off the vehicle are likely to be traveling at 
lower speeds.
    (5) The effect of the rule on cost and availability of ROVs is 
expected to be minimal. The average manufacturer's suggested retail 
prices (MSRP) of ROVs, weighted by units sold, was about $13,100 in 
2013, with a range of about $3,600 to $20,100. The preliminary 
regulatory analysis estimates the per-unit cost to ROVs of the rule to 
be $61 to $94. Because this per-unit cost resulting from the rule is a 
very small percentage of the overall retail price of a ROV, it is 
unlikely that the rule would have more than a minimal effect on the 
cost or availability of ROVs.
    (e) Other means to achieve the objective of the rule, while 
minimizing the impact on competition and manufacturing. (1) The 
Commission does not believe the rule will have adverse impact on 
competition. The preliminary regulatory analysis estimates the per-unit 
cost to ROVs of the rule to be $61 to $94. The average manufacturer's 
suggested retail prices (MSRP) of ROVs, weighted by units sold, was 
about $13,100 in 2013, with a range of about $3,600 to $20,100. The 
per-unit cost resulting from the rule is a very small percentage of the 
overall retail price of a ROV and is unlikely to have any impact on 
competition.
    (2) The Commission believes that some but not all ROV models 
already meet the rule's requirement that the speed of the vehicle be 
limited if the driver's seat belt is not fastened. Before implementing 
any changes to their vehicles to meet the requirement, manufacturers 
whose ROVs do not meet the seatbelt speed limiter requirement would 
have to analyze their options for meeting the requirement. This process 
would include developing prototypes of system designs, testing the 
prototypes and refining the design of the systems based on this 
testing. Once the manufacturer has settled upon a system for meeting 
the requirement, the system will have to be incorporated into the 
manufacturing process of the vehicle. This will involve producing the 
engineering specifications and drawings of the system, parts, 
assemblies, and subassemblies that are required. Manufacturers will 
need to obtain the needed parts from their suppliers and incorporate 
the steps needed to install the system on the vehicles in the assembly 
line. The Commission believes that manufacturers should be able to 
complete all of these activities and be ready to produce vehicles that 
meet the requirement within 12 calendar months. The Commission is 
proposing a 12-month effective date for the occupant protection 
requirements to minimize the burden on manufacturing.
    (f) Unreasonable risk. (1) CPSC received 428 reports of ROV-related 
incidents from the Injury and Potential Injury Incident (IPII) and In-
Depth Investigation (INDP) databases that occurred between January 1, 
2003 and December 31, 2011, and were received by December 31, 2011. 
There were a total of 826 victims involved in the 428 incidents. Within 
the 428 ROV-related incidents, there were a total of 231 reported 
fatalities and 388 reported injuries. Seventy-five of the 388 injuries 
(19 percent) could be classified as severe, that is, the victim has 
lasting repercussions from the injuries received in the incident, based 
on the information available.
    (2) The estimated cost and benefits of the rule on an annual basis 
can be calculated by multiplying the estimated benefits and costs per 
unit by the number of ROVs sold in a given year. In 2013, 234,000 ROVs 
were sold. If the proposed rule had been in effect that year, the total 
quantifiable cost would have been between $14.3 million and $22.0 
million ($61 and $94 multiplied by 234,000 units, respectively). The 
total quantifiable benefits would have been at least $515 million 
($2,199 x 234,000). Of the benefits, about $453 million (or about 88 
percent) would have resulted from the reduction in fatal injuries, and 
about $62 million (or about 12 percent) of the benefits would have 
resulted from a reduction in the societal cost of nonfatal injuries. 
About $47 million of the reduction in the societal cost of nonfatal 
injuries would have been due to a reduction in pain and suffering. We 
conclude preliminarily that ROVs pose an unreasonable risk of injury 
and that the proposed rule is reasonably necessary to reduce that risk.
    (g) Public interest. (1) This proposed rule is in the public 
interest because it may reduce ROV-related deaths and injuries in the 
future. The Commission believes that improving lateral stability (by 
increasing rollover resistance) and improving vehicle handling (by 
correcting oversteer to sub) are the most effective approaches to 
reduce the occurrence of ROV rollover incidents. ROVs with higher 
lateral stability are less likely to roll over because more lateral 
force is necessary to cause rollover. ROVs exhibiting understeer during 
a turn are also less likely to rollover because lateral acceleration 
decreases as the path of the ROV makes a wider turn, and the vehicle is 
more stable if a sudden change in direction occurs.
    (2) The Commission believes that, when rollovers do occur, 
improving occupant protection performance (by increasing seat belt use) 
will mitigate

[[Page 69030]]

injury severity. CPSC analysis of ROV incidents indicates that 91 
percent of fatally ejected victims were not wearing a seat belt at the 
time of the incident. Increasing seat belt use, in conjunction with 
better shoulder retention performance, will significantly reduce 
injuries and deaths associated with an ROV rollover event.
    (h) Voluntary standards. (1) The Commission is aware of two 
voluntary standards that are applicable to ROVs, ANSI/ROHVA 1, American 
National Standard for Recreational Off-Highway Vehicles and ANSI/B71.9, 
American National Standard for Multipurpose Off-Highway Utility 
Vehicles. As described in detail in the preamble, the Commission 
believes that the current voluntary standard requirements not 
adequately reduce the risk of injury or death associated with ROVs. 
Neither voluntary standard requires that ROVs understeer, as required 
by the proposed rule. According to the ES staff, drivers are more 
likely to lose control of vehicles that oversteer, which can lead to 
the vehicle rolling over or to other types of accidents.
    (2) Both voluntary standards have requirements that are intended to 
set standards for dynamic lateral stability. ANSI/ROHVA 1-2011 uses a 
turn-circle test for dynamic lateral stability that is more similar to 
the test in the proposed rule for whether the vehicle understeers than 
it is to the test for dynamic lateral stability. The dynamic stability 
requirement in ANSI/OPEI B71.9-2012 uses a J-turn test, like the 
proposed rule, but measures different variables during the test and 
uses a different acceptance criterion. However, ES staff does not 
believe that the tests procedures in either standard have been properly 
validated as being capable of providing useful information about the 
dynamic stability of the vehicle. Moreover, the voluntary standards 
would find some vehicles acceptable even though their lateral 
acceleration at rollover is less than 0.70 g, which is the acceptance 
criterion in the proposed rule.
    (3) Both voluntary standards require that manufacturers include a 
lighted seat-belt reminder that is visible to the driver and remains on 
for at least 8 seconds after the vehicle is started, unless the 
driver's seatbelt is fastened. However, virtually all ROVs on the 
market already include this feature and, therefore, relying only on the 
voluntary standards would not be expected to raise seatbelt use over 
its current level.
    (4) The voluntary standards include requirements for retaining the 
occupant within the protective zone of the vehicle in the event of a 
rollover including two options for restraining the occupants in the 
shoulder/hip area. However, testing performed by CPSC identified 
weaknesses in the performance-based tilt table test option that allows 
unacceptable occupant head ejection beyond the protective zone of the 
vehicle Rollover Protective Structure (ROPS). CPSC testing indicated 
that a passive shoulder barrier could reduce the head excursion of a 
belted occupant during quarter-turn rollover events. The Commission 
believes that this can be accomplished by a requirement for a passive 
barrier based on the dimensions of the upper arm of a 5th percentile 
adult female, at a defined area near the ROV occupants' shoulder as 
contained in the proposed rule.
    (i) Relationship of benefits to costs. (1) The estimated cost and 
benefits of the rule on an annual basis can be calculated by 
multiplying the estimated benefits and costs per unit by the number of 
ROVs sold in a given year. In 2013, 234,000 ROVs were sold. If the 
proposed rule had been in effect that year, the total quantifiable cost 
would have been between $14.3 million and $22.0 million ($61 and $94 
multiplied by 234,000 units, respectively). The total quantifiable 
benefits would have been at least $515 million ($2,199 x 234,000).
    (2) On a per unit basis, we estimate the total cost of the proposed 
rule to be $61 to $94 per vehicle. We estimate the total quantifiable 
benefits of the proposed rule to be $2199 per unit. This results in net 
quantifiable benefits of $2105 to $2138 per unit. Quantifiable benefits 
of the proposed rule could exceed the estimated $2199 per unit because 
the benefit associated with the vehicle handling and lateral stability 
requirement could not be quantified.
    (j) Least burdensome requirement. The Commission considered less 
burdensome alternatives to the proposed rule regarding ROVs, but 
concluded that none of these alternatives would adequately reduce the 
risk of injury.
    (1) Not issuing a mandatory rule, but instead relying upon 
voluntary standards. If CPSC did not issue a mandatory standard, most 
manufacturers would comply with one of the two voluntary standards that 
apply to ROVs. The Commission does not believe either voluntary 
standard adequately addresses the risk of injury and death associated 
with ROVs.
    (2) Including the dynamic lateral stability requirement or the 
understeer requirement, but not both. The Commission believes that both 
of these characteristics need to be addressed. According to CPSC's 
Directorate for Engineering Sciences, a vehicle that meets both the 
dynamic stability requirement and the understeer requirement should be 
safer than a vehicle that meets only one of the requirements. Moreover, 
the cost of meeting just one requirement is not substantially lower 
than the cost of meeting both requirements. The cost of testing a 
vehicle for compliance with both the dynamic lateral stability and 
vehicle handling/understeer requirement was estimated to be about 
$24,000. However, the cost of testing for compliance with just the 
dynamic stability requirement itself would be about $20,000, or only 
about 17 percent less than the cost of testing for compliance with both 
requirements together. This is because the cost of renting and 
transporting the vehicle to the test site, instrumenting the vehicle 
for the tests, and making some initial static measurements are 
virtually the same for both requirements and would only have to be done 
once if the tests for both requirements were conducted on the same day. 
Moreover, changes in the vehicle design that affect the lateral 
stability of the vehicle could also impact the handling of the vehicle. 
For these reasons, the proposed rule includes both a dynamic stability 
and vehicle handling requirement.
    (3) Loud or intrusive seatbelt reminders instead of seatbelt/speed 
limitation requirements. (i) Currently, most ROVs meet the voluntary 
standards that require an 8-second visual seatbelt reminder. Some more 
intrusive systems have been used on passenger cars. For example, one 
system resumes warning the driver after about 65 seconds if his or her 
seatbelt is not fastened and the car is traveling at more than 3 mph. 
The system flashes a warning light and sounds a chime for 6 seconds 
every 30 seconds for up to 5 minutes so long as the car is operating 
and the driver's seatbelt is not fastened. A similar system is used in 
which the warning could last for longer than 9 minutes if the driver's 
seatbelt is not fastened. Although studies of both systems found an 
increase in the use seatbelts, the systems' effectiveness was limited. 
Moreover, audible warnings are not likely to be effective in ROVs. ROVs 
are open vehicles and the ambient noise is higher than in the enclosed 
passenger compartment of a car. ROV drivers would not hear the warning 
and be motivated to fasten their seatbelts unless the warning was 
substantially louder than the systems used in passenger cars.
    (ii) In contrast, these more intrusive seatbelt warning systems are 
unlikely to be as effective as the seatbelt speed limitation 
requirement in the proposed rule. The Commission believes that the

[[Page 69031]]

requirement in the proposed rule will cause most drivers and passengers 
that desire to exceed 15 mph to fasten their seatbelts. Research 
supports this position. One experiment used a haptic feedback system to 
increase the force the driver needed to exert to depress the gas pedal 
when the vehicle exceeded 25 mph if the seatbelt was not fastened. The 
system did not prevent the driver from exceeding 25 mph, but it 
increased the amount of force required to depress the gas pedal to 
maintain a speed greater than 25 mph. In this experiment all 7 
participants chose to fasten their seatbelts.

    Dated: October 31, 2014.
Todd A. Stevenson,
Secretary, Consumer Product Safety Commission.
[FR Doc. 2014-26500 Filed 11-18-14; 8:45 am]
BILLING CODE 6355-01-P