[Federal Register Volume 62, Number 148 (Friday, August 1, 1997)]
[Notices]
[Pages 41477-41481]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 97-20295]


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

DEPARTMENT OF TRANSPORTATION


National Highway Traffic Safety Administration (NHTSA) Denial of 
Motor Vehicle Defect Petition

AGENCY: National Highway Traffic Safety Administration (NHTSA), 
Department of Transportation.

ACTION: Denial of motor vehicle defect petition.

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

SUMMARY: This document denies a November 8, 1996 petition from the 
Center for Auto Safety requesting that the agency commence a proceeding 
to determine the existence of defects related to motor vehicle safety 
in the air bag systems in certain vehicles. After reviewing the 
petition and other information, NHTSA has concluded that further 
investigation of these vehicles is unlikely to result in a 
determination that such safety-related defects exist. The agency 
accordingly denies the petition.

FOR FURTHER INFORMATION CONTACT: Mr. Thomas Cooper, Chief, Vehicle 
Integrity Branch, Office of Defects Investigation, NHTSA, 400 Seventh 
Street, SW, Washington, DC 20590. Telephone: (202) 366-5218.

SUPPLEMENTARY INFORMATION: On November 8, 1996, the Center for Auto 
Safety (CAS) submitted a petition requesting the agency to investigate 
certain motor vehicles for ``defective airbag crash sensor and/or 
deployment systems which result in airbag induced injuries in crashes 
below 12-mph.'' 1 The vehicles identified in the CAS letter 
are the 1991 through 1992 Chevrolet Corsica, the 1990 through 1992 Ford 
Taurus, and the 1994 through 1996 Chrysler minivan. CAS alleges that 
the subject vehicles are over-represented for air bag-caused crash 
deaths at low speeds when compared to other vehicles. CAS cites the 
number of deaths of adult drivers in the Corsica and Taurus vehicles 
and of child passengers in the Chrysler minivan vehicles.
---------------------------------------------------------------------------

    \1\ Footnote 1 in the CAS petition states: ``All speeds referred 
to are change in velocity of the striking vehicle as opposed to its 
impact speed.''
---------------------------------------------------------------------------

    In the same letter, CAS separately requested the agency to ``amend 
FMVSS 208 to set a 12 mph threshold below which an airbag cannot 
trigger unless the manufacturer establishes it will not injure an out-
of-position occupant.'' This petition analysis evaluates only that 
portion of the petition requesting a defect investigation. The 
rulemaking requested by CAS was initially discussed in the Notice of 
Proposed Rulemaking dated January 6, 1997 (62 FR at 827).
    Following receipt of the petition, NHTSA officials requested CAS to 
clarify the specific issue(s) it wished the agency to investigate. In 
response to this request, on December 12, 1996, CAS clarified its 
November 8, 1996 letter:

    CAS does not request NHTSA to investigate the design threshold 
trigger speed specified by the manufacturer for these vehicles but 
rather two other factors which combined have resulted in 19 of the 
53 known deaths caused by bags--(1) the wide variability from the 
design threshold trigger speed which may result in injurious airbag 
deployments at barrier equivalent velocities (BEVs) as low as 7-mph 
when the design speed may be 12 to 14-mph BEV and (2) the 
aggressivity of the airbag itself because of its deployment 
velocity, trajectory or proximity to the occupant.

    After reviewing the petition, as well as information furnished by 
Chrysler Corporation (Chrysler), Ford Motor Company (Ford), General 
Motors Corporation (GM), the Insurance Institute for Highway Safety 
(IIHS), and data in the agency's possession from ongoing rulemaking 
proceedings and other sources, NHTSA has concluded that further 
investigation of these vehicles is unlikely to result in a 
determination that the air bag systems in the vehicles identified in 
the petition contain safety-related defects as alleged by the 
petitioner, and that a further commitment of agency resources in this 
effort is not warranted. The agency accordingly denies the petition.

Alleged Problem And Safety-Related Consequences

    The petitioner alleges that specific Chrysler, Ford and GM vehicles 
present

[[Page 41478]]

an unreasonable risk to safety from a combination of low speed and 
aggressive air bag deployment. The petitioner bases this view on 
statistics concerning the number of fatalities caused by air bags in 
the subject vehicles compared with the number of such fatalities in 
other vehicle models.
    Fatalities due to air bags are principally due to the air bag 
itself or the air bag module cover making aggressive contact with the 
vehicle occupant during the inflation phase of air bag deployment. 
Those individuals who are extremely close to the air bag module at the 
time of deployment are most likely to be adversely affected.

Vehicles Involved

    The petition identifies the subject Chrysler vehicles as the 
Chrysler Town and Country, Dodge Caravan and Grand Caravan, and 
Plymouth Voyager and Grand Voyager minivans for model years (MY) 1994-
96. Chrysler produced a total of 1,569,027 such vehicles.
    The petition identifies the subject Ford vehicles as the Ford 
Taurus for MY 1990 through 1992. Ford produced a total of 931,900 such 
vehicles.2
---------------------------------------------------------------------------

    \2\ The petition did not identify the ``corporate siblings'' of 
the designated Ford and GM vehicles; i.e., the Mercury Sable and the 
Chevrolet Beretta. If those ``corporate siblings'' are included, the 
vehicle populations would be 1,232,000 and 412,000, respectively.
---------------------------------------------------------------------------

    The petition identifies the subject GM vehicles as the Chevrolet 
Corsica for MY 1991 and 1992. GM produced a total of 290,145 such 
vehicles.2

System Description

    An air bag is a fabric bag that inflates extremely rapidly in the 
event of a frontal crash to protect an occupant from moving forward and 
striking solid parts of the interior of the vehicle and also to absorb 
and distribute crash forces. The air bag supplements the protection 
provided by the manual lap and shoulder belt. It also must protect 
unbelted individuals in crashes the equivalent of up to 30 mph into a 
fixed barrier. Each air bag system contains two or more sensors to 
detect the initiation of a crash and provide an electrical signal that 
a crash is occurring.3 The sensors are designed to 
discriminate between crashes or impacts that warrant air bag inflation 
and those that do not warrant air bag inflation. At least two sensors 
(one safing sensor 4 and one crash sensor) must be activated 
to inflate the bag. The type, location, number, and calibration of 
sensors vary by vehicle model.
---------------------------------------------------------------------------

    \3\ There are also a small number of all-mechanical crash sensor 
systems which use the mechanical energy of a firing pin (rather than 
an electrical switch) to signal that a crash is occurring. The 
subject vehicles contain systems using electromechanical sensors 
which provide an electrical signal.
    \4\ The safing sensor is a redundant sensor to add reliability 
to prevent the system from deploying an air bag in response to a 
faulty signal from the crash sensor.
---------------------------------------------------------------------------

    Each manufacturer provides a specification for the vehicle's air 
bag system that establishes both a ``low limit,'' below which the air 
bag must not deploy, and a ``threshold,'' above which the air bag must 
deploy. Both the low limit and the threshold are expressed in terms of 
change in velocity of the vehicle. Chrysler reports that for its 1994 
through 1996 minivans, the air bags must not deploy at 8 mph or below 
and must deploy at 14 mph or above. Ford reports that the air bag in 
1990 through 1992 Taurus vehicles must not deploy at 8 mph or below and 
must deploy at 14 mph or above. GM reports that the air bag in the 1991 
and 1992 Corsica must not deploy at 7 mph or below and must deploy at 
14 mph or above.5
---------------------------------------------------------------------------

    \5\ In 1970, as part of a comprehensive rulemaking on automatic 
restraints, NHTSA proposed to require that air bags not deploy when 
the vehicle impacts a fixed barrier at any velocity less than 15 
miles per hour, at any angle. 35 FR 16937, at 16938 (November 3, 
1970). However, after considering opposing comments from vehicle 
manufacturers, the agency did not adopt this requirement because it 
determined that it was preferable to allow manufacturers freedom in 
the design of their protective systems at all speeds. 36 FR 4600, at 
4602 (March 10, 1971).
---------------------------------------------------------------------------

    All air bag systems include a diagnostic module which performs two 
functions. (1) It provides a back-up energy source in case the battery 
voltage is cut during a crash. The back-up voltage is available to 
trigger deployment of the air bag if the sensors detect a crash 
requiring air bag inflation. (2) It also performs a diagnostic check of 
the electrical system components and connections both when the ignition 
is first turned on and periodically during vehicle use.
    The air bag inflator module contains the air bag, the inflator, and 
the initiator. The inflator module is located on the steering wheel for 
the driver and in the instrument panel for the passenger (many vehicles 
have only a driver side air bag).6 When the sensors detect a 
crash that meets the pre-established criteria for deployment of the air 
bag, an electrical current flows through the initiator and ignites the 
material in the inflator. Or, in the case of all-mechanical sensors, a 
firing pin activates a detonator that ignites the inflator material. 
The gas generated by the inflator rapidly inflates the air bag.
---------------------------------------------------------------------------

    \6\ The subject Chrysler minivans have driver and front 
passenger air bags. The subject GM vehicles have an air bag on the 
driver side only. The subject Ford vehicles have standard driver 
side only with optional passenger side available in MY 1992.
---------------------------------------------------------------------------

    In order to provide continuous electrical connection to the air bag 
in the steering wheel, each driver air bag connects to a coil assembly 
or clock spring. This device allows rotation of the air bag module in 
the steering wheel while maintaining continuous electrical contact with 
the sensors, the diagnostic module, and the vehicle's electric power 
supply.

Modifications

    The manufacturers' responses to information request letters from 
NHTSA's Office of Defects Investigation (ODI) provided the following 
information on vehicle modifications related to the air bag systems in 
the subject vehicles.

Chrysler

    Chrysler reports that 1994 was the first year for passenger air 
bags in its minivans. In developing its passenger air bags, Chrysler 
made modifications to its pre-1994 driver air bag system.

    * * * Chrysler introduced single point electronic crash sensing 
internal to the airbag [electronic] control module (AECM) rather 
than the distributed system of prior years. No other changes were 
made relating to deployment or aggressivity through the end of the 
vehicles' production, [i.e., through the end of MY 1995].

    The 1996 minivan is a totally new vehicle; with new crash sensing 
algorithms in the AECM. In addition, the 1996 Chrysler minivan 
passenger air bag is mid-mounted on the instrument panel compared with 
the 1994 and 1995 passenger air bags, which are top-mounted. Neither 
the 1996 AECM nor the 1996 passenger air bag is interchangeable with 
those from the earlier model years.

Ford

    The 1990 Taurus is the first model year Taurus with an air bag. 
Ford's response to NHTSA's information request (IR) states,

    Ford Taurus vehicles were equipped with five sensors in model 
year 1990, four in model year 1991, and three in model year 1992.
    Ford adds,
    Although there were a number of minor changes made to 1990-1992 
model year Taurus vehicles and the driver side air bags installed in 
those vehicles, there were no modifications or changes in the design 
or manufacture of the vehicles or air bag system components made to 
address either air bag system ``deployment variability or air bag 
aggressivity,'' as no need has been identified for such changes or 
modifications.

    Ford states that it reduced the number of sensors from five to 
three to reduce

[[Page 41479]]

system complexity and cost. According to Ford, these reductions ``were 
made possible by ongoing improvements in technology of air bag crash 
sensor design.'' \7\
---------------------------------------------------------------------------

    \7\ In a letter dated March 24, 1997, CAS asserted that Ford 
failed to fully respond to NHTSA's IR letter. CAS requested NHTSA to 
ask Ford for air bag system modifications to its Taurus vehicles 
subsequent to the subject MY 1990-92 Taurus. NHTSA responded to CAS 
indicating that such information was not needed for the purpose of 
evaluating this petition.
---------------------------------------------------------------------------

GM

    The 1991 Corsica is the first model year Corsica with an air bag. 
In its response to the agency's IR, GM states,

    The vehicle design or the subject components were not changed or 
modified on the 1991-92 Corsica vehicles to address either air bag 
deployment variability or air bag aggressivity.

Crash Reports

    As of May 15, 1997, NHTSA is aware of the following reports of low 
speed motor vehicle crashes in the subject vehicles in which fatalities 
and injuries have been attributed to deployment of the air bags. These 
reports have been obtained from agency field crash investigations and 
from data provided to the agency by the respective vehicle 
manufacturers. For the MY 1994-96 Chrysler minivan vehicles, the agency 
has reports of 11 such crashes, resulting in 9 deaths and 2 injuries to 
occupants in the front passenger seat. For the MY 1990-92 Ford Taurus, 
the agency has reports of 24 such crashes, causing 5 deaths and 19 
injuries to drivers. For the MY 1991-92 GM Corsica, the agency has 
reports of 11 such crashes, resulting in 4 deaths and 7 injuries to 
drivers. The agency is aware of no fatal crash reports for the MY 1990-
92 Sable or MY 1991-92 Beretta.8
---------------------------------------------------------------------------

    \8\  The Chevrolet Corsica and Chevrolet Beretta share the same 
vehicle platform and restraint system (with the Beretta having only 
two doors). Likewise, the Ford Taurus and Mercury Sable share the 
same vehicle platform and restraint system.
---------------------------------------------------------------------------

    NHTSA prepared a Special Crash Investigation (SCI) report for each 
of the air bag-related fatal crashes. In most, the report provides a 
calculation of the change of velocity in the longitudinal direction of 
the vehicle. For the subject Ford Taurus, there are 5 fatal crashes in 
which the change of velocity ranges from a low of 7.7 mph to a high of 
12.5 mph, with an average of 10.4 mph. For the GM Corsica, there are 4 
fatal crashes in which the change of velocity ranges from a low of 7 
mph to a high of 16 mph, with an average of 11.2 mph. For the Chrysler 
minivans, there are 9 fatal crashes in which the change of velocity 
ranges from a low of 8 mph to a high of 17.5 mph, with an average of 
13.3 mph.
    For the non-subject vehicles, there are 33 fatal crashes for which 
the change in velocity is reported. The change of velocity ranges from 
a low of 7 mph to a high of 20 mph, with an average of 12.6 mph. The 
SCI reports indicate that the average change in velocity for the 
crashes leading to air bag-related fatalities in the subject vehicles 
is not significantly different from the change in velocity in crashes 
in which there were air bag-related fatalities in non-subject vehicles.

Testing

    Manufacturers have performed considerable testing in the process of 
developing and improving air bag systems. NHTSA has also performed 
tests in research programs, in its compliance programs, and as part of 
the New Car Assessment Program (NCAP). This data has provided 
additional technical understanding of both the potential benefits of 
air bags and the potential risks associated with their deployment.

FMVSS 208--NHTSA

    All new passenger cars and light trucks must comply with Federal 
Motor Vehicle Safety Standard (FMVSS) 208, ``Occupant Crash 
Protection.'' This standard specifies minimum occupant protection 
performance levels for the restraint systems in those vehicles. 
Starting with MY 1987, manufacturers were required to phase in 
automatic occupant restraints to meet specified injury criteria. 
Beginning with MY 1990, all vehicles were required to meet the 
automatic restraint injury criteria and manufacturers began to make 
significant numbers of vehicles with driver air bags. Then in 1991, the 
Intermodal Surface Transportation Efficiency Act directed the agency to 
amend FMVSS 208 to require air bags as the form of automatic crash 
protection in light vehicles. The agency accordingly amended the 
standard in 1993 to require air bags for both front seat positions in 
all automobiles by MY 1998, and in all light trucks by MY 1999. 
Subsequently, manufacturers continued on an accelerated program to 
install both driver and passenger air bags.
    On March 19, 1997 the agency temporarily amended FMVSS 208 to allow 
manufacturers to quickly depower air bags so that they inflate less 
aggressively. The agency took this action in response to the growing 
number of air bag-related fatalities and injuries to young children and 
adults. Prior to March 19, 1997 in a FMVSS 208 test, the test vehicle 
carried instrumented unbelted test dummies and the vehicle impacted a 
fixed flat barrier at 30 mph. Data recorded from the test dummies 
provided information to determine if the vehicle complied with the 
Standard. All of the subject vehicles were produced prior to the March 
1997, FMVSS 208 amendment.
    Table 1 presents a summary of the results of the relevant FMVSS 208 
compliance testing performed on the subject vehicles by NHTSA. The air 
bag principally acts on the surfaces of the upper body to protect it 
from harm. The injury criteria, ``Head Injury Criterion (``HIC''), and 
``Chest G,'' predict the level of human injury to the upper body that 
could occur in a crash. The maximum allowable HIC and Chest G is shown 
at the bottom of Table 1. All of the subject vehicles met the dummy 
injury criteria requirements when tested by NHTSA. The FMVSS also 
contains injury criteria for the lower body (femur), but because the 
air bag does not impact this region of the body, the femur data results 
are not shown.

                 Table 1.--FMVSS Compliance Test Results                
------------------------------------------------------------------------
                                                                 Chest G
                        Model                          HIC \9\    \10\  
------------------------------------------------------------------------
1990 Taurus (driver)................................       249        53
1991 Beretta \11\ (driver)..........................       172        52
1994 Caravan (pass.)................................       148        51
1996 Caravan (pass.)................................       129        39
FMVSS Requirement...................................      1000       60 
------------------------------------------------------------------------
\9\ HIC is a measure of the potential for injury to the brain.          
\10\ Chest G is a measure of the potential for injury to the chest.     
\11\ NHTSA did not conduct FMVSS 208 compliance tests on either the 1991
  or 1992 Corsica.                                                      

NCAP--NHTSA

    NHTSA conducts the New Car Assessment Program (NCAP) to provide 
information on the crash performance of vehicles at speeds greater than 
the test speed specified in FMVSS 208. The program is different from 
the FMVSS testing in two principal areas: (1) the vehicle crash speed 
into the barrier is higher (35 mph), and (2) the test dummies are 
restrained with manual safety belts in addition to the passive 
restraint system provided in the vehicle. The NCAP results for the 
subject vehicles are shown in Table 2.

                       Table 2.--NCAP Test Results                      
------------------------------------------------------------------------
                        Model                            HIC     Chest G
------------------------------------------------------------------------
1990 Taurus (driver)................................       735        46
1991 Corsica (driver)...............................       493        41
1991 Taurus (driver)................................       480        44
1994 Caravan (pass.)................................       422        45

[[Page 41480]]

                                                                        
1996 Caravan (pass.)................................       403        46
------------------------------------------------------------------------

Leading Edge Deployment Speed--IIHS

    In a March 21, 1997 letter to ODI, Ford provided a copy of a report 
issued by IIHS in January 1995 titled ``Leading Edge Deployment Speed 
of Production Air Bags.'' The purpose of this report was ``To assess 
the relative potential of different air bag designs to cause skin 
abrasions * * *.'' The report describes a series of tests and presents 
data showing the speed at which the front or leading surface of the air 
bag emerges from the air bag cover during the inflation process. The 
data also reports the maximum excursion values--the distance that the 
bag reaches beyond the plane of the steering wheel during its inflation 
process. The report provides data on driver side air bags in 15 MY 1993 
passenger cars, including the Ford Taurus. 12
---------------------------------------------------------------------------

    \12\ Ford reports, ``the bags installed in 1993 Taurus vehicles 
are the same configuration as those installed in 1990-92 Taurus 
vehicles.''
---------------------------------------------------------------------------

    Although there are many variables that could affect the likelihood 
of air bag injuries, contact with a deploying air bag may be more 
likely in those systems with greater air bag excursion distances during 
inflation. In the same manner, the impact force of air bag contact 
during inflation with the occupant might be greater in those systems 
presenting a higher leading edge speed.
    The data show that the leading edge speed of the 1993 Ford Taurus 
fell within the range of leading edge speeds measured for all of the 
tested vehicles. For all vehicles, the measured leading edge speed 
ranged from a low of 170 km/h to a high of 328 km/h. For the Ford 
Taurus, the speeds were recorded as 266 and 291 km/h (two tests).
    The maximum excursion of the Ford Taurus air bag also was within 
the range of all of the models tested. The Ford Taurus air bag 
excursion values were 370 and 378 mm (two tests). The values for all 
models tested ranged from 297 to 487 mm.
    Neither the Corsica nor the Chrysler minivans were included in this 
IIHS study.

Rulemaking Research--NHTSA

    In preparation for possible regulatory action to amend FMVSS 208, 
the agency conducted tests to determine the effectiveness of depowering 
air bags on adult drivers and child passengers who are in close 
proximity to the air bag at the time of deployment. These tests are 
described in NHTSA's February 1997 report, ``Actions to Reduce the 
Adverse Effects of Air Bags, FMVSS No. 208, Depowering'' by the Office 
of Regulatory Analysis, Plans and Policy (p. III -1 to III-29). The 
tests included driver and passenger static air bag deployments in sled 
bucks representing vehicles of MY 1994 and 1996 vintage. The testing 
focused on the effects of air bag inflation on out-of-position (OOP) 
occupants. International Standards Organization (ISO) procedures were 
used as a guideline for determining the positioning of the OOP test 
dummy. This testing program gathered baseline data representing 
production air bag systems from different manufacturers and also 
gathered data on depowered inflators for comparison to the baseline 
results.

OOP Testing--Ford

    Although ODI did not specifically request data on OOP occupant 
testing, Ford provided such data comparing the performance of the air 
bag system on the 1992 Taurus with that of seven other passenger 
vehicles. This testing was conducted by Ford for potential use in 
litigation. The results show the Taurus performance as within the range 
of results for the other vehicles. Neither GM nor Chrysler conducted 
OOP testing of the subject vehicles.

Crash Data

NASS

    NHTSA's National Automotive Sampling System (NASS) has records of a 
sampling of crashes and an analysis that includes a computation of the 
change in velocity (based on measurements of vehicle crash damage) of 
the vehicle during the crash impact. A review of this data shows that 
the air bags in many vehicles, in addition to the subject vehicles, 
deploy in low speed crashes. The NASS file for calendar years 1989 
through 1996 (a partial file for 1996 is now available) contains 
reports on 412 vehicles in which the air bag deployed in a crash at a 
computed longitudinal change in velocity of less than 10 mph. More than 
50% (221) of the 412 crashed vehicles involve a 7-mph change in 
velocity or less and these include over 70 different non-subject model 
vehicles.

IIHS

    In a letter dated December 20, 1996, IIHS submitted an analysis of 
all crashes in which a driver or passenger death has been attributed to 
the air bag. IIHS provided a series of tables ``listing airbag 
inflation fatalities and the exposure of the vehicles in which these 
fatalities occurred.'' The IIHS data is based on NHTSA data (SCI). The 
IIHS analysis compares the subject vehicles identified in the CAS 
petition with other vehicle models for which deaths have been 
attributed to air bags. The air bag-related fatality rate for each 
model is expressed as the number of deaths, as of September 30, 1996, 
divided by the number of million registered vehicle years. This 
provides an exposure-adjusted basis for comparing the air bag-related 
fatality rates of one model to another.
    In its analysis, IIHS constructed vehicle groups from the subject 
vehicles and from other vehicles in which there has been an air bag 
fatality. The IIHS placed the MY 1991-92 Corsica into a group 
consisting of the MY 1991-93 Corsica and Beretta; placed the MY 1990-92 
Taurus into a group consisting of the MY 1990-93 Taurus and Sable; and 
split the MY 1994-96 Chrysler minivans into two groups, the MY 1994-95 
minivans, and the MY 1996 minivans. Non-subject vehicles were placed 
into appropriate groups as well. IIHS's rationale for placement of 
vehicles into particular groups is presented in its December letter and 
elaborated on in its April 8, 1997 letter to ODI. In the December 
letter, IIHS states, ``Additional model years without known changes in 
the airbag sensor or deployment system are included in the exposure 
counts to provide a complete picture of the relative exposure. Also 
included are corporate ``sibling'' models with the same airbag systems 
during the same model years (for example, Taurus and Sable) for a true 
reflection of the exposure of each airbag system.'' The April letter 
provides more a specific description of the reasons for the choice of 
groups for the subject Taurus, Corsica and Chrysler minivans vehicles. 
ODI has reviewed the IIHS methodology and finds it to be a reasonable 
approach for evaluating the relative performance of the subject 
vehicles compared to other air bag-equipped vehicles.
    The air bag fatality rates for various vehicles as stated below are 
derived from a small number of fatalities spanning as much as 6 years. 
To expand on this further, consider that in just calendar year 1995 
alone,13 NHTSA's Fatal Analysis Reporting System (FARS) 
files indicate a total of 41,798 traffic crash fatalities from all 
causes. Of these, NHTSA is aware of 12 air bag-related fatalities. The 
total traffic fatality rate in 1995 for all causes is 212.6 fatalities 
per million registered vehicles. The specific

[[Page 41481]]

air bag-related fatality rates for the subject vehicles compared with 
various other vehicles is discussed below.
---------------------------------------------------------------------------

    \13\ The agency's latest full year of fatal crash data is 
calendar year 1995.
---------------------------------------------------------------------------

    Adjusting for exposure, the subject vehicles do not have the 
highest driver air bag-related fatality rates, even though they have 
more individual fatalities than other vehicles. For example, the MY 
1990-93 Taurus shows a lower driver air bag fatality rate per million 
registered vehicle years (0.80) than the MY 1990-91 Cadillac El Dorado 
(4.57), MY 1990-92 Pontiac Firebird (2.20), MY 1993-94 Toyota Tercel 
(1.82), MY 1989-93 Dodge Daytona (1.16), MY 1994 Ford F150 (0.95) and 
the MY 1990-96 Mazda Miata (1.28). Similarly, the rate of the MY 1991-
93 Corsica (2.18), while higher than that of the Taurus, is lower than 
that of the El Dorado and Firebird. Thus, the data indicate that the 
rates for many vehicles, while subject to the uncertainties due to the 
extremely limited amount of data, are as high or higher than those of 
the subject vehicles. Furthermore, adding corporate siblings to the 
rate calculation for each of the above subject vehicles brings down the 
rate for each of them, since there have been no air bag-related 
fatalities in either of those siblings.
    The IIHS analysis of child passenger deaths (excluding rear-facing 
infant seats because the injury mechanism is different) reveals that 
models other than the Chrysler minivans present the highest air bag-
related fatality rates per million registered vehicle years. The 
highest is the MY 1995-96 Hyundai Sonata (29.01) followed by the MY 
1995-96 Isuzu Trooper (22.94), the MY 1995-96 Hyundai Accent (13.32), 
the MY 1995-96 Toyota Avalon (8.86), the MY 1993-94 Lexus LS400 (7.41), 
the MY 1995 Geo Metro (6.90), and the MY 1995-96 Mazda Protege (6.76). 
The highest rate for one of the subject Chrysler minivans is 6.67 
deaths per million registered vehicle years for the MY 1996 Dodge 
Caravan and the rate is 3.48 for the MY 1996 subject minivans as a 
whole. The rate for the MY 1994-95 Dodge Caravan is 3.88 and for the MY 
1994-95 Plymouth Voyager is 2.52, while the overall rate for the MY 
1994-95 Chrysler minivans is 3.06. Again, as with the driver air bag-
related fatalities, the data indicate that the rates for many vehicles, 
while subject to the uncertainties due to the extremely limited amount 
of data, are as high or higher than those of the subject vehicles.
    Using the known deaths and the registered vehicle years for each 
model, and the grouping of models, model years and siblings as listed 
in the IIHS letter, NHTSA applied a statistical test to ascertain 
whether any of the subject vehicles is over-represented compared with 
all other vehicles (compared as a single group) having at least one air 
bag-related fatality. The data do not demonstrate that any subject 
vehicle is over-represented.
    IIHS also provided crash data concerning the likelihood of the air 
bags in the subject vehicle deploying in crashes, compared with the air 
bags in other vehicles. Based on insurance crash data, the rates of air 
bag deployments per 100 collision claims in frontal crashes are 
essentially the same for the subject vehicles as for several other 
identified models. In its December 20, 1996 letter, the IIHS reports, 
``The deployment rates per 100 collision claims for 1990-93 Ford 
Taurus/Mercury Sable models are no different from those for the 1992-96 
Toyota Camrys or 1994-96 Honda Accords.'' The rate per 100 collision 
claims for the MY 1990-93 Taurus/Sable is 12, for the MY 1991-93 
Corsica /Beretta is 10, for the MY 1992-96 Camry is 12, and for the MY 
1994-96 Accord is 12. The same holds true for the Chrysler minivans. 
The rate per 100 collision claims for the individual models of the MY 
1994-95 Chrysler minivans ranges from 6 to 9 and for MY 1996 Chrysler 
minivans ranges from 8 to 11. By comparison the rate for the MY 1995-96 
Ford Windstar is 22 (The IIHS did not provide data for any other non-
Chrysler minivan).

Findings

    1. When adjusted for exposure, the air bag-related fatality rates 
for the subject vehicles are not statistically different from the air 
bag-related fatality rates for non-subject vehicles.
    2. NHTSA's SCI reports indicate that the average change in velocity 
for the crashes leading to air bag-related fatalities in the subject 
vehicles is not significantly different from the change in velocity in 
crashes in which there were air bag-related fatalities in non-subject 
vehicles.
    3. NHTSA's SCI reports indicate that in the air bag-related fatal 
crashes involving the subject vehicles, the average change in velocity 
is within the design range specified by each manufacturer.
    4. The NASS data indicates that the air bags in many non-subject 
vehicles have deployed in crashes having a 7 mph or less change in 
velocity. Those crashes involved over 70 non-subject model vehicles.
    5. IIHS data show that the subject vehicles have rates of air bag 
deployments per 100 collision claims that are similar to that of many 
other vehicles.
    Based on the information available at the present time, there is no 
reasonable possibility that an order concerning the notification and 
remedy of a safety-related defect in the 1990 through 1992 Ford Taurus, 
the 1991 and 1992 Chevrolet Corsica, or the 1994 through 1996 Chrysler 
minivan vehicles would be issued at the conclusion of an investigation. 
Therefore, in view of the need to allocate and prioritize NHTSA's 
limited resources to best accomplish the agency's safety mission, the 
petition is denied.

    Authority: 49 U.S.C. 30162 (d); delegations of authority at CFR 
1.50 and 501.8.

    Issued on: July 28, 1997.
Kenneth N. Weinstein,
Associate Administrator for Safety Assurance.
[FR Doc. 97-20295 Filed 7-31-97; 8:45 am]
BILLING CODE 4910-59-P