Motor Vehicle Safety: Information on Recent Controversy Between NHTSA and Consumer Group

Published by the Government Accountability Office on 1990-09-27.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

            1.                                                            . .   I   --
                        United   States   General   Accounting   Office
                        Report to the Chairman, Committee on
GAO                     Energy and Commerce, House of

September        1990
                        Information on Recent
                        Controversy Between
                        NHTSA and Consumer

                   United States
GAO                General Accounting  Office
                   Washington, D.C. 20648

                   Resources, Community,   and
                   Economic Development    Division


                   September 27,199O

                   The Honorable John Dingell
                   Chairman, Committee on Energy and Commerce
                   House of Representatives

                   Dear Mr. Chairman:

                   As requested in your letter of November 7, 1989, and subsequent discus-
                   sions with your office, this report provides information on various alle-
                   gations involving the Department of Transportation’s National Highway
                   Traffic Safety Administration (NHTSA) and the Center for Auto Safety
                   (Center). NHTSA is responsible for ensuring highway and vehicle safety;
                   the Center is a private consumer group concerned with vehicle safety
                   issues. You enclosed a copy of the Center’s September 1989 Lemon
                   Times newsletter that alleged that NHTSA had allowed unsafe vehicles to
                   remain on the roads. The article listed 25 defect investigations involving
                   roughly 37 million vehicles that NH-KU’S Office of Defect Investigation
                   (ODI) had asked manufacturers to recall voluntarily. Later, NETSA closed
                   these investigations without ordering a recall. Also, the Center stated
                   that the Congress should provide judicial review of NHTSA decisions to
                   deny petitions to open safety investigations, or to close safety defect
                   investigations without a recall. Largely in response to this article, 50 ODI
                   employees sent a letter to the Center defending NH--EN’S decisions to close
                   the 25 defect investigations and charged the Center with eroding public
                   confidence in NHTSA.

                   We focused our review on (1) the reasons for the controversy between
                   NHTSAand the Center over the 25 closed safety defect investigations, (2)
                   the issues surrounding proposed judicial review of NHTSA’S decisions to
                   deny petitions to open safety defect investigations or close investiga-
                   tions without a recall (nonenforcement decisions), (3) whether the
                   Center’s activities hindered ODI’S work performance or damaged public
                   confidence in NHTSA, and (4) whether the Center’s sale of data obtained
                   from NHTSA is legal, and whether the Center receives preferential treat-
                   ment in obtaining those data.

                   ODI  employees said that they responded to the September Lemon Times
Results in Brief   article because the director of the Center had visited ODI before the
                   article was published, and the employees believed that a number of their

                   Page 1                         GAO/WED-W221   NHl’SA/Consumer   Group (bntroversy

differences had been resolved. ODI employees saw the subsequent publi-
cation of the article as a betrayal because it did not portray ODI'S investi-
gative process accurately. A particular misunderstanding concerns the
role of the recall request letter, which is the letter ODI sends to manufac-
turers asking them to recall a vehicle model voluntarily.

Another major source of conflict between ODI and the Center concerns
the role of the recall request letter. In its Lemon Times article, the
Center noted that ODI sent a recall request letter for each of the 25 inves-
tigations and later closed these investigations without ordering a manu-
facturer recall. According to ODI officials, however, the recall request
letter is often followed by ODI'S obtaining additional information from
the manufacturer or conducting its own tests. ODI then uses this addi-
tional information to decide whether the safety defect warrants further
investigation or action. NHTSAclosed 21 of the 25 safety defect investiga-
tions cited in the article because the agency decided that the safety risks
were insufficient to warrant pursuing the matter any further. NHTSA
closed one investigation because it lost a federal court appeal, and the
other three because they were similar to the case NHTSA lost (see app. I).

A 1985 Supreme Court ruling (Heckler v. Chaney) set a general pre-
sumption of unreviewability for federal agency nonenforecment deci-
sions. NHTSA'S denying a petition to open a particular safety defect
investigation fits within this category. A 1988 District of Columbia Cir-
cuit Court case (Center for Auto Safety v. Dole) applied the Heckler v.
Chaney ruling to a NHTSAdecision to deny a petition to open a defect
investigation, finding it inherently unreviewable because NHTSA’S gov-
erning statute and regulations provide a reviewing court “no law to
apply.” The Center supports legislation under which the court could
examine the substance of such denials to determine if they were “arbi-
trary and capricious.” Currently, enacting legislation to change NHTSA'S
governing statutes is the only way to permit review of NHTSA’Snonen-
 forcement decisions under the “arbitrary and capricious” standard. We
 found no court decisions discussing the right to judicial review of deci-
 sions to close investigations without a recall.

 We found no evidence that the September controversy hindered ODI
 employees from performing their work. ODI officials believe, however,
 that the Center’s activities over the past several years have caused a
 loss of public confidence in NHTSA, resulting in fewer calls to NHTSA’S
 Auto Safety Hotline on potential safety defects, and fewer returns of
 0~1’s vehicle owner questionnaires. Our examination of 6 years of ODI
 statistics found that questionnaire returns have declined from a high        in

 Page 2                        GAO/RcEDW221    NEISA/Consumer   Group Controversy

             1987, and that Hotline defect calls have been declining since early 1989.
             There is no clear evidence, however, to indicate what role, if any, the
             Center’s activities may have played in this decline.

             The Center does not violate any laws by selling information it receives
             from NHTFA and does not seem to be treated differently from other
             public interest groups requesting information from MIT%. According to
             its director, the Center does not charge the general public for its infor-
             mation packets on specific safety defects and recalls, but does charge
             reporters, attorneys, and others for a complete list of defect complaints.
             The price roughly covers the Center’s costs.

             Under the National Traffic and Motor Vehicle Safety Act of 1966, as
Background   amended, NHTSA has the authority to recall motor vehicles found to have
             a safety defect that poses an unreasonable risk of accident. From 1967
             through 1989, approximately 140 million vehicles have been recalled for
             safety defects, many as a direct result of NWIW’Sinfluence.

             The Center for Auto Safety is a nonprofit consumer group founded by
             Ralph Nader and the Consumers Union in 1970.* The group has been
             independent of its founders since 1972. The Center has been both a
             major critic of NHTSA and a frequent user of NHTSA information and data.
             The Center also supplies information to NHTSAfrom consumer com-
             plaints it receives on possible vehicle defects.

             In an article in its September 1989 Lemon Times, the Center accused
             NHTSAof allowing vehicles it knows to be unsafe to remain on the roads.
             The article lists 25 investigations involving roughly 37 million vehicles
             in which NHTSA asked for a voluntary manufacturer recall and later
             closed the investigation without a recall. The Center contends that the
             Congress should provide for judicial review of agency decisions to close
             safety defect investigations (such as the 25) without a recall or to deny
             petitions to open defect investigations. Fifty NHTSAODI employees signed
             a response letter to the Director of the Center defending NHTSA’S record
             on the 25 cases and charging the Center with undermining the defect
             investigation program by eroding public confidence in NHTSA.

             ‘The Consumers Union is a nonprofit organization established in 1936 which provides mfomuuon to
             consumers on goods, services, health, and personal finance.

             Page 3                               GAO,‘RCED9&221 NWI%/Conmme            r Group (bstmwwny

                              officials and employees told us that although they expect criticism
Several Reasonsfor        ODI
                          from outside groups in the course of their work, the Lemon Times article
the NHTSA/Center          provoked a response because the Director of the Center had visited ODI
Controversy               prior to its publication. ODI employees thought they had reached a con-
                          sensus on several issues of disagreement during this visit, but the crit-
                          ical article came out shortly afterwards. NHTSA and the Center continue
                          to differ on their interpretations of the recall request letter and its role
                          regarding the 25 defect investigations cited by the Center. Beyond this,
                          the different roles and responsibilities of NHTSA and the Center in pro-
                          moting auto safety cause a continuing tension between the two groups.

Center Director’s Visit   The main catalyst for writing a letter responding to the Lemon Times
                          article was the fact that the Director of the Center had visited ODI
Provided Impetus for      shortly before its publication to resolve differences between the two
NHTSA ResponseLetter      groups. ODI officials invited the Director to visit the agency and talk to
                          ODI staff about how the two groups could work together more effec-
                          tively, hoping to arrive at a compromise in regard to information
                          sharing. Specifically, ODI employees have been trying to get the Center to
                          standardize the information contained in consumer complaints that it
                          forwards to ODI, thereby saving what staff characterize as many extra
                          hours of work tracking down additional information and weeding out
                          duplicate complaints.

                          ODI employees told us that after the meeting, they were left with the
                          impression that the Center would be more willing in the future to work
                          with the agency for their mutual goal of promoting vehicle safety.
                          Hence, the subsequent publication of the Lemon Times article caused
                          many employees to feel betrayed because they believed the Director
                          deliberately misrepresented 0~1’s investigative process.

Recall Request Letter’s   As outlined in ODI’S formal procedures, a request letter for voluntary
                          recall is sent when ODI investigators have made a preliminary determi-
Role in NHTSA’s           nation that a defect exists. This determination is made during the engi-
Investigative ProcessIs   neering analysis phase of an investigation. A recall request letter is only
Disputed                  one step in an ongoing safety investigation, and subsequent information
                          from the manufacturer or from ODI’S own testing may lead investigators
                          to conclude that the defect does not pose an unreasonable risk of acci-
                          dent. Also, a manufacturer may take other action, short of a formal
                          recall, that satisfies ODI investigators (See app. 11for a summary of !)DI’S
                          investigative procedures).

                          Page 4                        GAO/WED-90-221 NHl’SA/Consumer   Group (bnuoversy

                       One ODI official acknowledged that the decisions made in regard to the
                       25 cases cited by the Center were very difficult, but that the process
                       depends on allowing engineers, managers, and legal staff to make the
                       hard decisions on the basis of their professional experience, knowledge,
                       and expertise. In each of the 25 safety defect investigations cited by the
                       Center, NHTSA sent a voluntary recall request letter to the manufacturer
                       and, after further information-gathering and investigation, later closed
                       the investigation without a recall. According to ODI closing documents,
                       21 of the 25 investigations were closed because the alleged defect did
                       not pose a safety risk significant enough to warrant further action. One
                       was closed because NHTSAlost a federal court appeal, and the remaining
                       three were closed because they were similar to the case NHTSA lost in
                       federal appeals court (see app. I).

                       For example, four of the defects in question resulted in a safety failure
                       early in the life of the vehicle. ODI officials said these types of defects
                       will not have a significant safety impact because the manufacturer most
                       likely will have repaired or replaced the defective part before a recall is
                       necessary. Also, the manufacturer will have changed the design on later
                       models to prevent future failures. In nine other investigations, manufac-
                       turers took corrective action, such as a partial recall or a service cam-
                       paign which ODI deemed sufficient to minimize any safety risk. Three
                       other investigations concerned defects that might cause a gradual equip-
                       ment failure, with warning to the driver. ODI officials consider gradual
                       equipment failure with some type of warning to be a mitigating factor
                       when evaluating potential safety risk. Each of these investigations
                       involved use of formal procedures, analysis, and professional judgment
                       by NHTSA'S engineers and attorneys that a recall was not warranted.

NHTSA and the Center   NHTSA  and the Center play different roles in promoting auto safety. As a
                       government     agency, NHTSA is bound by its authorizing legislation, which
Have Different Auto    directs it to find and remedy defects that pose an unreasonable risk to
Safety Roles and       safety. The definition of “unreasonable risk,” however, is left open, and
Responsibilities       the tension between NHTU and the Center arises in part because of dif-
                       ferences in the way each interprets this standard. The consumer-
                       oriented Center is more inclined to consider the existence of any defect
                       to be “unreasonable,” while NHTSA takes into account such factors as the
                       severity of safety hazards caused by the defect, and how much warning
                       the driver receives before a significant safety failure occurs.

                       Page S                        GAO/BCEJM@221 NETSA/Consumer   Group   (‘ontmversy

                     The Lemon Times article expresses the Center’s support for legislation
Judicial Review of   to permit judicial review of NI-ITSA decisions to deny petitions to open
Federal              defect investigations, or to close investigations without recalls. The
Nonenforcement       article discusses several high-profile investigations and lists the 25 that
                     NHTSA closed without a recall to demonstrate the need for such legisla-
Decisions            tion. The Senate bill cited, however (S. 673), referred only to judicial
                     review of NHTSA decisions to deny petitions to open defect investigations.
                     Currently, NHTSA’S governing statutes would have to be changed to
                     permit either type of review.

                     NHT% is almost completely exempt from review of its decisions to deny
                     petitions to open defect investigations since a 1985 Supreme Court
                     ruling that a federal agency’s nonenforcement decisions are not review-
                     able by the courts unless an agency’s statute provides for it (Heckler v.
                     Chaney).2 Heckler v. Chaney also found that the general presumption of
                     unreviewability did not apply where (1) the agency’s inaction was based
                     on a conclusion by the agency that it did not have the statutory
                     authority to deal with an issue, (2) the agency’s policy was so extreme
                     as to amount to an abdication of its statutory responsibilities, or (3) con-
                     stitutional rights are being violated. The District of Columbia Circuit
                     Court applied the principles set forth in Heckler v. Chaney when it spe-
                     cifically held in 1988 that nothing in NHTSA'S governing statute or regu-
                     lations allowed for judicial review of a NliTSA decision to deny a petition
                     to open a safety defect investigation (Center for Auto Safety v. Dole).3
                     None of the three exceptions outlined in Heckler v. Chaney were found
                     to be applicable in this case.

                     The Director of the Center for Auto Safety said he supports judicial
                     review of NHTSA nonenforcement decisions using the “arbitrary and
                     capricious” standard set out in Title 5 [6 U.S.& sec. 706 (2)(a)]. To this
                     end, the Center supported a judicial review provision included in S. 673
                     (National Highway Traffic Safety Administration Authorization Act of
                     1989), as approved by the Senate Committee on Commerce, Science and
                     Transportation. The judicial review provision was removed from the bill
                     on the Senate floor, and the bill as amended was passed by voice vote on
                     August 3, 1989. The Center for Auto Safety seeks to have the provision
                     reinstated in the House version.

                     2Heclder v. Chaney, 470 U.S. 821(1986).

                     3C.enterfor Auto Safety v. Dole, 846 F.2d 1532 (DC. Cir. 1988).

                     Page 6                                 GAO/WED-XL221        NHTSA/Consume r Group ( mvmrrny

The judicial review provision contained in S. 673, however, referred
only to agency denials of petitions to open investigations, not to agency
decisions to close an investigation without a recall. We found no court
decisions discussing the right to judicial review of decisions to close
investigations without a recall. The Director of the Center pointed to a
 1970 suit brought by the Center in which the group had sued over just
such an issue and obtained a preliminary injunction against NHTSA The
Director considers this suit precedent for allowing judicial review.
Before the suit was decided, however, the manufacturer recalled the
vehicles in question voluntarily. Therefore, the court did not reach a
conclusion on this issue. We believe that the 1970 suit does not set a
precedent for judicial review of NHTSA's decisions to close an investiga-
tion without a recall.

Department of Transportation officials oppose judicial review of NHTSA
petition denials because they fear an onslaught of litigation over investi-
gations they are not pursuing that will force them to divert resources-
engineers and lawyers- away from the problems they do choose to
investigate. The Director of the Center believes that the “arbitrary and
capricious” standard is tough enough to prevent a large influx of court
cases against NHlS4, and he said that previously, the Center sued NHTSA
over a petition denial only once (the l-988 Center for Auto Safety v. Dole
case). The Senate Committee report on S. 673 confirms that only one
NHTSA petition denial has been challenged in court since the National
Traffic and Motor Vehicle Safety Act of 1966 was amended in 1974 to
provide for citizen petitions, among other changes. The Director of the
Center told us that the real worth of a judicial review provision is that it
forces NHTSA to consider possible public opposition when making deci-
sions not to open certain safety defect investigations.

As it stands, the Congress would have to change the law to provide for
judicial review of NHTSA’S decisions to deny a petition to open a defect
 investigation, or to close an investigation without a recall. As noted
 above, agency nonenforcement decisions are not reviewable since
Heckler v. Chaney unless set out in agency statute or regulation. The
Senate has considered and rejected legislation providing for such
 reviews of NHTSA.

Page 7                        GAO/RCJ3lM@221 NEllSA/Conmmer   Group Controvemy

                     The controversy surrounding the September Lemon Times article and
Impact of Center’s   ODI employees’ letter to the Center did not hinder ODI employees in their
Actions on NHTSA’s   work. ODI officials continue to believe, however, that the Center’s activi-
Work                 ties over the past few years have reduced public confidence in NHTSA,
                     resulting in fewer calls to the agency to report potential safety defects
                     and fewer returns of ODI’S vehicle owner questionnaires. We found that
                     certain indicators ODI officials use to measure public confidence in NHTSA
                     have been decreasing recently, but there is no clear evidence linking this
                     drop to the Center’s activities.

                     Federal government employees are subject to review and criticism of
                     their work by the public, and ODI employees are often criticized by the
                     Center and other groups. When we spoke with ODI employees, we found
                     that while some employees were angry about what they believed were
                     distortions of their work, most said they recognize the legitimate role the
                     Center plays as consumer advocate. Also, there was a consensus among
                     the employees that the particular September 1989 Lemon Times article
                     was no worse than previous or subsequent criticism from the Center.

                     We found no evidence that publication of the article hindered ODI
                     employees from performing their work. On the contrary, several told us
                     that they were motivated to work even harder. One member of the com-
                     plaint-screening staff said that he “took it as a challenge,” and another
                     said that she responded by taking more time with public callers so that
                     they would have a good impression of the agency. The incident seemed
                     to improve office morale, and the employees generally believed the
                     resultant publicity was good for NHTSA because it raised public aware-
                     ness of the agency’s existence and mission.

                     In their September letter to the Center, 50 ODI employees alleged that
                     their work has been hindered by a loss of public confidence in the
                     agency. ODI officials fear that if the public loses confidence in NHTSA,
                     fewer people will report potential safety defects. Several 0~1 employees
                     stressed the fact that their work depends on reports of safety defects
                     from the public for early clues to potential auto safety problems, via
                     their Auto Safety Hotline and Vehicle Owner Questionnaires. To support
                     their case that the Center’s activities in recent years, including the Sep-
                     tember article, have had a negative effect on public confidence in NHTSA,
                     ODI officials gave us statistics on the number of calls to the Auto Safety

                     Page 6                        GAO/BCZBW221   NHlSA/Consumer   Group Controversy

Hotline and the number of Vehicle Owner Questionnaires returned.4 ODI
officials use these numbers as indicators of public willingness to contact
NHTSAabout auto safety problems.

We analyzed ODI’S numbers and found that questionnaire returns are
down substantially since 1987. It should be noted, however, that in
1987, questionnaire returns were much higher than in any other year
from 1984 to 1989 (see fig. 1). ODI officials suggested that several high-
profile defect investigations in 1987 may have caused the high return
rate, but they could not pinpoint any specific reason for the increase. ODI
officials continue to be concerned by the numbers for the first 6 months
of 1990, which are lower than those for the same months in 1989.5 We
found no clear evidence linking the fluctuations in questionnaire returns
to the Center’s activities.

4NHTSA sends out a questionnaire to every person who calls with a complaint about a safety tlrt&
or an information request on a recall effort. The questionnaire asks for specific informatlon aknjut the
complainant’s allegedly defective vehicle, for use in safety defect investigations.

51nresponse to what they see as a sharp decrease in questionnaire returns, ODI officiab ha\ v twgun
sollclting returns over the phone (“ekctronlc” returns). Ekctronic returns have ken PCPI\ t-d In the
months of April, May, and June 1990. We did not include these electronic returns m our analysts If
we had included them, the number of questionnaire returns would have been slightly h+$wr m I hese
3 months than in the same months in 1989.

Page 9                                   GAO/RCED90-221 NHlSA/Consumer              Group (‘~mtmversy

Figure 1: Vehicle Owner Questionnaire
Returns, 1984-89                        16   VOO Rstums     (bmaands)

                                        Source: 001 data.

                                        ODI officials also were concerned that while Hotline calls were up sub-
                                        stantially in 1989, those calls relating to safety defects were down. We
                                        found that Hotline calls did rise substantially in 1989, most likely
                                        because of a controversy over child safety seats in December, during
                                        which ODI received substantially more hotline calls than in previous
                                        months. Because ODI employees do not send a vehicle owner question-
                                        naire in response to every call, it is difficult to make a direct comparison
                                        between the increased number of calls and the number of questionnaire
                                        returns. Even without reference to total Hotline calls, however, it
                                        appears that calls on safety defects are dropping (see fig. 2). Since these
                                        calls are often 0~1’s first tip on potential safety problems, any sustained
                                        decrease is cause for concern among ODI officials. Again, we found no
                                        clear link between the Center’s actions and the drop in defect calls.

                                        Page 10                         GAO/EEBWM21   NElSA/Consumer   Group   Cmaroversy
Figure 2: Number of Calls to NHTSA’s Auto Safety Hotlhw Reporting Potential Safety Defects, Jan. 1989-June 1990
2060 Dsfutcsus





 606                                                                                                                     ---



   lm#    2189     m9   me    was    6/w     rm       Bma      San   1omB   11188   1211)s lrn    2m      s/o0   us0    6m6m
                                           Source: ODI data.

                                           The Center’s sale of information available from NHTSA is not illegal. If a
Center’s Use of                            third party is willing to pay the Center for information that one can get
NHTSA Information                          free from NHTM, there is no legal prohibition on the Center’s selling it to
                                           the third party.

                                           The Center receives information from NHTSA by (1) reviewing the public
                                           files kept in W’S    Technical Reference Library and noting or copying
                                           the needed information and (2) filing a Freedom of Information Act
                                           (IUA) request with NHTU’S Executive Secretariat. Although fees are
                                           associated with obtaining information by FUA request, the Department
                                           of Transportation waives all fees that total less than $10.

                                           While the Director of the Center said that the group often pays FOIAfees,
                                           NHTSAofficials told us that in most cases, the Center also gets a waiver
                                           for costs in excess of $10 under a section of the Department of Trans-
                                           portation’s VIA regulations which states that fees may be waived if dis-
                                           closure of the information “is in the public interest because it is likely to
                                           contribute significantly to public understanding of the operations or
                                           activities of the government and is not primarily in the commercial
                                           interest of the requester” (49 C.F.R. Subtitle A, section 7.97 (e)). The

                                           Page 11                              GAO/lUXD9&221    NJTl’SA/Consnmer Group (‘mwoversy
Center does pay for a yearly computer search and downloading of ODI’S
complaint database,including names and addressesof complainants.~
The Center does not seemto be treated differently from other nonprofit
or public interest groups that request NITI%information, although it is
the most frequent requester of information from the Office of Defect
Investigation. ODI officials estimated that 90 percent of the time that
employeesspend answering public inquiries is devoted to the Center. ODI
officials also noted that the type of inquiry the Center makes generally
requires the time of technical staff, not Hotline or other information-
In regard to NXArequests, the Center is by far the largest public interest
group requesting this type of information, making it difficult to compare
its treatment with that of similar groups. ODI officials told us that, since
the Center requests FDIAinformation in its complaint data baseso fre-
quently, the information is sometimes releasedto it without going
through the formal FUIAprocess.It should be noted, however, that auto
manufacturers also are granted this informal waiver. No other group
requests information as frequently as the Center and the manufacturers.

The Director of the Center says that the group does not charge the gen-
eral public for consumer packageson auto safety defects and recalls
compiled from information obtained free from NITISLThe Center asks
that information requests be in writing and that the requester include a
self-addressedenvelope with 46 cents postage.According to the
Director, the Center sendsout about 60,000 information packagesevery
year. The packagesinclude referrals to NITW’Stechnical reference divi-
sion (which charges a fee for searcheson specific defects and recalls)
and the Auto Safety Hotline if additional information is needed.
Information sold to reporters, attorneys, and others from ODI’S complaint
data basecosts the Center $80 per search after the initial, annual
downloading. The Director of the Center says that those who request
this information are charged the search fee plus a postage and handling
fee of $20. He claims the Center takes a loss much of the time even with
this fee schedule and, in someinstances (e.g., to public interest research
groups), provides the information free.

?he Center files a FDIA each year to obtain the names, addresses, and specitic complamts of those
people who have contacted NJTISA in the preceding year. The Center hires a contractor w go through
NH’ISA’s files and compile this information into a computer data base. K&i charges and (-ontractor
fees total about $2,000. Subsequent searches of that data base cost the Center $80 each

Page 12                               GAO/WED-W221        NE’ISA/Conmuner     Group (‘cmtroversy

              To determine the reasons for the conflict over the 25 safety defect
Scopeand      investigations cited by the Center, we met with NHTSA and ODI officials
Methodology   and the Director of the Center. We reviewed and analyzed the closing
              documents for the 25 investigations and the recall request letters for 20
              of the 25.7 Also, we interviewed three panels of ODI employees, including
              (1) seven employees from the complaint-screening staff, (2) four
              employees from the engineering analysis staff, and (3) seven senior ODI
              officials (one of the officials was also a member of the complaint-
              screening staff panel). These interviews helped us determine how ODI’S
              work had been affected by the controversy. In addition, we analyzed
              Auto Safety Hotline calls received by ODI from January 1984 through
              June 1990, and the number of Vehicle Owner Questionnaires returned
              between January 1984 and June 1990 to determine if the Center’s activ-
              ities had hindered ODI’s work by unde rmining public confidence in NHTSA.
              We reviewed Department of Transportation FOIAregulations and spoke
              to NI-I'ISA officials responsible for the disposition of FOIAand other infor-
              mation requests to determine how the Center compared with other
              public interest groups in requesting information from NHTSA.Finally, we
              examined legal matters principalIy relating to judicial review, but also
              encompassing the legality of the Center’s sale of NHTSAinformation. It
              was not within the scope of our review to determine whether NH-ISA’S
              decisions in regard to the 25 safety defect investigations were correct.

              We discussed the information in this report with NHTSAofficials and the
              Director of the Center, who agreed with the facts presented. As
              requested, however, we did not obtain official agency comments on a
              draft of this report. Our work was conducted primarily between
              December 1989 and June 1990 in accordance with generally accepted
              government auditing standards.

              As arranged with your office, unless you publicly announce its contents
              earlier, we plan no further distribution of this report until 30 days after
              the date of this letter. At that time, we will provide copies to the Secre-
              tary of Transportation, the Administrator of NHTSA,    the Director of the

              WHTSA officials were unable to provide recall request lette13 for five of the inveS$atbm   GUIstated

              P8ge 13                                 GAO/RCEDWPl         NEISA/ckmsnme      r Gruap I-aawrrsy

Center for Auto Safety, and any other interested parties. If you have
any questions about this report, please contact me at (202) 275-1000.
Major contributors to this report are listed in appendix III.

Sincerely yours,

Kenneth M. Mead
Director, Transportation   Issues

 Page 14                     GAO/lUXB~Z21   NElSA/Consnmer   Groap (bnmversy
Page 16   GAO/RCE@9M!21 IWR3A/Consnme   r Gmap   (batrovemy

Appendix I                                                                         18
Details on the 25
Investigations Cited
by the Center
Appendix II                                                                        22
Office of Defect        Preliminary Evaluation
                        Engineering Analysis
Investigation’s         Formal Case                                                24
Procedures for
Conducting Safety
Defect Invktigations
Appendix III                                                                       25
Major Contributors to
This Report
Related GAO Products                                                               28

Figures                 Figure 1: Vehicle Owner Questionnaire Returns, 198489
                        Figure 2: Number of Calls to NHT!W’s Auto Safety
                            Hotline Reporting Potential Safety Defects, Jan.
                            1989-June 1990


                        Dar       Department of Transportation
                        EA        engineering analysis
                        NnA       Freedom of Information Act
                        GAO       General Accounting Office
                                  National Highway Traffic Safety Administration
                        ODI       Office of Defect Investigation
                        PE        preliminary evaluation
                        VOq       Vehicle Owner Questionnaire

                        Pa@! 16
Page 17   GAO/RCED9@221 NEl’SA/Conmmer   Group Controversy
Appendix I

Details on the 25 Investigations Cited by
the Center

                                   ,. _.,, - _,,,   -_   ,.                                ..,,                    -_      -
             l$wt;ffice   action
File order                           Affected vehicles               Alleged defect
1            EA78-118                1975-84 Ford trucks & vans      Loose or missing stud nuts & broken wheel studs; can result m a
             C85-10                                                  set of dual rear wheels falling off

2            EA79-057                197578 Cadillacs                Headlight adjustor failures; headlight drops down or to the side
                                                                     when plastic adjusting nut breaks; some visibility problems no
3            EA79-079                1974-79 Ford vehicles           Loss of power steering/brake assist; increases force/time needed
                                                                     for steering/braking

4            EA80-009                1976-78 Chevettes               Tire separates from rim, resulting in a flat or complete detachment;
                                                                     potential loss of vehicle control
5            EA80-085                1978-79 Cadillac Sevilles       Broken wire wheel spokes cause flat or wheel separation; potential.
                                                                     loss of vehicle control

6            EA80-024                1980-85 GM X-Body vehicles      Rear brake lock-up; potential loss of control
             EABO-112                197980 Dodge and                Failure of hatchback support; door falls without warnrng
             C82-18                  Plymouth Horizons
             EA81-015                197880 Chevy Malibu, Monte Carlo & El      Rear axle separation
             C82-03                  Camino, Pontiac Lemans & Grand Prix, Olds.
                                     Cutlass & Cutlass Supreme, Buick Century
                                     & Reaal. GMC Caballero
9            EA81-010                197980 Mustang and Capri                   Failure of hatchback support; door falls wlthout
             C82-19                                                             warning
10           E82-028                  1981-82 Ford Escort, EXP & Mercury Lynx,        Engine compartment       fires

11           EA83-014                 1978-82 Chrysler vehicles with 1.7L and 2.2L Carburetor isolator failure; results in stalling
                                      engine     .
12           EA80-100                 1980 GM X-Body vehicles                      Loss of power brake assist; increased force/
             c84-01                                                                distance needed for braking

13           EA83-013                 198182 Ford vehicles                            Front seatback failure

14           E80-087                  197980 Ford Mustang & Mercury Capri             Rear brake lock-up; potential loss of control
15           EA84-019                 1982 Chevy Camaro and Pontiac Firebird          Failure of rear seatbelt anchorage

16           EA84-028                 1981-82 diesel Chevy Chevettes                  Rear brake lock-up; potential loss of control

                                             Page 18                              GAO/ECJCD-9@221NE’l’SA/Consnmer          Group (imtrovcrsy
                                          Appendix I
                                          Detaila on the 2ti Investigationa   Cited by
                                          the Center

Date defect office   Da;e~bcall request      Date defect off ice
opened actionb                               closed actionb              Reason/s defect office closed action
09/l 4178            10/31/84                08/02/88                    1. No difference seen between this vehicle and total vehrcle
                                                                            population-no     unreasonable risk
                                                                         2. Complaints down
                                                                         3. New ‘design
03/l 8179            02/06/80                1O/27/81                    1, No unreasonable risk-unlikely     that two headlight adjusters
                                                                            will fail at once
                                                                         2. New design
05/l 8179            04/03/80                04/21/82                    1. Problem occurs early in life of vehicle-little likelihood of
                                                                            future incidents
                                                                         2. New design on later models
                                                                         3. Complain& down
11/12/79             04/17/80                02/03/82                    1. Slow loss of air, providing warning to driver
                                                                         2. Identifiable problem only in 1976 model, corrected
05/22/80             05/l l/81               1l/04/83                    1. Sufficient driver warning
                                                                         2. Tire air loss usually gradual, vehicle control maintained
                                                                         3. Complaints down
11/26/79             07/06/81                05/18/89                    GM 1980 X-Car case, which DOT lost in federal appeals court

09/02m               1l/20/81                Qxw=                        1. No unreasonable risk
                                                                         2. Complaints down
                                                                         3. New design
04/07/81             12/02/8 1               ll/26/84                    1. Safety recall was conducted on 3 million vehicles produced at
                                                                            one plant
                                                                         2. No defect found on vehicles produced at other plants

06/23/81             06/01 f82               01/03/85                    1. Ford took steps to correct problem
                                                                         2. Complaints down
                                                                         3. No unreasonable risk
05/27/82                                     07126184                    1. No specific defect found
                                                                         2. No unreasonable risk
                                                                         3. Symptom occurs early in life of vehicle-little likelrhood of
                                                                            future incidents
                     05/12/83                12/11/88                    1. No unreasonable risk
                                                                         2. New design
07/24/80             08/l 2183               IO/l l/88                   1. Complaints down
                                                                         2. GM conducted an intensive information campargn on
                                                                            servicing these vehicles
                                                                         3. No unreasonable risk
01/12/83             11/01/83                Ol/26/89                    1. Ford conducted a recall that addressed the specrfic problem
                                                                         2. No unreasonable risk on rest of vehicle
05/27/80             11fO3l83                05/l 8189                   Similar to 1980 GM X-body case, which DOT lost In federal
                                                                            appeals court
                                             05/01/85                    1. Complaints down
                                                                         2. Passes dynamic test
                                                                         3. No unreasonable risk                             --        ..__
                                             WW~                         1. Complaints down
                                                                         2. Similar to 1980 X-Body case, which DOT lost In federal
                                                                            appeals court

                                          Page 19                                  GAO/WED-9&221      NEISA/Ckmsumer       Group   (‘mtroversy
                                         Details on the 25 Investigations   Cited by
                                         the Center

             lh3;h3~pffice action
File order                          Affected vehicles                                  Alleged defect
17           EA84-023               1982-85 GM A-Body vehicles                         Rear brake lock-up; potential loss of control
18           EA84-008               Certain GM A-Body vehicles                         Upper control arm mounting failure; possible
                                                                                       steering control problems

19           PE85-047               198285 Cadillac Eldorado                           Door fires

20           EA82-040               197981 Honda Accord & Civic                        Seatbelt retractor failures

21           EA88-002               1982-83 Toyota Celica                              Malfunction of front seatbelts

22                                  Certain 198184 Toyota Cressidas                    Sudden acceleration

23           PE88-051               1985 GM A,C,G,F and N-          Windshield wiper failure
             EA88-025               Body vehicles

24           EA85-041               1985 GM A,C,X-Body              Failure of linear seatback recliners; potential loss of control
25           EA85-018               1978-88 Chevy Chevette,         Crankshaft pulley bolt failure; potential loss of power steering
                                    198188 Pontiac T-1000           assist, engine overheating/stalling

                                          Page20                                  GAO/E~W221            NlI’l?SA/Conenmer Group (bntnwersy
                                           Details on the 25 lnve&igation~      cited by
                                           the Center

Date defect office   Date recall request      Date defect office
opened actionb       isruedb                  closed actionb               Reason/s defect office closed action
12/l 3183            01/03/85                 04/ 18/89                    Similar to 1980 X-Body case, which DOT lost in federal appeals
1l/23/83             01 116185                12/08/85                     1. No unreasonable risk
                     03/25/85 .                                            2. Problem occurs early in life of vehicle-little likelihood of
                                                                              future incidents
01/28/85             09/04/85                 0211a/88                     1. GM instituted a safety recall for defective part or cars with a
                                                                              high fail rate        -
                                                                           2. No iniuries/deaths
                                                                                     .     .     reborted
07/20/82             11/04/85                 05/l 2186                    1. Honda took corrective action
                                                                           2. No unreasonable risk
1o/ 18185            12/05/85                 05/08/86                     1. Toyota conducted service recall
                                                                           2. No unreasonable risk
09/20/83             05/06/86                 02~18~88                     1. Toyota conducted recall to replace potentially defective part
                                                                           2. Besides above part, no other defect found
                                                                           3. Complaints down
04/28/86             09/05/86                 03/02/88                     1. GM issued six service bulletins on repair, which seemed
                                                                           2. No unreasonable risk
                                                                           3. Problem occurs early in life of vehicle-little likelihood of
                                                                              future incidents
                                                                           4. New design on later models
08/l 9185            09/03/86                 11/12/87                     1. No unreasonable risk-seatback       movement is gradual
                                                                           2. New design
03/l 3185            05/09/86                 09/04 /87                    1. No unreasonable risk
                                                                           2. Manufacturer service campaign initiated
                                                                           3. New design
                                           Bathe defect office action numbers in this appendix may differ somewhat from ODI source documents
                                           because the same investigation sometimes was referred to differently in various ODI sources For
                                           example, an investigation might appear as E82-05 in one source, and EA82-005 in another The numbers
                                           do follow a consistent format, however, with a letter(s) to identify the level of investrgation (prelrmrnary
                                           evaluation, engineering analysis, or formal case), a 2-digit number identifying the year in which the
                                           investigation was opened, and a 2- or 3-digit number giving the chronological order wrthrn that year.
                                           bathe dates in this appendix have been compiled from several sources. ODI officials confirmed these
                                           dates, but in some cases, we were not able to document them independently. The dates are not a
                                           matter of dispute between NHTSA and the Center, however, and do not affect the content of thus report.

                                           Page 21                                    GAo/Bclxhm221         NJm3A/conrmme r Group Controversy
Appendix II

Office of Defect Investigation’s Procedures for
Conducting Safety Defect Investigations

                       Office of Defect Investigation (ODI) officials outlined their investigative
                       process, which consists of three main steps, as detailed below:

                       Most of 0~1’s work comes from consumer complaints via the Auto Safety
Preliminary            Hotline, but the agency also gains information on potential safety
Evaluation             defects from letters and from monitoring manufacturer service bulletins
                       (manufacturers are required to send the National Highway Traffic
                       Safety Administration (NI-ITSA)a copy of any service bulletins they send
                       out to their dealers, even if the bulletin is not related to safety). ODIstaff
                       screen the information for patterns and trends that would indicate a
                       potential safety defect. If they see such a pattern, a preliminary evalua-
                       tion (PE) is opened. When a PEis opened, ODIsends a letter to the manu-
                       facturer requesting information on any complaint received on the
                       particular problem noted. Manufacturers are required by law to provide
                       this information (The National Traffic and Motor Vehicle Safety Act of
                       1966 requires manufacturers to notify NHTSAof any potential safety
                       defects, not only those about which NHTSAsends an information
                       request). NHTSAgenerally requests that the information be returned
                       within 6 weeks, but the agency has no power to enforce this time limit.

                       The PEis designed as an information-gathering process. When the manu-
                       facturer responds, ODIstaff analyze the information and determine
                       whether there is a potential safety problem or whether the manufac-
                       turer has satisfactorily addressed its concerns. If the manufacturer’s
                       response is adequate, a PEclosing document is prepared (this is a public
                       document). If ODIstaff believe the problem warrants further investiga-
                       tion, the case is upgraded to an “engineering analysis” (EA). The PE
                       phase of a safety defect case usually takes about 4 months. ODIstaff told
                       us that about one-third of all PEcases go on to the engineering analysis

Engineering Analysis   EA. In this phase, a more lengthy information request is sent to the man-
                       ufacturer. This request asks for information on any lawsuits pending
                       against the manufacturer relating to the potential defect being investi-
                       gated, its own engineers’ testing results, sales figures for replacement
                       parts, subsequent design changes, and the manufacturer’s own appraisal
                       of the potential safety consequences. ODImay also do its own tests at
                       this point to determine whether the defect does have an adverse effect
                       on safety. The tests may be done in-house at NHTSA’Slab in Ohio (which

                       Page 22                        GAO/ItCED~Z21   NHlSA/Cawumer   Group   Controversy
Oflk   of Defect Inveatigation’r ProaWes
for coaeting    safety Defect lnV~tlOM

requires only an internal memorandum) or may be contracted out
(which requires a formal procurement contract).

ODI may have several rounds of correspondencewith the manufacturer
to clarify specific points in the EAphase. Although it is possible that
NHTS will bring something to the manufacturer’s attention of which it
was unaware, generally, the manufacturer has much more information
on a potential safety defect than NI-ITSA. ODI staff gather the relevant
information to determine if a serious problem exists. This task is compli-
cated by the fact that the quality and type of information available vary
among manufacturers. For example, somecompaniesmay have informa-
tion on warranties readily available but no information on replacement-
part sales.

Manufacturers will acknowledge that a defect exists much of the time,
but will often claim it doesnot adversely affect safety. Also, manufac-
turers will argue with ODI over the severity and frequency of the defect.
ODI staff set up their own criteria, on the basis of the warranty, sales,
and internal testing information, to determine if the defect is severe
enough to pose a safety threat. If there is such a threat, ODI issues a
recall request letter asking the manufacturer to recall the model for
repair. A recall request letter is not an order, but a request for voluntary
recall. In the past, NHTSA did not get a very good responseto its recall
request letters. Now, however, manufacturers act on recall request let-
ters much more often and voluntarily recall models containing safety
defects. If they do not, however, NHTSA   must decide whether to bring the
defect investigation up to the next level-a formal defect case.

At this point, it is still possible for manufacturers to provide NH’ISAwith
new or updated information that can changethe status of a case.Manu-
facturers are continually testing and surveying their dealers during the
investigation process,and they have far greater resourcesat their dis-
posal than NH’IW They may comeup with new or more detailed infor-
mation that convincesODI either that the defect in question is not a
safety hazard, or that it has been properly taken care of by the company
(e.g., if the manufacturer has isolated and corrected the problem and
performed a limited recall). If this happens, the casewill be closed and a
closeout memo will be prepared for the public file (not all portions are
public-any internal engineering opinions are deleted from the public
file, as well as any confidential businessinformation). If the manufac-
turer doesnot provide a satisfactory reason for refusing to recall defec-
tive cars voluntarily, NHTSA has the option of opening a formal case

Pyle 28                               GAO/RCED-W221 NETSA/Chmmer   Group ~batnwrmy
              Appendix If
              Off& of Defect Investigation’s Procedures
              for Conducting Safety Defect Investigations

              If NHTSA decides that a recall request has been refused without a good
Formal Case   reason, it will convene a defect review panel consisting of representa-
              tives from ODI, Office of the Chief Counsel, and the Office of the Admin-
              istrator. NHTSA attempts to reach a consensus on the severity of the
              safety problem at hand and the strength of NHTSA’S case if the agency
              were to press the issue in federal court. ODI officials also consult with
              the legal officials informally before the defect review panel convenes to
              get an idea of the strength of any potential court case. In the interest of
              agency credibility, NH%% is trying to send recall request letters only if it
              believes that they have a strong enough case to press the issue in court
              if necessary. NHTSA cannot force a manufacturer to recall vehicles except
              by court order on a case brought by the government. A formal case is
              the final stage of the safety defect investigation process, with the ulti-
              mate decision on recall resting with the federal courts.

              Page 24                               GAO/RCED-W221 NHlBA/Consumer   Group   (im~mversy
Appendix III

Major Conttibutors to This Report

                          John W. Hill, Associate Director
Resources,                Ron E. Wood, Assistant Director
Community, and            J. Erin Ebzik, Assignment Manager
                          Cheryl A. Donahue, Evaluator-in-Charge
Economic - ’              Curtis L. Groves, Senior Operations Research Specialist
Development Division,
Washington, DC.

                          Martin J. Fitzgerald, Special Assistant to the General Counsel
         of w~hington,
            the General       .
                          David K. How=-, Attorney


                          Page 25                      GAO/RCEM@221   NETSA/Consumer   Gnmp Chmwexuy
Page 26   GA0jRCED-W221   NETSA/-   r Group (bsrrovemy
Pqge 27   GAo/lK?EDw221   Nmswon-   r Groap Controversy
Related GAO Products

              Changes to the Motor Vehicle Recall Program Could Reduce Potential
              Safety Hazards (GAO/CED-82-99, Aug. 24, 1982).

              Department of Transportation’s Investigation of Rear Brake Lockup
              Problems In 1980 X-Body Cars Should Have Been More Timely (GAO/
              RCED-83-196,Aug. 5, 1983).

              Department of Transportation’s Motor Vehicle Defect-Testing Proce-
              dures Were Reasonable-Public Announcements of Potential Safety
              Defects Could Be Improved (GAO/RCED-&-28, Nov. 9, 1984).

              Auto Safety: Effectiveness of Ford Transmission Settlement Still at
              Issue (GAO/RCED%-62, June 10, 1986).

              Motor Vehicle Safety: Enforcement of Federal Standards Can Be
              Enhanced (GAO/RCED437-2,Dec. 15,1986).

(342810)      Page28                      GAO/RCED9@221NHTSA/Chnsumer Group Contnweray
Ordering   Information

The first five copies of each GAO report are free. Additional copies
are $2 each. Orders should be sent to the following address, accom-
panied by a check or money order made out to the Superintendent
of Documents, when necessary. Orders for 100 or more copies to be
mailed to a single address are discounted 25 percent.

U.S. General Accounting Office
P.O. Box 6015
Gaithersburg, MD 20877

Orders may also be placed by calling   (202) 275-6241.