- .._--.-- -.__ GAO - -. .-.-...--..ll..-...-.___....- ~- ..--____. -_--_I_~ “__.-_I __-__ .I II I \ 1!WO MOTOR VEHICLE SAFETY NHTSA Should Resume Its Support of State Periodic Inspection Programs 5 I I 141954 General Accounting olYke unless specifically approved by the Office of Cmqressional ~t~Fh!~:, United States GAO General Accounting Office Washington, D.C. 20648 Resources, Community, and Economic Development Division B-223736 July 6, 1990 The Honorable John D. Dingell Chairman, Subcommittee on Oversight and Investigations Committee on Energy and Commerce House of Representatives Dear Mr. Chairman: In response to your July 26, 1989 request, this report evaluates the Department of Transportation’s exercise of its state motor vehicle inspection program responsibilities through its National Highway Traffic Safety Administration (NHTSA). We focused our work on determining whether (1) NHTSA'S 1989 report accurately represented the safety benefits of state inspection programs, (2) available evidence indicates that state inspection programs reduce accident rates, and (3) NHTSA appropriately carried out its legislative responsibilities toward inspection programs. Our report recommends that NHTSA resume its support of periodic inspection programs, and we provide suggestions for ways NHTSA can promote these programs. As agreed with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time, we will send copies of this report to the Secretary of Transportation, the Administrator of NHTSA, and other interested parties. This work was performed under the direction of Kenneth M. Mead, Director, Transportation Issues, who can be reached at (202) 2751000. Other major contributors to the report are listed in appendix III. Sincerely yours, J. Dexter Peach Assistant Comptroller General l3xecutive Summq Currently, 21 states and the District of Columbia require annual inspec- Purpose tions of motor vehicles as part of their safety programs. The effective- ness of periodic motor vehicle inspection programs has been a controversial issue for many years. In 1988, the Congress requested the Department of Transportation’s (DOI') National Highway Traffic Safety Administration (NHTSA) to study state inspection programs to determine whether they improve highway safety. NHTSA'S report, issued in 1989, concluded that periodic inspection programs reduce the number of poorly maintained vehicles on the highways, but that available data did not conclusively demonstrate that inspection programs significantly reduced accident rates. NHTSA'S report has been criticized by various industry groups for not accurately representing the safety benefits of inspection programs. Because of this, the Chairman, Subcommitee on Oversight and Investiga- tions, House Committee on Energy and Commerce, asked GAO to evaluate NHTSA'S 1989 report to determine whether: (1) NHTSA'S report accurately represented the safety benefits of state inspection programs, (2) avail- able evidence indicated that state inspection programs reduce accident rates, and (3) NHTSA carried out its legislative responsibilities toward inspection programs. The Highway Safety Act of 1966 required the Secretary of Transporta- Background tion to prescribe uniform standards for mandatory state highway safety programs. A state not complying with the standards could lose its highway safety grant funds and 10 percent of its federal highway con- struction funds. From 1967 to 1972, DOT issued 18 standards, 1 of which required states to conduct periodic motor vehicle inspections. In 1973, NHTSA issued specific inspection standards, such as minimum thickness of brake linings and minimum tire tread depth. By 1975, 31 states and the Distict of Columbia had periodic inspection programs. NHTSA attempted to use funding sanctions against certain states to enforce the adoption of its various highway safety standards. However, the Congress passed the Highway Safety Act of 1976, deleting the Secretary’s authority to withhold highway construction funds and providing that state safety programs could be approved without meeting every program standard. Subsequently, 10 states repealed their periodic inspection programs. For its 1989 report, NHTSA considered most of the available studies of periodic inspection programs and also did five analyses of accident data Page 2 GAO/RCED99-176Periodic Inspection Programe Faecutive Summary available at its headquarters. NHTSA acknowledged that all of these studies, including NHTSA'S analyses, had limitations of scope, age, and/or methodological completeness. NIITSA'S 1989 report accurately concluded that state periodic inspection Results in Brief programs reduce the number of poorly maintained vehicles on the high- ways. This is an important finding because vehicles with worn or defec- tive brakes, tires, lights, or other safety-related components are a hazard to both their owners and other motorists. For example, Virginia officials provided data showing that 25 percent of the vehicles inspected in 1986 had brake defects. NHTSA'S report showed that accidents involving vehicle defects occur less often in states requiring periodic inspections. NHTSA'S conclusion that available data did not conclusively demonstrate that inspection programs significantly reduced accident rates was based primarily on two analyses it did using fatal accident data. Whether intended or not, this conclusion conveyed undue skepticism about the effectiveness of inspection programs and tended to overshadow NHTSA'S finding that inspection programs improve the safety condition of vehi- cles. GAO found that analyses such as NHTSA'S have been hindered by the limitations of available accident data. For example, police acci,dent reports are the source of most data, but they tend to understate the number of accidents in which defective vehicle components contributed to the cause. If driver error or poor road conditions are involved, the investigating officer may not recognize that worn brakes or tires helped cause or aggravate the accident. GAO considered all the studies and analyses in NHTSA'S report and others not discussed by NHTSA. Even taking into account the limitations of indi- vidual studies, their relative consistency in pointing to a safety benefit from periodic inspection justifies a conclusion that these programs reduce accident rates. The magnitude of accident reduction could not be determined because of the data limitations and the methodological problems encountered by those who have studied it. While NHTSA met its obligations under the 1966 legislation to prescribe standards for state inspection programs, the agency did not promote periodic inspection after the Congress restricted its sanction authority in 1976. Although NHTSA was not required after 1976 to support periodic inspection, it could have sponsored research and provided information to help states initiate and improve programs. Recently, NHTSA indicated Page 3 GAO/RCEDQO-175Periodic Inspection Programs Executive Summary a renewed interest in inspection programs and is considering how it can provide such assistance to states. GAO is making a recommendation in this regard. Principal Findings NHTSA’s Report Indicated NHTSA reviewed eight studies which compared the condition of safety- That Inspection Programs related components on vehicles subject to periodic inspection with those in non-inspection jurisdictions. All eight showed that vehicles subject to Have Safety Benefits periodic inspection had fewer defective components than those in areas not requiring inspections. For example, one study in the 1970s found that Pennsylvania, when it still required semiannual inspections, had 46 percent fewer vehicles with defective equipment than California, which used random police inspections. NHTSA'S report also indicated that existing state programs vary in their reliability in detecting and cor- recting vehicle defects. NHTSA'S report also discussed an Indiana study that investigated 420 accidents in depth and found that 12.6 percent were caused or aggra- vated by defective or worn vehicle equipment. Several other studies, including two by NHTSA itself, showed that accidents involving worn or defective equipment occurred less in states requiring periodic inspections. Studies that have compared fatal accident rates between inspection and non-inspection states have found mixed results in trying to estimate the effect of inspection programs. These studies have been hindered by the limitations of available accident data and the difficulty of accounting for the various factors that can affect accident rates. Also, fatal acci- dents are less than 1 percent of all accidents and are not typical of the universe of accidents. NHTSA found that fatal accident rates were about the same in inspection states as in non-inspection states. NHTSA also compared total accidents in four inspection states with those in six non-inspection states, and con- cluded that there was no significant difference in accident rates. In fact, the reported accident rate was 17 percent lower in the inspection states, but NHTSA questioned the comparability of the data and adjusted it to largely eliminate the difference between the two groups of states. For several reasons discussed in this report, GAO did not agree with NHTSA'S Page 4 GAO/RCED-90-1’75 Periodic Inspection Programs Executive Summary adjustment, and considered the 17-percent difference to be further evi- dence that inspection programs reduce accident rates. Taken together, the studies discussed in NHTSA'S report as well as several other studies identified by GAO indicated that inspection programs reduce accident rates. These studies included estimates of accident reduction ranging from less than 1percent to as high as 27 percent. The actual magnitude of the reduction is unknown. GAO agrees with NHTSA that all of the studies had limitations either of scope, age, or method- ological completeness. Thus, while the large majority of studies point to a safety benefit from inspection programs, they do not provide a reliable basis for judging how much effect the programs have on accident rates. NHTSA Intends to Resume NHTSA officials told GAO that the agency intends to resume its support of Its Support of Vehicle periodic vehicle inspection. The contribution these programs make to highway safety provides a basis for NHTSA'S support of such programs. Inspection Programs After reviewing NHTSA'S actions under the 1966 and subsequent legisla- tion, GAO concluded that the agency met its legislative obligations. As mandated, NHTSA issued a standard requiring states to inspect vehicles at least annually, and in 1973 issued specific standards for vehicle inspection. NHTSA officials acknowledged, however, that the agency did not promote inspection programs after the Congress in 1976 deleted NHTSA'S authority to withhold highway construction funds. Under the law, NHTSA was not required to support inspection programs, but it could have continued to promote and help improve programs by sponsoring research and providing information to states on effective ways to operate programs. is recommending that the Secretary of Transportation direct NHTSA Recommendation GAO to support state periodic motor vehicle inspection programs through such actions as (1) sponsoring research, (2) assisting inspection states to share their experiences and adapt to changing automotive technology, and (3) promoting public awareness of the need to properly maintain the safety-critical components of vehicles. As requested, GAO did not obtain official agency comments on a draft of Agency Views this report. However, GAO discussed the report’s contents with NHTSA Y officials, and they generally agreed with GAO'S findings. GAO incorpo- rated their clarifying comments as appropriate. These officials said that NHTSA welcomes suggestions for ways to promote and improve inspec- tion programs. Page 5 GAO/RCED-90-176Periodic Inspection Programs Contents Executive Summary 2 Chapter 1 8 Background NHTSA’s Involvement in Periodic Motor Vehicle Inspection Programs 8 Objectives, Scope, and Methodology 11 Chapter 2 NHTSA’s 1989 Report NHTSA’s Report Indicated That Inspection Programs Have Safe and Other Data Show Additional Stu UltjJIJ”LzUAJLU33~U vyLYll IUAfi1.7” That Periodic Indicated That Periodic Inspection Programs Reduce Accidents Inspection Programs ReduceAccident Rates Conclusion Chapter 3 21 NHTSA Intends to NHTSA Met Its Original Obligations but Has Not Actively Supported Vehicle Inspections Since 1976 21 ResumeIts Support of Inspection Programs Could Be More Effective 22 Vehicle Inspection Conclusions 23 Recommendation 24 Programs Agency Views 24 Appendixes Appendix I: States Requiring Periodic Vehicle Inspections 26 Appendix II: Principal Studies Discussed in NHTSA’s 27 Report Appendix III: Major Contributors to This Report 33 Figures Figure 2.1: Defective Equipment Reported on Vehicles 16 Involved in Accidents, 1984-86 Figure 2.2: Defective Equipment Reported on Vehicles 17 Involved in Fatal Accidents, 1985-87 Page 0 GAO/RCEDSO-175Periodic Inspection Programs Contents Abbreviations * DOT Department of Transportation GAO General Accounting Office NHTSA National Highway Traffic Safety Administration Page7 GAO/RCED-90-17sPeriodic Inspection Program Chapter 1 Background In order to reduce the toll of highway accidents, the Congress created the National Highway Traffic Safety Administration (NHTSA) within the Department of Transportation’ (DCT) to enforce federal motor vehicle safety standards, sponsor safety research and development, and sup- port state highway safety programs. In 1966, traffic accidents killed 50,894 persons in the United States. In 1988, the traffic death toll had decreased to 47,093. The number of vehicle miles traveled, estimated at 926 billion in 1966, had increased to over 2 trillion by 1988. Measured in terms of vehicle miles traveled, the fatality rate in 1988 was 58 percent lower than in 1966. Many factors have contributed to this decline, including improved vehicles, improved highways, and programs to raise the safety consciousness of drivers. NHTSA has estimated that one major initiative, safety belts, saved 4,500 lives in 1988. Among the many programs intended to reduce the number and severity of accidents are the requirements in 21 states and the Dis- trict of Columbia for periodic inspection of motor vehicle brakes, tires, steering, and other safety-related components. The Highway Safety Act of 1966 (P.L. 89-564) and the National Traffic NHTSA’s Involvement and Motor Vehicle Safety Act of 1966 (P.L. 89-563) both established in Periodic Motor responsibilities for the Secretary of Transportation.” The National Vehicle Inspection Traffic Safety Bureau (later NHTSA) was created to administer these acts on behalf of the Secretary. Programs The first act required the Secretary to prescribe uniform standards for mandatory state highway safety programs. The Secretary was required to approve each state’s program and withhold highway safety grant funds and 10 percent of highway construction funds from states not complying with the program standards. The act specifically mentioned vehicle inspection among the potential subjects for state program stan- dards. The second act required the Secretary to establish safety stan- dards for new vehicles, and standards for the inspection of vehicles in use. ‘The agency was originally called the National Highway Safety Bureau and was under the Depart- ment of Commerce. It moved to the Department of Transportation (when that agency began opera- tions on Apr. 1, 1967), where it was under the Federal Highway Administration until it was established as a separate agency within the Department in 1970. ‘The legislation actually referred to the Secretary of Commerce, but these responsibilities were trans- ferred to the newly created Secretary of Transportation in 1967. Page 8 GAO/RCED-90-176Periodic Inspection Programs Chapter 1 Background NHTSA’s Issuance and In carrying out the Highway Safety Act of 1966, DOT issued 18 standards Enforcement of State for state highway safety programs from 1967 through 1972. The first standard required each state to have a program for periodically Safety Program Standards inspecting all registered vehicles or an experimental, pilot, or demon- stration program approved by the Secretary. NIITSA officials said that in 1973, under pressure from a court order, the agency established the spe- cific standards for inspecting vehicles required by the National Traffic and Motor Vehicle Safety Act of 1966. These standards applied to brakes, tires, wheels, and steering and suspension components, and included such items as minimum brake lining thickness and tire tread depth. In response to the DUT requirements, 11 states adopted periodic inspec- tion laws between 1967 and 1972, bringing the total to 31 plus the Dis- trict of Columbia. In 1975, NIITSA prepared to use the authorized funding sanctions to enforce state compliance with safety program standards, particularly those involving blood alcohol content for drunk driving, motorcycle helmet use, and periodic vehicle inspection. The sanction process was suspended when the Congress passed the Highway Safety Act of 1976, deleting the Secretary’s authority to enforce the safety pro- gram standards by withholding highway construction funds. The act also specified that the Secretary should not “require compliance with every uniform standard, or with every element of every standard, in every state.” NHTSA’s Reaction to the While the 1976 act did limit NIII'SA'S authority to require state program Highway Safety Act of activities, it did not eliminate the Secretary’s authority to approve state highway safety programs and withhold highway safety program funds 1976 from states not having approved programs. And it did not repeal the statement in the National Traffic and Motor Vehicle Safety Act of 1966 that “it is the policy of Congress to encourage and strengthen the enforcement of State inspection of used motor vehicles.” Nonetheless, em adopted a policy that all highway safety program standards would be optional and states could determine their own priorities. Dm reported to the Congress in 1977 that it could not statistically demonstrate the effectiveness of any of its 18 program standards, including periodic vehicle inspections. DO?' stated: “This is not to say that the highway safety program and the standards do not improve safety. Rather, this is an admission of our inability to produce statistically verifiable data which convincingly demonstrate what our common sense tells us.” Since 1977, NHTSA has not withheld highway safety funds from any state for Page 9 GAO/RCED-90-175Periodic Inspection Programs ,- Chapter 1 , Background noncompliance with a safety program standard.:’ Regarding periodic inspection programs, NHTSA officials agreed that the agency stopped pro- moting them after 1976. From 1976 through 1982, 10 states repealed their periodic inspection laws, including 7 of the states that adopted their laws in response to the Highway Safety Act of 1966. Currently, 21 states and the District of Columbia require annual motor vehicle safety inspections. All but three of these programs predate the 1966 act, with some going back to the 1930s and one to 1929. These states are primarily on the eastern sea- board or the Gulf Coast. (See app. I.) New Jersey, Delaware, and the District operate facilities which perform the inspections. The remaining 19 states license private garages to do inspections, monitored by the state police or another state agency. For a fee, the mechanics inspect the condition of brakes, tires, steering components, lights, and other safety- related equipment on the vehicles. Vehicles not meeting state standards must be repaired and reinspected. Federal funds are not used to support state inspection operations. NHTSA Required to Report We reported in 1977 that many states were reluctant to adopt periodic on Inspection Program inspection programs because they were not convinced of the benefits of such programs.4 Therefore, we recommended that NHTSA undertake pri- Effectiveness ority research to demonstrate program effectiveness. NIITSA did not, however, undertake any new research on the effects of inspection programs. In 1988, the Congress requested that NIITSA study existing state inspec- tion programs and determine whether they reduce the number of poorly maintained vehicles on the highway and help reduce accident rates. NIWSA reviewed prior studies, surveyed the current status of vehicle inspection programs, and performed analyses using data available at NIIWA headquarters. NHTSA also held public hearings and solicited com- ments from state officials and other interested persons. NIITSA reported in 1989” that periodic inspection programs reduce the number of poorly “In 1987, the Congress changed the highway safety program standards to guidelines. “Effectiveness of Vehicle Safety Inspections Neither Proven Nor Ilnproven (CED-78-18, Dec. 20, 1977), pp. 20-22. “Study the Effectiveness of State Motor Vehicle Inspection Programs. NIITSA (Washington, D.C.: A?g. 1989). Page 10 GAO/RCED-90-176Periodic Inspection Programs Chapter 1 Background maintained vehicles on the highways, but reported that it could not con- clusively demonstrate that the programs significantly reduce accident rates. Various organizations have criticized NHTSA for alleged shortcomings in Objectives, Scope,and its 1989 report and for its lack of support for periodic vehicle inspec- Methodology tion Consequently, the Chairman, Subcommittee on Oversight and Investigations, House Committee on Energy and Commerce, asked us to assess whether NHTSA has adequately carried out its vehicle inspection responsibilities and what safety benefits can be attributed to such programs. In agreement with the Chairman’s office, we focused our work on deter- mining whether: (1) NHTSA'S 1989 report accurately represented the safety benefits of state inspection programs, (2) available evidence indi- cated that state inspection programs reduce accident rates, and (3) NHTSA appropriately carried out its legislative responsibilities toward inspection programs. To carry out the first objective, we reviewed NHTSA'S 1989 report and discussed it with the NHTSA personnel who prepared it. We involved methodological experts on our staff in assessing NHTSA'S analyses of available data. We reviewed some of the prior studies cited by NHTSA, most of which were done before 1980, and in other cases, accepted NIITSA'S summarization of them. We considered whether, given the infor- mation contained in NHTSA'S report, we would have arrived at similar conclusions. For the second objective, we reviewed comments submitted to NHTSA by states and other interested parties to determine if there was other infor- mation or studies that NHTSA did not consider in its 1989 report. We also reviewed an available literature search and asked officials from NHTSA, states, and interested organizations if they were aware of other relevant studies or analyses. From this effort, we identified four studies not dis- cussed by NHTSA in arriving at its conclusions. We used this additional information along with the studies discussed by NHTSA to assess the rela- tionship between periodic vehicle inspection programs and accident rates. For the third objective, we reviewed legislation, regulations, and other documents relating to NHTSA'S safety programs and discussed their implemention with officials of NHTSA; the American Association of Motor Page 11 GAO/RCED-90-176Periodic Inspection Programs ‘4 Chapter 1 Background Vehicle Administrators; and the Coalition for Safer, Cleaner Vehicles. Specifically, we considered whether NHTSA met its minimum obligations under the 1966 legislation and whether it adopted an appropriate role in response to the 1976 legislative changes. We also considered whether NHTSA should encourage periodic inspection programs and how the programs could be improved, As requested by the Chairman, we interviewed officials of interested organizations and visited states with inspection programs (New Jersey, Pennsylvania, and Virginia) as well as states without them (Florida, Indiana, and Ohio). In Indiana and Florida, we discussed the reasons why previous inspection laws were repealed. We also attended a conference on vehicle inspection at which officials from a number of other states participated. We conducted our audit work between September 1989 and February 1990 in accordance with generally accepted govern-tent auditing stan- dards, We discussed the report’s contents with NHTSA officials and incor- porated their clarifying comments as appropriate. However, as requested, we did not obtain offical NHTSA comments on a draft of this report. Page 12 QAO/IUXD4W176 Periodic Impection Progmmii Chapter 2 NHTSA’s 1989 Report and Other Data Show That Periodic Inspection Programs Reduce Accident Rates NHTSA'S 1989 report accurately concluded that state periodic inspection programs reduce the number of poorly maintained vehicles on the high- ways. This is an important finding because vehicles with worn or defec- tive brakes, tires, lights, or other safety-related components are a hazard to both their owners and other motorists. NHTSA'S report also showed that accidents involving vehicle defects occur less often in states requiring periodic inspections. NHTSA'S conclusion that available data did not conclusively demonstrate that inspection programs significantly reduced accident rates was based primarily on two analyses it did using fatal accident data. Whether intended or not, this conclusion conveyed undue skepticism about the effectiveness of inspection programs and tended to overshadow NHTSA'S finding that inspection programs improve the safety condition of vehi- cles. Analyses such as NHTSA'S have been hindered by the limitations of available accident data. We considered all the studies and analyses in NHTSA'S report and others not discussed by NHTSA. Even taking into account the limitations of individual studies, their relative consistency in pointing to a safety benefit from periodic inspection justifies a conclu- sion that these programs reduce accident rates. The magnitude of acci- dent reduction could not be determined because of the data limitations and the methodological problems encountered by those who have studied it. 1989 report demonstrated that periodic inspection programs NHTSA’s Report NHTSA'S improve the condition of the safety-related components of vehicles sub- Indicated That ject to inspection. The report also contained consistent evidence that Inspection Programs fewer accidents involving defective or worn vehicle equipment occur in states requiring inspections. While this would seem to be persuasive evi- Have Safety Benefits dence that the programs reduce overall accident rates, comparisons of fatal accident rates do not always show such an effect. Unfortunately, most comparisons have been confined to fatal accident data because of their availability and reliability. However, fatal accidents represent less than 1 percent of all accidents and may not be the type of accidents most affected by defective vehicle equipment. Also, it is difficult in any comparison of accident rates to control the various other factors that can influence them. In assessing NHTSA'S report, we considered all the evidence and noted that most of the studies indicated a safety benefit from inspection pro- grams. We believe NHTSA may have focused too much on its own compar- isons of state accident rates, considering the limitations of such Page 18 GAO/RCEDQO-176Periodic Inspection Programs , Chapter 2 ‘ NIIWA’a 1989Report and Other Data Show That Periodic Inspection Program Reduce Accident Rates comparisons. While NHTSAmay not have intended to draw negative con- clusions about the effectiveness of inspection programs, it did seem to place emphasis on the analyses that did not support the programs. This left the impression that NHTSAwas skeptical of the benefit of inspection programs. NHTSA’s Report Showed To determine whether periodic inspections improve vehicle condition, That Inspection Programs NHTSAreviewed eight studies which compared the condition of safety- related components on vehicles subject to periodic inspection with those Improve Vehicle Condition in non-inspection jurisdictions. All eight studies showed that vehicles not subject to periodic inspection had more defective components than those in areas requiring inspections. For example, Tennessee found in the 1970s that vehicles in Memphis and Chattanooga, which required inspections, had fewer safety defects than those in Knoxville, which did not. As another example, two NHTSA-sponsored studies in the early 1970s compared results from diagnostic centers in 10 states. The three states with the lowest defect rates were states that required periodic inspections. Pennsylvania, which at that time required semiannual inspections, had 45 percent fewer vehicles with defective equipment than California, which used random police inspections. These studies indicated that semiannual inspections were more effective than annual inspections, which in turn were more effective than random inspections. NHTSAconcluded from its review of these eight studies that periodic inspection programs limit the number of poorly maintained vehicles on the highways. NHTSA’s Report If periodic inspection programs are effective, they should reduce the Demonstrated That number of accidents caused or aggravated by worn or defective vehicle equipment. To determine if this was so, NHTSAreviewed a study done in Inspection Programs Indiana on the causes of accidents, and three studies comparing the rate Reduce Accidents Caused of vehicle defects cited in accident reports. NHTSAalso performed two by Vehicle Defects data analyses comparing accident reports from inspection and non- inspection states. NHTSAsponsored a study in Monroe County, Indiana, in the early 1970s to determine the causes of traffic accidents. As part of that study, large, multidisciplinary teams conducted in-depth follow-up investigations of 420 accidents. The study concluded that defective vehicle components were the sole cause or a contributing cause of 12.6 percent of the acci- dents (4.5 percent definite, 8.1 percent probable). Failure or Page 14 GAO/RCED-90-175Periodic Inspection Programs chapter 2 NEITSA’s1989Report and Other Data Show That Periodic Inspection Progranu Reduce Accident Rates underperformance by brakes and tires were the most commonly noted deficiencies that contributed to accidents. The three other studies of vehicle-defect involvement in accidents all indicated that periodic inspection programs reduce accidents caused by vehicle defects. For example, a 1975 NHTSA study cited data showing that reported vehicle-defect accidents declined from 12 to 4 percent of all accidents in Texas and from 6.1 to 2.6 percent in rural Nebraska in the years following implementation of inspection programs. Another of these studies found a lower rate of vehicle-defect accidents on the Penn- sylvania Turnpike than on the Indiana and Ohio Turnpikes during a period when Pennsylvania was the only one of these states requiring inspections. NHTSA did two analyses of computerized accident reports to determine how often vehicle equipment failures were noted on vehicles involved in accidents. One involved all accidents in four states and the other involved fatal accidents in all states. In the four-state study, NHTSA found that vehicle equipment failures were reported on about 1 percent of the vehicles involved in accidents in the inspection states (Penn- sylvania and Texas) and about 2 percent in the non-inspection states (Maryland and Washington). As shown in figure 2.1, the difference between the states was greater for older vehicles than for newer ones, which indicates that inspection programs have more effect on older vehicles. This confirmed NHTSA'S hypothesis that the effect of inspection programs would be most evident for older vehicles. Page 15 GAO/RCED-SO-175 Periodic Inspection Programs Chapter 2 NHTSA’e 1989Report and Other Data Show That Periadic Inspection ProgramsReduce Accident Rates Figure 2.1: Defective Equipment Reported on Vehicles Involved in Accidents, 1984-86 Percent Accident-Involved Vehicles /I 0 1 2 3 4 S 6 7 8 9 10 Vehicle Age (Yrs.) m Inspection m Non-Inspection Source: NHTSA NHTSA'S second analysis, using nationwide fatal accident data from 1985 through 1987, also showed that defects were noted on vehicles involved in fatal accidents less often in inspection states than in non-inspection states. Once again, as shown in figure 2.2, NHTSA found that the differ- ence between inspection and non-inspection states widened for older vehicles. Thus, NHTSA'S analysis of fatal accidents in all states confirmed its four-state analysis of total accident data. Page 16 GAO/RCED-90-176Periodic Inspection Programs Chapter 2 NIiT8A’s 1989Report and Other Data Show That Periodic Inspection ProgramsReduce Accident Rates Figure 2.2: Defective Equipment Reported on Vehicles Involved in Fatal Accidents, 1985-87 Peroent Fatal Accident-Involved Vehicles 0 1 2 3 4 5 6 7 8 9 10 Vehicle Age (Yrs.) Source: NHTSA NHT~A considered the differences between the states too small to be of any practical significance. However, police accident reports may under- state the percentage of accidents caused by defective vehicle equipment. The Indiana study found that police officers did not identify all of the vehicle defects that contributed to the accidents studied. The Indiana researchers noted that police officers must try to determine who was legally responsible for an accident, and may not look further for causal relationships. As New York officials pointed out in comments to NHTSA, police officers are not mechanics, and their first concern at an accident scene must be the care of the injured and clearing the site of hazards. Accident Rate In addition to studying the role of vehicle defects in accidents, NHTSA Comparisons Have Been also reviewed studies which compared accident rates (mostly fatal acci- dent rates) between inspection and non-inspection states. Fatal accident Inconclusive Because of data are more readily available and more reliable, but fatal accidents Data Limitations represent less than 1 percent of all accidents. On the other hand, total accident data are difficult to interpret because of different reporting * practices among the states. Among eight studies NHTSA reviewed, six compared fatal accident rates between states with and without inspection programs. Three found Page 17 GAO/RCED-90-175Periodic InspectionPrograms Chapter 2 NHTSA’s 1999Report and Other Data Show That Periodic hpection ProgramaReduce Accident Rates lower accident rates in states requiring periodic inspections, one found higher rates, and two found inconclusive results. The remaining two studies were done within single states. One of these, done in Huntsville, Alabama, compared a sample of inspected vehicles with other vehicles in Huntsville. The authors estimated that the inspected vehicles were involved in 9-21 percent fewer accidents. The other, a historical study done in New Jersey, compared total accident rates for a number of years before and after the state’s adoption of an inspection program. Control- ling for a number of other factors, the study estimated an accident reduction of 23 percent from the state’s inspection program. While each of the previous studies had limitations, taken together, they suggested that periodic inspections reduce accident rates. However, NHTSA did three data analyses comparing accident rates among the states which produced apparently conflicting evidence. Two of these used fatal accident data from 49 states and found fatal accident rates to be about the same in inspection states as in non-inspection states. In the first of these two analyses, a comparison by age of vehicle also showed little difference, although vehicles 9-12 years old were more likely to be involved in fatal accidents in states not requiring periodic inspection. In the second analysis, NHTSA looked at fatal accidents involving 1976 vehi- cles over an 1 l-year period, and did not find a trend favoring inspection states as the vehicles got older. NHTSA'S third analysis of accident rates used state accident data files to compare total accidents in four inspection states with those in six non- inspection states. The inspection states showed a 17-percent lower acci- dent rate than the non-inspection states. However, NHTSA doubted the comparability of the data because it showed that relatively new vehicles (O-21 months old) also had higher accident rates in the non-inspection states. NHTSA assumed that newer vehicles have few defects and thus should not have higher accident rates in non-inspection states. NHTSA adjusted the data from this analysis and largely eliminated the difference between the two groups. NHTSA justified this in its report by stating that the inspection states were not reporting as many accidents because they had higher damage thresholds for reporting accidents than the non-inspection states. We found the reverse to be true: the inspection states in NHTSA'S sample had lower thresholds for reporting accidents than the non-inspection states. We also question NHTSA'S reason for adjusting the data. According to the U.S. Department of Commerce, more than a third of new vehicles are Page 18 GAO/RCED-90-176Periodic Inspection Programs chapter2 NIi’JX3A’s1989Report and Other Data Show That Periodic Inspection ProgramsReduce Accident Rates used for business purposes. Such vehicles may be driven 25,000 miles in their first year, according to the American Automotive Leasing Associa- tion They could develop problems with brakes, tires, or steering at an early age, problems that an inspection program would identify. More- over, older vehicles in non-inspection states may involve new vehicles in more accidents than would occur if the older vehicles were subject to periodic inspection. NHTSA'S comparisons of accident rates are also limited by the fact that, except for vehicle age, NHTSA did not control for other factors that can influence state accident rates. We found, for example, that traffic den- sity was higher in the inspection states. Motorists in these states trav- eled 20 percent more per mile of roadway in 1988 than motorists in non- inspection states. NHTSA also reviewed several studies that had attempted to estimate costs and benefits of periodic inspection programs. The majority of these studies, including NHTSA'S own 1975 study, indicated that the programs were cost-effective. NHTSA questioned their assumptions, however, and ’ concluded that “none of the reviewed studies provide credible evidence that current programs are cost-effective.” NHTSA officials told us they did not do a new cost-benefit analysis for their report. In addition to the information discussed in NHTSA'S report, we identified Additional Studies Not four other studies, all of which indicated an accident-reduction benefit Discussedby NHTSA from periodic inspection programs: Also Indicated That Florida officials provided us two studies showing that the percentage of Periodic Inspection l accidents caused by vehicle defects decreased when periodic inspections Programs Reduce were begun, and increased when the inspection law was repealed. Accidents . In comments submitted to NHTSA for its report, the Illinois Department of Transportation discussed the effect of terminating the semiannual inspection of pickup trucks and vans in 1984. It reported a sharp increase in the rate of accidents and injuries involving pickup trucks and a small increase for vans in the 3 years following repeal. . A study done at Rutgers University used 12 different variables in a regression analysis of fatal accident data from 1979.’ The author con- cluded that inspection programs can reduce fatal accident rates by a sig- nificant amount. ‘Peter D. Loeb, “The Determinants of Automobile Fatalities With Special Consideration to Policy Variables,” Journal of Transport Economics and Policy, Sept. 1987, pp. 279-287. Page 19 GAO/RCEDdO-175Periodic Inspection Programs . Chapter 2 NIi’IWA’s 1989Report and Other Data Show That Periodic Inspection ProgramsReduce Accident Rates A large majority of the studies NHTSA reviewed, and four additional ones Conclusion that we identified, indicated that inspection programs improve highway safety. We believe that when all the studies and analyses are considered together, even taking into account their individual limitations, their rela- tive consistency justifies a conclusion that periodic inspection programs reduce accident rates. None of the studies, however, produced a reliable estimate of the magnitude of accident reduction that can be expected from an inspection program. Various studies have placed it as low as less than 1 percent to as high as 27 percent. While it would be reason- able, on the basis of current evidence, for NHTSA to encourage the adop- tion of periodic inspection programs, states would have a better basis for considering such programs if NHTSA sponsored a carefully controlled research project to estimate their accident-reduction potential. Ideally, such research would follow the accident experience of a randomly selected group of inspected vehicles and a control group of vehicles not subject to inspection. We discussed our interpretation of the data with NHTSA officials respon- sible for the 1989 report, who said that their views are now close to ours. According to these officials, NHWA agrees that periodic inspection programs contribute to highway safety and should be supported. Page 20 GAO/RCED-90-175Periodic Inspection Programs Chaoter 3 NHTSA Idends to ResumeIts Support of Vehicle Inspection Programs NHTSA met its obligations under the 1966 legislation to prescribe uniform standards for state inspection programs. However, when the Congress restricted its sanction authority in 1976, the agency chose not to con- tinue promoting vehicle inspection programs. Some states have recently considered initiating or reinstating inspection programs. It also appears that inspection programs could be improved to enhance their contribu- tion to traffic safety. NHTSA could promote and help improve inspection programs by sponsoring research and providing information to states on effective ways to operate inspection programs. After reviewing the history of NHTSA actions under the 1966 legislation, NHTSA Met Its we conclude that NHTSA met its legislative obligations. As required by Original Obligations the Highway Safety Act of 1966, NHTSA issued standards for state but Has Not Actively highway safety programs, including a requirement that states inspect motor vehicles at least annually. In 1973, NHTSA complied with a require- Supported Vehicle ment in the National Traffic and Motor Vehicle Safety Act of 1966 by Inspections Since 1976 issuing specific standards for vehicle inspection. NHTSA was prepared to enforce its safety program standards through federal funding sanctions until the Congress deleted its authority to withhold highway construc- tion funds in the Highway Safety Act 1976 and provided that state safety programs could be approved without meeting every program standard. NHTSA officials acknowledged that after 1976, the agency did not con- tinue promoting vehicle inspection programs. NHTSA sponsored only one more piece of original research: a study in Idaho that showed that the condition of vehicles’ brakes, steering, and suspension deteriorated after inspections were discontinued in the state in 1977. In 1978, NHTSA abol- ished the office responsible for vehicle inspection. NHTSA'S withdrawal from the area of periodic inspection may have reflected its earlier reservations about inspection programs. As early as 1972, NHTSA testified before the Senate Commerce Committee that it was skeptical about the effectiveness of vehicle inspection programs. While acknowledging that four studies from the 1960s had shown positive cor- relations between inspection programs and lower traffic fatality rates, NHTSA said that more recent data did not support such a relationship. It said that an analysis with more variables was needed to determine whether inspection programs reduce fatalities. NHTSA'S subsequent actions, issuing vehicle-in-use inspection criteria and beginning the sanc- tion process to enforce safety program standards, were taken under pressure from court orders. Page 21 GAO/RCED-90-176Periodic Inspection Programs chapter 3 NIfBA Intends to ResumeIts Support of Vehicle Inspection Programs NHTSA officials believe their withdrawal from the area was supported by comments they received from states in 1981. The Congress had directed NHTSA to determine by rulemaking which of its 18 state safety programs should be eligible for continued federal funding. When vehicle inspec- tion was not one of the programs the states identified, NHTSA assigned it to the category of non-priority programs. A NHTSA official commented recently that the agency may have misinterpreted this response, since inspection programs are generally self-supporting and do not require federal funds. Of the 11 states that initiated periodic inspection programs from 1967 through 1969,7 repealed their programs after NHTSA'S authority to with- hold highway construction funds was deleted by the Highway Safety Act of 1976. No states have initiated mandatory safety inspection pro- grams since 1969. Florida and Colorado officials told us that safety testing may be reinstated in the near future, while Connecticut has initi- ated demonstration safety inspection facilities. Michigan’s written com- ments for NHTSA'S 1989 report indicated a possible interest in starting a periodic safety inspection program. In Missouri, on the other hand, the state’s inspection program has been questioned in the state legislature. Several state officials told us that NHTSA could be helpful by sponsoring research to test new technologies and determine the most effective approaches to vehicle inspection. They also said there is a need to dis- seminate information on state program experiences. For example, Penn- sylvania and New Jersey officials said that information on the hazards of modified (raised) vehicles would be useful to many states. Other states might also profit from New Jersey’s approach of using its inspec- tion procedure to check drivers’ licenses, registrations, license plates, and mandatory insurance coverage. As NHTSA pointed out in its report, existing state inspection programs Inspection Programs vary in their effectiveness. For example, Pennsylvania allows its Could Be More licensed private garages to charge one-half hour of shop labor for an Effective inspection and requires removal of two wheels and a road test for brake inspection. Pennsylvania rejects about 17 percent of its vehicles for brake problems. Virginia requires its licensed private garages to pull one wheel but some pull two anyway. Virginia’s stations found a 25-percent deficiency rate for brakes. By improving its monitoring of stations, Vir- *I ginia raised its overall vehicle rejection rate from 22 percent in 1982 to 34 percent in 1986, and recorded a 48-percent decline in accidents that it attributed to vehicle defects over the same period. New Jersey charges Page 22 GAO/RCED90-175Periodic Inspection Programs Chapter 3 NHTSA Intends to ResumeIts Support of VeNcle Inspection Programs $2.50 and takes about 5 minutes to move a vehicle through its state- operated safety lanes, including a brake test on an old style of platform tester. New Jersey does not pull wheels and rejects about 12 percent of vehicles for brake problems (including the parking brake). Some officials believe that more effective and efficient testing can be achieved with the application of new technology. For example, Florida has contracted with private companies to build and operate facilities for its required emissions testing. Each facility must include one lane which offers free voluntary safety tests using modern equipment for detecting the wear of brakes, steering linkage, and alignment. Florida officials hope to build public support for reinstatement of periodic safety inspec- tions. Connecticut, which does not require safety inspections, has a pilot project to demonstrate the operation of safety inspection facilities using modern testing equipment. For states which rely on private garages to perform inspections, vig- orous monitoring and adequate fees are important for an effective pro- gram. Studies have shown that lax inspections are more often a problem than unneeded repairs. Indiana and Colorado officials told us that public support for their programs was undermined by reports of perfunctory inspections and garage owners selling inspection stickers without per- forming the inspections. Indiana had 19 state police officers assigned to monitor 4,500 stations. In the final years of the program, they spent much of their time investigating allegations of stickers being sold without inspections. As previously mentioned, some states have set very low inspection fees. Political considerations may make it difficult to raise fees, but such states run a risk of losing the credibility of their inspections. If inspec- tion fees are too low, garage owners may be tempted to do lax inspec- tions or to reject vehicles unnecessarily in hopes of getting repair orders. Ideally, the fee should be set high enough to cover the cost of a legiti- mate inspection that would enhance highway safety and also give the individual motorist timely advice about the condition of safety-related equipment on his or her vehicle. The experience of 11 states which initiated programs in the late 1960s Conclusions under threat of federal funding sanctions demonstrated the importance Y of building public support for inspections. Seven of these states repealed their programs when the threat was lifted in 1976. State officials told us that NHTSA'S lack of a positive stance on the value of periodic inspection Page 23 GAO/RCED-90-176Periodic Inspection Programs . Chapter 3 NHTSA Intends to ResumeIts Support of Vehicle Inspection Progranw programs has made it more difficult to develop support for inspection programs. NHTSA met its original legislative obligations, but it could do more to sup- port inspection programs by taking a positive position, supporting research, and providing information services to the states. Through such efforts, NHTSA could help improve the effectiveness of existing programs and encourage other states to initiate or reinstate periodic inspections. We recommend that the Secretary of Transportation direct NHTSA to sup- Recommendation port periodic motor vehicle inspection through such actions as (1) spon- soring research that would assist states considering the initiation or reinstatement of inspection programs, (2) assisting inspection states so that they share their experiences and adapt to changing automotive technology, and (3) promoting public awareness of the need to properly maintain the safety-critical components of vehicles. As requested, we did not obtain official agency comments on a draft of Agency Views this report. However, we discussed the report’s contents with NHTSA offi- cials, and they generally agreed with our findings. We incorporated their clarifying comments as appropriate. They indicated that NHTSA has reconsidered its position on the value of periodic motor vehicle inspec- tions and that NIITSA now welcomes suggestions for activities it can undertake in support of inspection programs. Page 24 GAO/RCED-90-175Periodic Inspection Programs Page 26 GAO/RCED-90-175Periodic Inspection Programs Appendix I States Requiring Periodic Vehicle Inspections ’ l States Currently Requiring Annual States That;rtp;izt Inspection Safety Inspections a State Started State Started Ended Pennsvlvania 1929 Colorado 1937 1981 Maine 1930 New Mexico 1953 1977 Massachusetts 1930 Georgia 1965 1982 New Hampshire 1931 Wyoming 1967 1977 Virainia 1932 Florida 1968 1981 Delaware 1933 Idaho 1968 1976 Utah 1936 Kentucky 1968 1978 Vermont --.- 1936 South Dakota 1968 1979 New Jersev 1938 Indiana 1969 1980 District of Columbia 1939 Nebraska 1969 1982 Texas _______ 1951 West Virainia 1955 New York ___--_____ 1957 Rhode -- Island 1959 --. Louisana 1961 Mississippi 1961 -_ -Hawaii 1961 North Carolina 1966 South Carolina 1968 Arkansas 1969 Missouri _-.~ 1969 Oklahoma 1969 Page 26 GAO/RCED-90-175Periodic Inspection Programs &pen& II Phncipd Studies Discussedin NHTSA’s Report The following eight studies compared the condition of safety-related components on vehicles subject to periodic motor vehicle inspections with those in non-inspection jurisdictions: 1. McCutcheon, Robert W. The Influence of Periodic Motor Vehicle Inspection on Mechanical Condition. Ann Arbor: Highway Safety Research Institute, July 1968. This study compared vehicle condition in the metropolitan areas of Washington, D.C.; Cincinnati, Ohio; and Memphis, Tennessee, which required inspections, to non-inspection Ann Arbor, Michigan. It showed that inspection leads to better maintained vehicles and that the condi- tion of the vehicles improves with the frequency of inspection. 2. Fisher, Franklin G., Jr., Randolph Eidemiller, and Peter Biche. Vehicle-in-Use Safety Standards Study: Summary and Final Report (also 12 other ~01s.). NIITSA Reports DOT HS-800 559, 560. Newport Beach, Calif.: Ultrasystems, Aug. 1971. By comparing various inspection and non-inspection areas, this study found that inspection states had fewer vehicle component defects. The areas included were six diagnostic centers in California, one in Penn- sylvania, one in NewJersey, and two city inspection stations in Wash- ington, D.C. 3. Fisher, Franklin G., Jr., Peter Biche, and Randolph Eidemiller. Status of Vehicle-in-Use Study: Summary Final Report and Final Contract Report. NIITSA Reports DOT HS-800 894,898. Newport Beach, Calif.: IJltrasystems, July 1973. The non-inspection states in this study, Alabama, Connecticut, Illinois, Maryland, and Washington, had higher defective vehicle component rates than the inspection states in the previous study. However, Mis- souri had higher defective component rates despite its inspection program. 4. Hatch, William, James De Armon, and Cheryl Louie. State Inspection Program Evaluation and Data Analysis: Vol. I, Summary Report; Vol. II, Technical Report. NHTSA Reports DOT HS-802 149, 150. Silver Spring, Md.: Automated Sciences Group, Inc., Dec. 1976. By comparing selected inspection and non-inspection states using NIITSA'S mobile inspection van, this study found that 16 components Page 27 GAO/RCED-90-176Periodic Inspection Programs Appendix II Principal Studies Discussedln NH’ISA’s Report were defective less often in inspection states and 6 were defective less often in non-inspection states. The mobile inspection van sampled five cities in each of the following six states: California, Illinois, Maryland, Missouri, Pennsylvania, and Texas. 5. Innes, Joseph J. and Leslie E. Eder. Motor Vehicle Diagnostic Inspec- tion Demonstration Program-Summary Report. NHTSA Report nor HS-802 760. Washington: Department of Transportation, Oct. 1977. By comparing the inspection and non-inspection states of Alabama, Ari- zona, the District of Columbia, and Puerto Rico, this study found vehicle condition in inspection states to be better than in non-inspection states. 6. Eder, Leslie E., Noel Bleich, and Mario Damiata. The NIITSA Trial Sub- stitute Motor Vehicle Inspection Programs. NIITSA Technical Report DCR I-IS-803 535. Washington: NHTSA, July 1978. The authors found that Cincinnati, which required inspections, had fewer vehicles with defects than did the rest of Ohio. Memphis and Chattanooga, with inspection programs, had fewer defective vehicles than non-inspection Knoxville. 7. Final Report on Motor Vehicle Inspection Experiment. California IIighway Patrol. Sacramento: Dec. 1974. This study is a comparison of vehicle defects in areas with different levels of random inspection in the state of California. It found that vehi- cles in counties with more frequent random inspections tended to be in better condition. 8. Eder, Leslie F:. Impact of Discontinuing Idaho’s Periodic Motor Vehicle Inspection Program (A Before and After Outage Rate Study). NIWSA Technical Report DOT IIS- 535. Washington: NIITSA, July 1978. By comparing vehicle condition before and after Idaho repealed its inspection law in 1976, this study found that repeal had a somewhat negative effect on vehicle condition. The following is the Indiana study of accident causation and the role of vehicle component failures in accidents: Page28 GAO/RCED-90-175Periodic Inspection Programs Appendix II Principal Studies Mscussed in NIiTSA’s Report 1. Treat, John R. and Ricky L. Stansifer. “Vehicle Problems as Accident Causes-An Overview of Available Information,” SAE Paper 770117. Warrendale, Pa.: Society of Automotive Engineers, Mar. 1977. In this study, in-depth follow-up accident investigations were conducted by multidisciplinary teams in Monroe County, Indiana. It concluded that vehicle defects were definitely causal or severity-increasing in 4.5 per- cent, probably causal or severity-increasing in a further 8.1 percent, and possibly causal or severity-increasing in a further 12.6 percent of the 420 crashes studied. The following three studies compared the rate of vehicle defects cited in accident reports: 1. Costs and Benefits of Motor Vehicle Inspection. NHTSA, Office of State Vehicle Programs. NHTSA Technical Note DOT HS-801-614. Washington: NHTSA, Jan. 1975. This analysis showed a decline in crashes involving defective vehicles in Nebraska and Texas after inspection laws were enacted in those states, 2. Eder, Leslie E., Noel Bleich, and Mario Damiata. The NHTSA Trial Sub- stitute Motor Vehicle Inspection Programs. NHTSA Technical Report HS- 803-535. Washington: NHTSA, July 1978. By comparing crashes caused by vehicle defects in Cincinnati, which required semiannual inspections, to Ohio as a whole, which had a random inspection program, this study showed that fewer crash- involved vehicles had defects in Cincinnati. 3. O’Day, James and William L. Carlson. “Detection of Defects in Acci- dents,” SAE Paper 730584. Warrendale, Pa.: Society of Automotive Engineers, 1973. By comparing vehicle defect-related crashes on the Ohio Turnpike, Pennsylvania Turnpike, Indiana Turnpike, and in the state of Texas, this study found that the areas with a periodic inspection program had fewer crash-involved vehicles with defects. The following two NHTSA data analyses compared accident reports from inspection and non-inspection states: Page 29 GAO/RCED-90-176Periodic Inspection Programs , Appendix II Principal Studies Discussedin NHT?SA'sReport 1. “Analysis Using CARDfile,” Study of the Effectiveness of State Motor Vehicle Inspection Programs. Washington: NHTSA, Aug. 1989, p. 45. By comparing accident reports from two inspection states (Pennsylvania and Texas) and two non-inspection states (Maryland and Washington), the authors found that the non-inspection states reported a significantly higher percentage of defects on crash-involved vehicles, The difference between the states was largest for older vehicles. 2. “Further Analysis Using FARs Data,” Study of the Effectiveness of State Motor Vehicle Inspection Programs. Washington: NHTSA, Aug. 1989, p. 49. By comparing fatal accident reports from all inspection and non-inspec- tion states, the authors found that vehicles involved in fatal accidents had more defects in non-inspection states than in inspection states. The following six prior studies compared fatal accident rates between states with and without inspection programs: 1. Mayer, Albert J., and Thomas F. Hoult. Motor Vehicle Inspection: A Report on Current Information, Measurement, and Research. Detroit: Wayne State University, Institute for Regional and Urban Studies, Jan. 1963. In this comparison of death rates per mile traveled during the period 1948-59, the authors reported that states with state-operated inspection programs had lower death rates than did states with other inspection systems, which in turn had lower death rates than states with no inspec- tion systems. 2. Buxbaum, Robert C. and Theodore Colton. “Relationship of Motor Vehicle Inspection to Accident Mortalitv.” American Journal of Public Health 197-(l). July 1966, pp. 101-107.” In this analysis of 1960 traffic death rates among men aged 45-54, the authors reported results favorable to periodic inspection programs. 3. Little, Joseph W. The Fallacy of Evaluating Periodic Motor Vehicle Inspection by Death Rates. Ann Arbor: Highway Safety Research Insti- tute, 1968. Page 30 GAO/RCED-90-176Periodic Inspection Programs . Appendix II Principal Studies Diecussedin NliTSA’e Report By comparing death rates in six states that introduced periodic inspec- tion after WWII, six states that already had inspection programs, and six states that never had programs, the author found inspections to have no effect on death rates. He also concluded that fatal crash rates are not a good measure for evaluating the effectiveness of inspection programs. 4. Wart, Larry F. “Periodic Motor Vehicle Inspection: Its Accident Pre- vention Potential, Costs, and Benefits.” Springfield: Illinois Department of Transportation, Apr. 1976. By analyzing death rate trends in inspection and non-inspection states from 1949 to 1973, the author found that the trend had changed in 1968 to favor the non-inspection states for the last 5 years of his series. 6. Motor Vehicle Inspection. Harrisburg, Pa.: Pennsylvania Office of Budget and Administration, Jan. 1981. This is a regression analysis of state injury and crash data. The authors did not find significant differences between inspection and non-inspec- tion states when controlling for other important factors. 6. An Assessment of Pennsylvania’s Periodic Motor Vehicle Inspection System. Pittsburgh: Carnegie-Mellon University, Program in Engineering and Public Affairs, Dec. 1975. This study found that random-inspection states had the lowest fatality rates in relation to miles traveled, followed by states with semiannual inspection, states with annual inspection, and states with no inspection program. The following two studies compared accident rates within single states: 1. Schroer, Bernard J. and William F. Peyton. The Effects of Automobile Inspections on Accident Rates. Huntsville, Ala.: University of Alabama, Aug. 1977. This comparison of crash rates of inspected and non-inspected vehicles in Huntsville, Alabama, found that inspected vehicles had a lower crash rate, estimated at between 9 and 21 percent. Page 31 GAO/RCED-SO-176 Periodic Inspection Programs Appendix II Principal Studies Discussedin NHlSA’s Report 2. Jackson, Barry, Peter D. Loeb, and Karen A. Franck. Comprehensive Analysis of the New Jersey Motor Vehicle Inspection System. Newark: New Jersey Institute of Technology, Aug. 1982. This is a regression analysis that uses data from 1929 to 1979 in the state of New Jersey. The authors concluded that the existence of an inspection program saved an average of 304 lives and avoided 37,910 crashes per year. The following three data analyses compared accident statistics among states: 1. “Comparison of Fatal Crash Rates Across Model Years in a One-Year Crash Period,” Study of the Effectiveness of State Motor Vehicle Inspec- tion Programs. Washington: NHTSA, Aug. 1989, p. 39. By using fatal accident crash data from 49 states, the authors concluded there is no clear indication that crash involvement rates across vehicle model years are consistently different in non-inspection states and inspection states. 2. “Comparison of Fatal Crash Rates Across Crash Years,” Study of the Effectiveness of State Motor Vehicle Inspection Programs, Washington: NHTSA, Aug. 1989, p. 41. Comparing fatal accident rates of 1975 vehicles from 1975 to 1986, the authors found no trends favoring inspection states as the vehicles got older. 3. “Analysis of Total Crash Involvement Rates,” Study of the Effective- ness of State Motor Vehicle Inspection Programs. Washington: NHTSA, Aug. 1989, p. 43. NHTSA found that 4 inspection states had a 17-percent lower total acci- dent rate than 6 non-inspection states. After adjusting the data, NHTSA concluded that there was no evidence in the data examined to suggest that periodic motor vehicle inspection programs affect the crash involvement rates of older vehicles compared with newer vehicles. Page 32 GAO/RCED-SO-176 Periodic Inspection Programs F Appendix III L Major Contributors to This Report Ron E. Wood, Assistant Director Resources, Cheryl A. Donahue, Staff Evaluator Community, and Economic Development Division, Washington, D.C. Donald J. Heller, Issue Area Manager Cincinnati Regional Kenneth R. Libbey, Evaluator-in-Charge Office Matthew Byer, Staff Evaluator (242004) Page 33 GAO/RCED-90-175 Periodic Inspection Programs i ‘I’t~ltq~llorlt~ 202~275-6241 c j . ._- .-.. .._-----.--- _..__ --.- ____ -- . . ..--- -__~.-~~._ .- .Y-
Motor Vehicle Safety: NHTSA Should Resume Its Support of State Periodic Inspection Programs
Published by the Government Accountability Office on 1990-07-05.
Below is a raw (and likely hideous) rendition of the original report. (PDF)