3

CURRENTLY AVAILABLE CONSUMER AUTOMOTIVE SAFETY INFORMATION

Considerable information about vehicle safety characteristics and features is available to consumers, but not always in a comparable or readily accessible form. An overview of the key sources of vehicle safety information and dissemination outlets through which information is currently made available to consumers is provided in this chapter. A critical assessment of the strengths and weaknesses of the information from the perspective of new car purchasers and a brief explanation of why there is limited market incentive to produce better data are given. Finally, improvements in both the substance and communication of vehicle safety information are suggested.

SOURCES OF CONSUMER VEHICLE SAFETY INFORMATION

Consumers need comparative information on vehicle safety to help select a safer car. The following paragraphs highlight sources of comparative information in describing vehicle safety information currently provided by the government, nonprofit organizations, and the private sector (Table 3-1).

Comparative Vehicle Safety Data

The New Car Assessment Program (NCAP) is the primary governmental consumer safety information program. Each year the National Highway Traffic Safety Administration (NHTSA) tests the frontal crash performance of a selected number of the most popular vehicles. Forty-one model year 1995 vehicles, representing passenger cars, pickup trucks, vans, and sport utility vehicles, were tested at three facilities (NHTSA 1995a, 1–2).1



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information 3 CURRENTLY AVAILABLE CONSUMER AUTOMOTIVE SAFETY INFORMATION Considerable information about vehicle safety characteristics and features is available to consumers, but not always in a comparable or readily accessible form. An overview of the key sources of vehicle safety information and dissemination outlets through which information is currently made available to consumers is provided in this chapter. A critical assessment of the strengths and weaknesses of the information from the perspective of new car purchasers and a brief explanation of why there is limited market incentive to produce better data are given. Finally, improvements in both the substance and communication of vehicle safety information are suggested. SOURCES OF CONSUMER VEHICLE SAFETY INFORMATION Consumers need comparative information on vehicle safety to help select a safer car. The following paragraphs highlight sources of comparative information in describing vehicle safety information currently provided by the government, nonprofit organizations, and the private sector (Table 3-1). Comparative Vehicle Safety Data The New Car Assessment Program (NCAP) is the primary governmental consumer safety information program. Each year the National Highway Traffic Safety Administration (NHTSA) tests the frontal crash performance of a selected number of the most popular vehicles. Forty-one model year 1995 vehicles, representing passenger cars, pickup trucks, vans, and sport utility vehicles, were tested at three facilities (NHTSA 1995a, 1–2).1

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information TABLE 3-1 PRIMARY SOURCES OF COMPARATIVE VEHICLE SAFETY INFORMATION INFORMATION SOURCE TYPE OF INFORMATION RATING CATEGORIES/COMPARISONS NHTSA Frontal crash test (NCAP) results Five stars/within vehicle class comparisons only IIHS Offset crash test results Four rating categories/within vehicle class comparisons only   Driver death rates Indexed to average for all passenger vehicles/within and across vehicle class comparisons HLDI Injury, collision, and theft losses Five rating categories indexed to average for all passenger vehicles/within and across vehicle class comparisons CU Braking and emergency handling tests (CU); frontal crash test results (NHTSA-NCAP); injury claim rates (HLDI) Five rating categories/within and across vehicle class comparisons except for crash test results Note: NHTSA = National Highway Traffic Safety Administration; NCAP = New Car Assessment Program; IIHS = Insurance Institute for Highway Safety; HLDI = Highway Loss Data Institute; CU = Consumers Union. The first program of its kind to provide comparative information on vehicle crashworthiness, the NCAP was recently redesigned to provide more consumer-friendly information. Crash test results are now provided in a simplified star-rating system, which indicates worst (one star) to best (five stars) crash protection for vehicles in the same weight class (NHTSA 1995a, 1). The Insurance Institute for Highway Safety (IIHS) published the results of offset frontal crash tests performed at its crash test facilities for the first time in 1995 (IIHS 1995a). The performance of each of 14 mid-

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information size, four-door cars for model year 1995 was compared and given an overall score on a four-point scale.2 IIHS and the associated Highway Loss Data Institute (HLDI) also provide comparative data on the outcomes of vehicle crashes. IIHS compiles information on vehicle occupant fatalities by vehicle size and body style. Driver death rates for specific makes and models are compared with those of other vehicles in their class and with overall passenger vehicle driver death rates (IIHS 1995b).3 HLDI compiles data on three insurance loss categories—injury, collision, and theft, standardized to reduce differences because of operator age and deductible amount (HLDI 1995).4 The results are rated according to five categories and grouped by vehicle size and body style. In 1995 NHTSA required dealers to provide prospective buyers with HLDI-generated comparative information on collision losses by vehicle make and model.5 The comparative data provided by IIHS and HLDI provide consumers with useful information about real-world crash experience. However, when new model designs are introduced, they do not provide new car buyers with prospective information. Nor do they isolate the vehicle-related characteristics contributing to crash likelihood and crash outcomes from the driver and environmental characteristics. Perhaps the best-known source of new car information is Consumer Reports. Consumers Union (CU) runs its own tests of such vehicle safety characteristics as emergency handling and braking performance. Other comparative safety information is drawn from several external sources, including manufacturer-provided data on vehicle safety features and equipment, NHTSA-provided NCAP ratings, and insurance industry-provided data on injury claim rates and insurance costs. These safety data have been modified to fit within Consumer Reports' typical five rating categories. The automobile manufacturers provide a wide range of consumer-oriented safety information, including details on new car labels about specific vehicle safety features such as air bags and antilock brakes, safety information in owner's manuals, and videos and cassettes showing vehicle handling and performance characteristics and the operation of specific safety features. The information is useful for owners of new vehicles but provides little in the way of comparative data to assist consumers in making purchase decisions. Safety advocacy groups also provide consumer information. For example, the American Coalition for Traffic Safety, Inc., has developed

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information several brochures providing consumers with information on such vehicle safety features as air bags and antilock brakes and on ways to prevent such crash types as rollovers. None of these publications, however, provide comparative vehicle safety information. Summary Compilations of Vehicle Safety Information Several publications are available in which an attempt is made to compile comparative vehicle safety information in one place for distribution to consumers. CU publishes passenger car and light truck vehicle evaluations in the issues of its monthly magazine, Consumer Reports, which include a summary of safety information for each vehicle. Its April issue is devoted entirely to cars and light truck vehicles and provides detailed comparisons of hundreds of new and used vehicles, including all available safety information. In addition, CU publishes an annual New Car Yearbook and New Car Buying Guide. The New Car Yearbook, as an example, provides summary automotive safety information and evaluations of performance, cost, and features in single-page vehicle profiles (Figure 3-1). The Car Book by Jack Gillis is another comprehensive source of information, including safety information, about new cars.6 A separate chapter on safety includes information on safety features and vehicle crashworthiness. NCAP crash test results, which are organized by vehicle class, are presented in several different ways—a numerical injury index, a five-category rating, and a probability-of-injury outcome rating. Other publications, including the American Automobile Association's (AAA's) Auto Test, Consumer Guide's automobile ratings, and automobile magazines such as Car and Driver, also provide comprehensive new car information, but with much less emphasis on safety.7 NHTSA and IIHS have produced small brochures that provide comparative information focused primarily on vehicle safety. NHTSA's recently developed consumer guide, Buying a Safer Car, was cosponsored by AAA and the Federal Trade Commission. The guide provides several safety buying tips. Vehicle-specific information follows on such topics as availability of safety features (i.e., driver and passenger air bags, antilock brakes, adjustable seat belt anchors, and 1997 side-impact protection), crash test results, and theft ratings, organized by vehicle class. IIHS has developed a booklet entitled Shopping for a Safer Car that introduces the consumer to such basic safety concepts as vehicle size

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information FIGURE 3-1 Example of comparative vehicle safety information provided by Consumers Union (1996 New Car Yearbook). Note: edited for illustrative purposes.

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information and structure. Various safety features are highlighted (dual air bags, side-impact protection, head and child restraints), and cars that have well-designed features are identified. The brochure directs consumers to other safety information sources, such as crash test results. Individual insurance companies also provide summary vehicle safety information to their policyholders. For example, the United Services Automotive Association (USAA), the nations' sixth-largest insurer of motor vehicles, publishes a booklet entitled The Car Guide. The guide provides a vehicle safety and insurance “report card” on specific vehicles organized by manufacturer. Performance ratings are given for crash test results, fatality rates, injury frequency and collision loss, theft loss, and claims experience. This brief overview indicates that some comparative vehicle safety information is available to consumers. As discussed in the following section, however, consumers must be knowledgeable and diligent to access the information and give it proper consideration. DISSEMINATION OUTLETS Little comparative vehicle safety information is available at the dealership (Table 3-2). Safety features, such as antilock brakes and air bags, are mentioned on new car vehicle labels along with many other features. More detailed information about the operation of safety features is contained in owner's manuals. The only comparative safety information is the booklet on differences in collision losses that NHTSA now requires all dealers to make available to prospective new car purchasers. More information is available to consumers before they go to the showroom (Table 3-2), but obtaining comparable safety data on a wide range of vehicles is difficult. Consumer Reports has a wide circulation. Nearly 5 million subscribers receive this publication, including libraries, and nonsubscription sales are large. 8 The New Car Yearbook is for sale on the newsstands in the fall, and New Car Buying Guide is sold by mail, in bookstores, and on newsstands. The same data are provided to consumers by the major commercial on-line services —America On Line, CompuServe, and Prodigy. Finally, CU recently published a CD-ROM entitled Consumer Reports Cars: The Essential Guide. With average sales of about 75,000 per year, Jack Gillis's The Car Book reaches a smaller but still significant audience. NHTSA has recently increased the visibility of its NCAP crash test results in response to a congressional request to revamp the communi-

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information TABLE 3-2 PRIMARY DISSEMINATION OUTLETS FOR CONSUMER VEHICLE SAFETY INFORMATION DISSEMINATOR TYPE OF COMMUNICATION MODE FREQUENCY Point-of-Sale Information Automobile manufacturers New car labela (list of safety features) Print N.A.   Owner's manuala Print N.A. NHTSA Collision loss booklet Print Annual Presale Information CU Magazine: Consumer Reports Print and on-line Monthly   Magazine: New Car Yearbook Print Annual   Book: New Car Buying Guide Print Annual   CD-ROM: Consumer Reports Cars: The Essential Guide Electronic Periodic AAA/NHTSA Brochure: Buying a Safer Car Print Periodic

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information NHTSA Pamphlet: NCAP test results Print and electronic (through NHTSA hotline) Periodic releases throughout the year IIHS Booklet: Offset frontal crash test results Print (publicized through NBC Dateline) Periodic   Pamphlet: Shopping for a Safer Car Print Annual   Special Issue: Death Rates by Vehicle Make, Series Print Annual HLDI Pamphlet: Injury, Collision, and Theft Losses Print Annual Note: N.A. = not appropriate; NHTSA = National Highway Traffic Safety Administration; CU = Consumers Union; AAA = American Automobile Association; NCAP = New Car Assessment Program; IIHS = Insurance Institute for Highway Safety; HLDI = Highway Loss Data Institute. a No comparative information.

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information cations aspect of the program. In addition to regular press releases, the agency inaugurated a hotline in 1994 through which electronic or printed versions of the crash test results are made available. Approximately 35,000 requests for NCAP data were received through the hotline in 1994 (NHTSA 1995a, 3). NHTSA may achieve wider circulation of NCAP test results from its new consumer guide, Buying a Safer Car, which incorporates the NCAP test results and is available through AAA as well as the NHTSA hotline. 9 Finally, NCAP test scores are picked up in both The New Car Yearbook and The Car Book. Consumer-oriented comparative vehicle safety information is also available in the summary brochures and special reports from IIHS and HLDI described earlier.10 Consumers, however, must know where to call to receive these publications. A few individual insurers like USAA send out comparative vehicle safety information with their premium notices. IIHS itself recently launched its offset frontal crash testing program with a highly visible NBC Dateline presentation, which resulted in numerous requests for its consumer-oriented booklet summarizing the test results.11 This brief overview of communication modes and dissemination outlets for consumer vehicle safety information suggests that the information is scattered among several providers. Little comparative information is available at the dealership. More presale information is available, much of it in print form—special magazine issues, brochures, and pamphlets. With the exception of Consumer Reports and perhaps The Car Book, however, which are widely known, consumers may not be aware of the publications or where to obtain the information. DEVELOPMENT OF THE NCAP STAR-RATING SYSTEM: A CASE STUDY Recent efforts to develop a more consumer-oriented presentation and communication of NCAP test results provide a good illustration of the difficulties involved in the development of meaningful consumer automotive safety information. In fiscal year 1992 the Senate and Conference Appropriations Reports requested that NHTSA implement improved methods to inform consumers of the comparative crashworthiness of passenger vehicles as measured by NCAP crash test results.12 Up to that time, NHTSA had simply published numerical injury scores indicating the likelihood of

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information head, chest, and upper leg injuries. Responding to focus groups calling for presentation of the crash test information “in a form that is non-technical and as short and simple as possible,” NHTSA developed a new star-rating format and inaugurated a hotline through which consumers could request the information (NHTSA 1993, 43, 58). Beginning with model year 1994 vehicles, the new rating system combines the head and chest injury crash test scores into a single rating of one to five stars, one star indicating the least crash protection for drivers and front seat passengers and five the most.13 The number of stars relates to the probability of serious injury. One star is equivalent to a 45 percent or greater chance of sustaining a serious injury in a frontal crash, and five stars is equivalent to a 10 percent chance of sustaining a serious injury. This information on injury likelihood, however, is not given to the consumer. A brief cover sheet provides basic information about the crash test and reiterates that the results should be used only to compare vehicles in the same weight class. Specific vehicles are then grouped by weight and rated accordingly. The new rating system provides consumers with a more readily understandable summary measure of vehicle occupant protection in frontal crashes than the old numerical scoring system.14 At the same time it has raised several concerns, many from the safety community. The first set of concerns relates to the presentation of the information. Some critics maintain that the use of stars as rating symbols is inappropriate. The public generally associates stars with good performance. Yet a one-star NCAP rating means a 45 percent or greater probability of sustaining a serious injury. In the critics' view, vehicles that receive this rating should not receive a star or should have a negative rating (IIHS 1995c, 3). A related concern is the technical basis for selecting five rating categories. Critics maintain that the star system translates relatively small differences in injury criteria into what can be perceived as more major differences—for example, three stars versus four. There appears to be no statistical evidence that the differences in the ratings are systematically related to differences in crashworthiness (IIHS 1995c, 3).15 In fact, an analysis of model year 1995 ratings suggests that most cars fall within a narrower rating band of three to five stars, although light trucks, vans, and sport utility vehicles exhibit a more even distribution among the five categories (Figure 3-2).16 Furthermore, no estimate is provided of the variance within each rating category, a nontrivial issue.

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information FIGURE 3-2 NCAP star ratings of the frontal crashworthiness of passenger cars (above) and light trucks, sport utility vehicles, and vans (below), model year 1995 (NHTSA 1995c).

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information Large variances in individual test scores that underlie the ratings because of vehicle or test differences could render distinctions among the rating categories meaningless. A final set of concerns relates to the potential for the star ratings to mislead. For example, consumers may compare ratings for vehicles in different weight classes, which is not valid. For example, they may conclude erroneously that a small compact with a five-star rating is as generally crashworthy as a large sedan with a similar rating. NHTSA warns the consumer against making such comparisons in its literature and groups vehicle ratings by vehicle weight class. This caveat can easily be overlooked, however, when vehicles and related rating information are listed alphabetically or by manufacturer, as they are in several consumer guides. Consumers may make the same mistake if numerical scores are provided, but the simplified star-rating scheme makes it even easier to mistakenly compare scores across car classes (IIHS 1994, 4). STRENGTHS AND WEAKNESSES OF CURRENT INFORMATION As the NCAP star-rating system illustrates, providing consumers with meaningful, yet not misleading, comparative information is a challenge. The available information, although limited, gives consumers some basis on which to compare vehicle safety. Most of the literature identifies passenger vehicles that have desirable safety features. Of course, once the vehicle is purchased, much information is available from owner's manuals and safety advocacy groups about how to use these safety features. Consumers can also use NCAP ratings to compare the frontal crashworthiness of vehicles within the same weight class. As indicated in the preceding chapter, there is evidence that the underlying test scores have some correlation with differences in real-world crash performance, at least for vehicles that rank at the top or the bottom of the rating scale.17 Finally, consumers can use IIHS and HLDI data to examine real-world outcomes, such as driver and occupant fatalities, by vehicle make and model. Death rates, however, are influenced by driver characteristics and driving conditions as well as vehicle design. Another frequently overlooked benefit of consumer information is the incentive provided to manufacturers to improve the safety design of vehicles. The NCAP is a case in point. NCAP scores have improved

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information steadily since the inception of the program, with the largest improvement in the early years. Most passenger cars now meet the lower [48-km/hr (30-mph)] regulatory standard at the higher [56-km/hr (35-mph)] NCAP test speed. The results are not as good for light trucks, vans, and sport utility vehicles, because crashworthiness standards were phased in later for light truck vehicles than for cars (Kahane et al. 1994, 13). Improvements in test performance have been matched by real-world reductions in fatality likelihood for drivers in head-on crashes similar to those simulated by the NCAP test.18 The NCAP is not the sole stimulus for this improvement in safety; the 1984 regulations leading to automatic passenger restraint systems and air bags were another important factor. But the program can claim part of the credit. Given all the data listed earlier, why do consumers not have adequate information on which to compare and make choices about vehicle safety? Several reasons are apparent. First, there are critical gaps in current information. Most significant, the important role of vehicle weight and size in crash outcomes is not emphasized in consumer safety literature, with the exception of the IIHS brochure Shopping for a Safer Car. Comparative safety information is also limited for vehicles within the same size and weight class. For example, NCAP crash protection ratings reflect vehicle crashworthiness for only one very limited crash type—head-on collisions. No comparative information is available about how vehicles perform in side-impact crashes and rollovers, two crash types resulting in significant numbers of fatalities. IIHS has begun to rate vehicle crashworthiness in offset crash tests, representing a more common type of frontal crash. The rating categories are different from the NCAP system. Moreover, the probability of head and chest injury measured by the two tests may differ. This result is not surprising because the tests measure different aspects of vehicle crashworthiness.19 In fact, one could argue that testing more crash types is beneficial because it encourages manufacturers to optimize safety design for more crash types. However, explaining these differences in any detail, or simply presenting the results without explanation, is likely to confuse the consumer. The results of comparing information from crash tests with real-world crash data for the same vehicle are frequently, but not always, consistent. For example, a comparison of the NCAP crash test ratings with actual fatality rates for the four-door Ford Escort, a popular car in the small car category, provides conflicting information. The 1995 model year Ford Escort receives a four-star NCAP rating for both driver

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information and front seat passenger, only one star below the highest rating. Yet real-world crash data indicate a driver death rate about 50 percent higher than for other vehicles in its class. Similar results can be shown for Chevrolet's 1995 model year Geo Metro four-door sedan (NHTSA 1995c; IIHS 1995b). Part of the difference may be attributed to the use of dual air bag-equipped cars for model year 1995 NCAP tests. Crash test results for 1990–1993 were based on belt-equipped vehicles. Part of the difference may reflect the influence of driver and vehicle use characteristics in the crash data. These distinctions serve to highlight the difficulties that consumers may find in attempting to reconcile such information. Where an attempt is made to bring safety data together in one place (e.g., CU's New Car Yearbook, New Car Buying Guide, and CD-ROM Consumer Reports Cars: The Essential Guide; NHTSA's Buying a Safer Car; IIHS's Shopping for a Safer Car; USAA's Car Guide), the information available to the public is often incomplete. Missing NCAP ratings are most noticeable, reflecting the relatively small number of vehicles tested and the timing of the release of test results, which is not necessarily coordinated with consumer or insurance industry publication deadlines. (Test dates can also vary widely, particularly for test results from prior years for models that have not changed substantially.) It is clear that the public would benefit from an increased number of models evaluated in the NCAP. Consumer Reports also does not rate every automobile yearly. Finally, even when the information is available, consumers are not always provided an explanation of the limited predictive power of the crash tests generally, the certainty of the ratings themselves, or how to combine the information. In summary, information is available that enables consumers to compare the safety performance of different vehicles, but much more could be done to make the information more comprehensive and easier for consumers to interpret and use. WHY BETTER INFORMATION IS NOT PROVIDED When considering the need for improvements in information about vehicle safety, one could ask why such information is not being provided by the market, particularly if consumer interest in automotive safety information is growing. The answer lies in part in the nature of the information. Product safety information is a typical example of what economists call a “public good.” The defining property of a public good

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information is that, once the good is provided, the consumption of that good by one individual does not diminish its availability for consumption by others (Rosen 1995, 61). A related characteristic is the difficulty of excluding those who are not willing to pay for consuming the good (Rosen 1995, 62). Thus, a consumer does not have an incentive to pay the full cost of obtaining the information and can become a “free rider.”20 As a result, firms providing information are unable to capture through their pricing the full value of their informational product to consumers. Thus they will provide less information than what consumers in the aggregate would be willing to pay for—an amount that an efficient market would supply. The insurance industry (through IIHS and HLDI) and CU provide comparative vehicle safety information for which consumers are willing to pay —directly in the case of CU and indirectly through insurance premiums in the case of the insurance industry. However, the large expense of crash testing—the capital expense of establishing a test facility and operating costs of approximately $15,000 per test plus the cost of the vehicle—is a barrier to any one organization conducting multiple tests per vehicle on the full range of vehicles available each year necessary to develop comprehensive comparative summary data. For example, CU regularly conducts tests on a wide range of products, but, as a nonprofit consumer organization, it could not absorb or recover the added costs of a complete crash testing program for passenger cars and light truck vehicles. In accord with its mission, CU publications are widely available to the public in libraries and through the media, which publicize CU's information, thus exacerbating the free rider problem. The insurance industry has funded a state-of-the-art crash test facility and begun an offset crash testing program, now being expanded. Here, too, the industry on its own, even through a pooled effort, cannot hope to recoup the cost of a major expansion of its crash testing program, in part because of the free rider problem and in part because of the structure of regulated rates, which are not closely linked with the crash performance or safety records of insured vehicles. Moreover, the direct benefits to automobile insurers from safer vehicles are affected by the tort liability system in effect in most states. Payments for injuries to “third parties” (occupants of other vehicles, pedestrians, bicyclists), not to the occupants of the vehicles they insure, are the largest injury-related claims costs for insurers. Finally, the automobile manufacturers conduct crash tests to meet regulatory standards, but test results are proprietary. The manufacturers also face significant barriers to providing accurate comprehensive

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information comparative safety data to consumers. The cost of testing would be high relative to the benefits that any one company would receive and consumers might perceive the test results to be biased toward the cars sold by that company. To date, NHTSA has paid for the major crash test program, NCAP, that provides consumers with crash test data. It is probably for these reasons that the market has failed to provide more comprehensive comparative safety information and that public funds have supported the NCAP. A cost-sharing approach involving the insurance and automobile industries and NHTSA could help provide the incentives and the resources to conduct the necessary crash tests and produce the comparative information. MAKING BETTER USE OF EXISTING INFORMATION The presentation and communication of existing consumer vehicle safety information could be improved in a number of ways. First, critical factors that affect vehicle safety should be highlighted in consumer safety publications. For example, the important relationship between vehicle size and weight and occupant protection is well understood but needs to be more prominently featured in consumer-oriented safety publications. Second, common misconceptions about use of vehicle safety features should be addressed. For example, it should be made clear that antilock brakes should not be pumped and that air bags do not replace seat belts. Third, further improvements could be made in the government's primary consumer safety information program, NCAP. Much time and effort have gone into reviews of the program, particularly the new star-rating system, but many issues remain unresolved. Several organizations have questioned the use of the star symbol and the basis for the selection of rating categories (NHTSA 1995d; IIHS 1995c). If the star symbol is retained, consumers should at least be provided information about the outcomes (i.e., likelihood of injury) associated with each rating category. They should also be given information on the frequency of the frontal crashes that the crash tests represent. Fourth, questions about the reliability of the underlying test data persist. These concerns may best be met by a new study, which would attempt to document the variance from test to test and the sources of that variance. Such a study was recommended by the General Accounting Office in its recent report (GAO 1995, 9).21 Fifth, conducting crash tests at different speeds and for other than frontal crashes could provide consumers with more comprehensive

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information crashworthiness data than are currently available, but expansion of crash testing is highly controversial. The manufacturers oppose any expansion of crash testing until the reliability of the new test modes and correlation with real-world crashes are better understood (NHTSA 1995d, 18–20). Safety groups, in contrast, believe that NCAP should continue to be a cutting-edge program. Thus, new crash tests should be considered to provide incentives for industry to improve vehicle performance in other than full-frontal crash types (NHTSA 1995d, 20–21). Offset crash tests are the most likely candidate in the near term. IIHS is gaining experience from its offset crash testing program, the Australians are conducting both full-frontal and offset crash tests for consumer information, and the European Union will soon introduce offset crash testing (NHTSA 1995d, 26; IIHS 1995d, 5). If information is shared, the results of these efforts could be sufficient to support an expanded offset crash testing program in the United States. As discussed in Chapter 2, introduction of consumer information programs on vehicle crashworthiness in side-impact and rollover crashes is probably premature. In the long run, entirely different types of crash tests may be developed using advanced computer techniques. Finally, providers of vehicle safety information could improve the presentation and dissemination of information to the extent funding permits. Current efforts to compile summary information in one place, such as CU's New Car Yearbook, New Car Buying Guide, and CD-ROM Consumer Reports Cars: The Essential Guide; Jack Gillis's The Car Book; and the NHTSA and IIHS brochures, should be continued. Improvements in the structure and format of brochures and print documents currently published could be made, with empirical evaluation and systematic application of principles from research on text comprehension (Kintsch 1986; Ericsson 1988; Schriver 1989; Atman et al. 1994). These publications could also be made more accessible and available to consumers. More information could be made available electronically. More general access to the media can be accomplished by staging newsworthy events (Maibach and Holtgrave 1995, 227), such as IIHS's offset crash tests, which were broadcast on NBC's Dateline. Of course, advertising the availability of vehicle safety information (e.g., through public service announcements and consumer magazine inserts) should increase consumer awareness of the information's existence and of where and how it can be obtained. Even with all these improvements, consumers are still likely to have difficulty integrating available information into a summary as-

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information sessment of the relative safety of various vehicles in a form that serves their needs. Much more could be done to provide consumers with summary measures of overall vehicle safety that are easier to interpret and use in making purchase decisions. Development of more comprehensive information requires a better understanding of what consumers believe about safety and how they consider safety in the process of choosing a new vehicle, the subject of the following chapter. NOTES 1. Published information also includes ratings from prior years for models that were not substantially changed. 2. Using the same four-point scale, the cars were also rated on their structural performance; effectiveness of the occupant restraint systems; injury likelihood, including measures of head, neck, chest, and leg injury; head restraint design; and bumper performance (IIHS 1995a). 3. Fatality rates are drawn from NHTSA's Fatal Accident Reporting System, and vehicle registration counts come from R.L. Polk. 4. Operator age is based on the age of the “rated driver.” The rated driver and the actual driver involved in a claim are not always the same. 5. The information reflects the average collision loss payment per insured vehicle year, which is presented as an index with 100 representing the average for all passenger vehicles (NHTSA 1995b, 1). The rating provides a measure of a vehicle's damage susceptibility, which only indirectly relates to its safety performance. 6. Separate books are available for passenger cars and for trucks, vans, and sport utility vehicles. 7. Auto Test, for example, does not include NCAP crash test results, although these may be found in the brochure AAA cosponsors with NHTSA entitled Buying a Safer Car. 8. CU estimates that the New Car Yearbook reaches about 14 million individuals if nonsubscription sales and indirect dissemination channels (e.g., libraries) are taken into account (personal communication, David Pittle, Consumers Union, Oct. 23, 1995). 9. For each brochure it disseminates, NHTSA must pay 50 cents. The copies are distributed from the National Consumer Information Center in Pueblo, Colorado, a repository of consumer information from all federal agencies. 10. For example, in 1995 there were requests for about 140,000 copies of IIHS's Shopping for a Safer Car and about the same number for HLDI's Injury, Collision, and Theft Losses. Copies of both of these publications were also distributed by sponsoring insurance companies, either under the name of the company or that of IIHS or HLDI. 11. IIHS's publications office estimated that it received 18,000 to 20,000 requests for the publication summarizing offset test results. Most requests followed the NBC Dateline presentation.

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information 12. The reports also requested a study of the validity of the test data and the efficacy of allowing manufacturers to choose between “high-tech” and “low-tech” dummies for the purposes of NCAP testing (NHTSA 1993). 13. The star system relates head and chest injury test scores to injury probabilities using injury risk functions developed by biomechanical experts from the Ford Motor Company and General Motors Corporation. The risk functions relate the dummy measurements to injury probabilities (Hackney and Kahane 1995, 1). To calculate the injury probability rating, the head and chest scores are combined (although each is treated as an independent event), because an individual who suffers multiple injuries is likely to have a higher risk of permanent disability or death (Hackney and Kahane 1995, 2). Upper leg loads were not included because femur injury is seldom life threatening (Hackney and Kahane 1995, 2). 14. The numerical scores are published and made available to consumers on request. 15. In the judgment of agency staff, five rating categories were needed. Three did not provide enough variation, and four did not give a midpoint (personal communication, James Hackney, NHTSA, Aug. 10, 1995). 16. The analysis includes the 41 tested model year 1995 vehicles plus results from previous years for tested vehicle types that have not changed. 17. The researchers cautioned that the findings apply on the average. There is no guarantee that every specific make-model with a very poor NCAP score or, alternatively, an excellent score has a higher-or lower-than-average fatality risk, respectively, in head-on collisions (Kahane et al. 1994, 13). 18. The researchers found a 20 to 25 percent reduction in fatality risk for belted drivers in actual head-on collisions in model years 1979 through 1991, with the largest decreases during the early 1980s (Kahane et al. 1994, 13). 19. Full-frontal tests are more demanding of occupant restraint systems, whereas offset crash tests are more demanding of the structural integrity of the occupant compartment. For a more detailed explanation, see Chapter 2. 20. The importance of the free rider problem is an empirical question. For example, consumers purchase Consumer Reports, and their insurance companies, acting as their agents, pay for product safety information. The contributions of the insurance companies to IIHS and HLDI are examples of how a public good can be paid for in a market economy. 21. GAO recommended an update of the NCAP test variability study that NHTSA conducted in 1982 (see Chapter 2 for more details about that study). REFERENCES Abbreviations IIHS Insurance Institute for Highway Safety GAO General Accounting Office HLDI Highway Loss Data Institute NHTSA National Highway Traffic Safety Administration

OCR for page 56
Shopping for Safety: Providing Consumer Automotive Safety Information Atman, C.J., A. Bostrom, B. Fischhoff, and M.G. Morgan. 1994. Designing Risk Communications: Completing and Correcting Mental Models of Hazardous Processes, Part I. Risk Analysis, Vol. 14, No. 5, pp. 779–788. Ericsson, K.A. 1988. Concurrent Verbal Reports on Text Comprehension. Text, Vol. 8, No. 4, pp. 295–325. GAO. 1995. Highway Safety: Reliability and Validity of DOT Crash Tests. GAO/ PEMD-95-5. May, 76 pp. Hackney, J.R., and C.J. Kahane. 1995. The New Car Assessment Program: Five Star Rating System and Vehicle Safety Performance Characteristics. No. 950888. Presented at Society of Automotive Engineers International Congress and Exposition, Detroit, Mich., 16 pp. HLDI. 1995. Injury, Collision, and Theft Losses by Make and Model. Arlington, Va., Sept. IIHS. 1994. Future of NCAP: NHTSA Mulls Options To Expand Crash Test Program. Status Report, Vol. 29, No. 7, June 25, pp. 4–5. IIHS. 1995a. Crashworthiness Evaluations. Arlington, Va. IIHS. 1995b. Special Issue: Driver Death Rates by Vehicle Make, Series. Status Report, Vol. 30, No. 9, Oct. 14. IIHS. 1995c. NCAP's Star Ratings Simply Don't Work, Safety and Consumer Groups Advise NHTSA. Status Report, Vol. 30, No. 1, Jan. 14, p. 3. IIHS. 1995d. European Union Moves Toward New Safety Standards with Dynamic Tests . Status Report, Vol. 30, No. 7, Aug. 12. Kahane, C.J., J.R. Hackney, and A.M. Berkowitz. 1994. Correlation of Vehicle Performance in the New Car Assessment Program with Fatality Risk in Actual Head-On Collisions. Paper No. 94-S8-O-11. National Highway Traffic Safety Administration, 17 pp. Kintsch, W. 1986. Learning from Text. Cognition and Instruction, Vol. 3, pp. 87–108. Maibach, E., and D.R. Holtgrave. 1995. Advances in Public Health Communication. Annual Review of Public Health, Vol. 16, pp. 219–238. NHTSA. 1993. New Car Assessment Program. Response to the NCAP FY 1992 Congressional Requirements. U.S. Department of Transportation, Dec., 125 pp. NHTSA. 1995a. The New Car Assessment Program. Office of Market Incentives, U.S. Department of Transportation, 5 pp. NHTSA. 1995b. Comparison of Differences in Insurance Costs for Passenger Motor Vehicles on the Basis of Damage Susceptibility. U.S. Department of Transportation, Feb., 10 pp. NHTSA. 1995c. New Car Assessment Program, Test Results, Model Year 1995. U.S. Department of Transportation, May, 10 pp. NHTSA. 1995d. New Car Assessment Program Public Meeting. U.S. Department of Transportation, Feb., 28 pp. Rosen, H.S. 1995. Public Finance (4th ed.). Richard D. Irwin, Inc. Schriver, K. 1989. Evaluating Text Quality: The Continuum from Text-Focused to Reader-Focused Methods. IEEE Transactions on Professional Communication, Vol. 32, No. 4.