Printer Friendly

All-terrain vehicles: market reaction to risk information.

I. INTRODUCTION

Several recent studies have shown a systematic relationship between well-publicized product recalls and capital market losses for the firms involved. Jarrell and Peltzman |1985~ found that drug recalls by the Food and Drug Administration (FDA) resulted in an average 6 percent loss in stock equity values for the firms affected, and that some of the effects of the recall on stock equity values spilled over to other drug companies not directly affected by the recall.(1) In a similar study, Rubin, Murphy, and Jarrell |1988~ estimated that product recalls by the U.S. Consumer Product Safety Commission resulted in an average 7 percent reduction in stock equity values of the firms involved.

The reduction in stock prices of firms involved in recalls is not surprising; with efficient capital markets, stock prices are expected to reflect the full costs of regulatory action on firms and shareholders. Interestingly, however, the magnitude of the reduction in equity values substantially exceeded the estimated direct costs of recalls and suggests that recalls may exert a powerful deterrent effect on the production of defective products. The explicit factors leading to these stock price reductions are unclear, but probably reflect anticipated reductions in demand for the firms' products, loss of reputation, and the increased risk of product liability suits, as well as the direct costs of recalls. The size of the loss also appears to be related to the degree of publicity; according to the Jarrell and Peltzman results, equity losses were higher for recalls publicized in the Wall Street Journal and for products that were completely withdrawn from the market.

In a similar vein, Viscusi and Hersch |1990~ evaluated stock market repercussions of product liability suits and regulatory violations reported in the Wall Street Journal. They found that product liability news coverage had a significantly negative effect on stock prices that varied with the amount of information and publicity transmitted to investors. Although there was more of a reduction in stock market returns in the case of product liability actions (when controlling for variations in information presented), the adverse effects of regulatory violations far exceeded the dollar value of the regulatory penalties.(2)

The focus of these studies has been on investor reaction to the anticipated impacts of recalls, product liability litigation, and other regulatory violations.(3) However, the studies are unable to measure directly the consumer market response to new risk information, which is presumably an important factor in investor calculations. This suggests the following questions: Are consumers influenced by regulatory actions as anticipated by investors? Do consumers in fact exhibit rational economic behavior when presented with new risk information?

To explore these questions this paper examines the market reaction to risk information implicit in a recent, well-publicized consent decree between the U.S. Consumer Product Safety Commission and the distributors of all-terrain vehicles (ATVs).(4) ATVs are three- and four-wheeled motorcycle-type vehicles intended for off-road use. The three- and four-wheeled models are functionally similar and are used for both recreational and nonrecreational activities. The consent decree stopped the sale of the allegedly more risky three-wheeled vehicles in the new product market, but did not affect the sale of new four-wheeled ATVs or the second-hand market for used ATVs. The analysis evaluates the impact of the consent decree in the market for used ATVs.

If, in fact, the consent decree and concomitant publicity provided new risk information to the public we would expect the relative price of three-wheeled ATVs to decline in the market for used ATVs. Such a decline would suggest a direct link between risk information and consumer market behavior. Moreover, since sellers of used ATVs (particularly individuals) are not liable for injury and have no reputations to lose, the results of the study should provide a pure measure of demand reduction.

Section II develops a simple model for estimating the impact of product risk on the price of substitute goods. Section III provides a background discussion of the ATV market and product risk. Section IV describes methodology and data sources. Results are described in section V and discussed in section VI.

II. THE IMPACT OF RISK ON THE PRICE OF SUBSTITUTES

To examine the impact of product risk on the price of substitute goods, assume that individuals maximize utility by consuming two risky products, |x.sub.1~ and |x.sub.2~, and a riskless composite consumption good, q. Let the utility function of a consumer be written as

U = U(|x.sub.1~, |x.sub.2~,q~).

Also assume that for each |x.sub.i~, i = 1,2, there is a given accident risk, |r.sub.i~, and a known loss in the event of accident, given by |c.sub.i~. Following Oi |1973~, if the market price of each |x.sub.i~ is given by |p.sub.i~, the expected full price, including expected accident losses, can be represented by |p.sub.i~ + |r.sub.i~|c.sub.i~.

If the time horizon is extended and there is a competitive insurance market, or if individuals can practice self-insurance with a program of sequential purchases combined with borrowing or lending, the consumer can demand any quantity of the risky products |x.sub.1~ and |x.sub.2~ at their expected full price. The Lagrangian function of the optimization model can be written as

|Mathematical Expression Omitted~

where y is income, |p.sub.q~ is the market price of the composite consumption good, and h is the Lagrangian multiplier.

Maximizing the constrained utility function yields the usual equality between the marginal rates of substitution and relative product prices. However, the relative prices for the risky products |x.sub.1~ and |x.sub.2~ include components for expected accident costs. That is,

|Mathematical Expression Omitted~

Assuming that |x.sub.1~ and |x.sub.2~ are close substitutes, the difference in market prices should approximate the difference in the expected accident costs. That is,

|P.sub.1~ - |P.sub.2~ = |r.sub.2~|c.sub.2~ - |r.sub.1~|c.sub.1~.

Since expected accident costs are related to perceived product risk, new risk information which effectively alters consumer perceptions should result in a change in the price differential.

III. MARKET BACKGROUND

ATVs are three- and four-wheeled motorized vehicles intended for off-road use on various types of nonpaved terrain. Although, as indicated by Deppa and Hauser |1989~, some of the vehicle dynamics of three- and four-wheeled ATVs differ, they are functionally similar vehicles and can generally be used interchangeably. They are produced primarily by Japanese motorcycle manufacturers and have been imported and sold in this country since the early 1970s. There are currently about 2.5 million ATVs in use, and prices of new ATVs range from about $2,000 to $4,000.

There exists an active market for used ATVs. A recent survey of consumers developed by the Consumer Product Safety Commission and conducted by Market Facts, Inc. |1988~ indicated that about half of the ATVs purchased in 1986 were purchased in the secondhand (or "used") market; this suggests sales of almost half a million ATVs in the used market in 1986. The size of the market for used ATVs is not surprising; ATVs are costly, durable consumer products that retain a significant residual value over time.

The survey also found that about 80 percent of ATVs purchased in the used market were purchased from individuals, and about 20 percent were purchased from franchised and nonfranchised motorcycle and ATV dealers. The used market appears to be well-functioning. Survey results indicated that about 76 percent of used ATVs were sold in less than one month, and about half (51 percent) were sold in less than one week. Moreover, the reported sales price averaged only about 7 percent less than the reported initial asking price.

Concern about the safety of ATVs grew during the 1980s as the level of injuries rose. In 1985, for example, the Consumer Product Safety Commission estimated that there were about 86,000 ATV-related injuries treated in hospital emergency rooms (about a 2 percent annual driver injury risk), and about 295 ATV-related deaths (Fulcher and Reiff |1991~). The Commission began a regulatory proceeding in 1985 to evaluate the hazards associated with ATVs and to determine what, if any, regulatory actions should be taken. The results of a staff evaluation were presented to the Commission in November 1986, and formed the basis for the preliminary consent decree between the Commission and ATV distributors which was signed on 30 December 1987 (Consumer Product Safety Commission, |1986~; U.S. District Court, |1987~).

A major component of the preliminary consent decree (the "consent decree") was an agreement by distributors not to sell new three-wheeled ATVs and to repurchase all new three-wheeled ATVs that remained in dealer inventories. The stop-sale followed directly from agency engineering and risk findings. Engineering analysis, described in Deppa and Hauser |1989~, showed that the lateral stability of three-wheeled ATVs, a measure of resistance to sideways rollover, was less than that of their four-wheeled counterparts.

The engineering findings were consistent with the results of a parallel risk analysis, described in Rubinfeld and Rodgers |1992~, which used binary qualitative response probability models to analyze cross-section data from surveys of users and accident victims. The probability models assumed that the risk of injury was a function of driver characteristics (e.g., age, gender), driver use patterns (e.g., riding time, recreational vs. nonrecreational use), and characteristics of the ATV driven (e.g., engine size, number of wheels). This analysis found that the injury risk on three-wheeled ATVs was about twice the risk on comparable four-wheeled ATVs.

IV. METHODOLOGY AND DATA

The change in the wheels-related price differential following the consent decree is estimated by means of a hedonic price analysis (Rosen |1974~) and is calculated with an ordinary least squares regression model. It is assumed that ATV prices are a function of a number of ATV attributes and characteristics such as (among others) engine size and number of wheels.

The primary model (model 1) pools cross-section data on the characteristics of ATV models with information on their prices in the used market for two time periods: the period prior to the consent decree (September-December 1987) and the period immediately following the consent decree (January-April 1988). A second model is also estimated to determine if the change in the price differential following the consent decree was permanent. The second model compares used market prices in the pre-consent decree September-December 1987 time period with prices in the January-April 1989 time period, a year following the consent decree.

The analysis is based on pre- and post-consent decree used prices for 108 ATV models produced by the five major manufacturers during the 1985-87 model years. Data were gathered from two sources. Data on the mechanical attributes of ATVs were gathered from the annual "buyer's guide" issues of ATV News, a publication intended for enthusiasts. Price data were gathered from several National Automobile Dealer Association appraisal guides for ATVs (National Automobile Dealer Association |1988-89~).

The Association's appraisal guides, often referred to as "blue books," are published three times annually. Price quotations are based on estimates of the average sales price for specific ATV models in the used market during the preceding time period.(5) For the periods under consideration, price estimates were based on actual sales data provided by about 150 motorcycle-ATV retailers from across the nation.

The price data were supplemented with interviews of dealers to determine market conditions. Some asking prices from newspaper listings were also factored into the price estimates.(6)

V. STATISTICAL RESULTS

Results of the analysis are reported in Table I. Both equations are significant, with adjusted |R.sup.2~ values of 0.94 or greater. Engine size, measured in cubic centimeters of displacement (ccs), is a key variable and indicates that ATV prices are directly related to engine size. All of the independent variables except engine size are defined as dummy variables, indicating the presence or absence of specific characteristics.

Three variables are of primary interest. The variable Four-Wh is a dummy variable defined to be one if the ATV had four wheels, zero if it had three wheels. The variable Post-Con is a dummy variable defined to be one if the used market price of the vehicle was from the post-consent decree time period, zero if it was from the pre-consent decree time period. The coefficient for this variable is expected to reflect the normal depreciation of used vehicles over time, as well as the possible effects associated with the consent decree. Finally, the variable Four-wh |Mathematical Expression Omitted~ Post-Con is an interaction term defined to be the product of the Four-Wh and Post-Con variables. Thus, it is equal to one if the ATV observation had four wheels and the used market price was from the post-consent decree time period.(7) All three variables are highly significant.

We are primarily interested in the change in the wheels-related price differential immediately following the consent decree, described in model 1. The results indicate that prior to the consent decree the price of a used four-wheeled ATV was about $275 more than the price of a similar three-wheeled ATV. Following the consent decree the price differential rose significantly to about $461, an increase of $186.

The change in used prices for three- and four-wheeled ATVs following the consent decree may also be of some interest. The results indicate that the price for a three-wheeled ATV of a given model year declined by about $234 in the post-consent decree time period, relative to the pre-consent decree price. In contrast, the price of a used four-wheeled ATV of a given model year declined by about $48 (se=21.6; p|is less than~0.05) over the same time period.(8) While the price of both types of ATVs declined in the latter time period, these estimates reflect normal depreciation as well as the effects of the consent decree; net of depreciation, the price of four-wheeled ATVs may have risen. Again, the main implication is that the price differential increased by about $186.

The results of model 2, which measure the wheels-related price differential one year after the consent decree, are consistent with those of model 1 and indicate that the price change was permanent. In fact, in model 2 the estimated increase in the wheels-related price differential in the post-consent decree time period was $223, about $37 (se=17.4; p |is less than~ 0.05) higher than the model 1 estimate. The total estimated wheels-related price differential in the early 1989 time period, including the pre-consent decree price differential, was $503, an increase of about $42 (se=14.6; p |is less than~ 0.05) over the model 1 estimate.

TABULAR DATA OMITTED

VI. DISCUSSION

Impact of the Stop-Sale

The results of the analysis show a clear change in the wheels-related price differential immediately following the consent decree: the market price of used three-wheelers declined relative to the price of used four-wheelers.

There are no formal market impediments or restrictions that can account for this change in price differential. The purpose of the stop-sale was to prevent the sale of new three-wheeled ATVs to the public; the market for used ATVs was not directly affected. The consent decree did not prohibit the sale of used three-wheelers, and used three-wheelers continued to be sold through motorcycle outlets and by individuals. Moreover, since the analysis was based on the secondhand market, the supply of ATVs was presumably price inelastic. It therefore seems likely that the decline in the price of three-wheeled ATVs relative to the price of four-wheeled ATVs resulted primarily from the implicit risk information conveyed by the stop-sale component of the consent decree and the concomitant publicity, which vividly conveyed the general message to consumers that three-wheeled ATVs were more risky than four-wheeled vehicles.(9)

The substantial (as well as statistically significant) increase in the price differential immediately following the consent decree may be somewhat surprising since publicity surrounding the project was not entirely new. The finding of a substantially higher risk of injury associated with three-wheeled ATVs was made public beginning in November 1986, following the presentation of the Consumer Product Safety Commission staff research findings.

Several points should be considered, however, in evaluating the apparently dramatic price impact following the consent decree. The consent decree represented the Commission's first formal action to discourage the use of three-wheelers. Prior to the consent decree a segment of the public may have been unaware of or uncertain about the importance or accuracy of the Commission's findings, which were, in fact, disputed by industry.(10) Moreover, although the industry did not admit fault, liability, or agreement with the Commission's legal or factual positions in accepting the consent decree, the fact that industry acquiesced in the stop-sale of three-wheeled ATVs may have given added credence, in the view of the public, to the Commission's contention that three-wheelers were less safe than four-wheelers.

In addition, the consent decree's stop-sale of new three-wheeled ATVs was itself heavily publicized in the news media. This is indicated by both the increasing quantity and changing content of newspaper reports concerning ATVs from November 1987 through January 1988, the months surrounding the signing of the consent decree. The newspaper articles were found by the Commission's clipping service and are not a random sample of newspaper accounts from around the country. However, they illustrate the changing quantity and content of newspaper accounts.

During November and December 1987, there were eighty-nine newspaper articles about ATVs: twenty-seven discussed ATVs generally; thirteen discussed the higher risks associated with three-wheeled ATVs; and forty-nine (55 percent) discussed the consent decree and stop-sale component. All of the articles about the consent decree followed a leak of the preliminary decree on about 21 December 1987 or its publication on 30 December. During January 1988, in contrast, there were 131 ATV articles: 15 discussed the risks of ATVs generally; 8 discussed the higher risk of three-wheeled ATVs; and 108 (82 percent) discussed the consent decree and stop-sale. Thus, the focus of the newspaper reports appear to have shifted from ATVs in general, prior to the consent decree, to three-wheeled ATVs and the stop-sale following the consent decree.

Market Efficiency

The increased price differential between used three- and four-wheeled ATVs following the consent decree is consistent with increased public awareness of the higher risk associated with three-wheeled ATVs and represents a possible measure of the post-consent decree implicit market valuation of the risk differential.(11) However, since information on the higher risk for three-wheeled ATVs was available to the public as early as 1986, and presumably had some effect on the pre-consent decree price differential, it probably represents a minimum market valuation.

On the other hand, assuming close substitutability of the wheel types (and the possibility that the pre-consent decree price differential was reflective of the valuation of the risk differential at that time), the post-consent decree market valuation of the risk differential could be as high as the full difference in their respective prices. Thus, the post-consent decree market valuation of the risk differential could range from an estimated $186 to $461, based on model 1, or from an estimated $223 to $503, based on model 2.

These implicit market valuation ranges are generally consistent with an earlier quantification of injury cost differentials, described in Rodgers and Rubin |1989~, which estimated directly the difference in injury costs between three- and four-wheeled ATVs. Estimates from the earlier analysis indicated that on an annual basis the expected injury costs associated with three-wheeled ATVs were approximately $120 higher than the expected injury costs of four-wheeled vehicles. These cost estimates were derived from the Consumer Product Safety Commission's Injury Cost Model which includes direct costs, such as medical costs for treatment of injury and foregone earnings during convalescence, and imputed costs for pain and suffering, based largely on jury verdicts (Technology and Economics |1980~).

According to the Motorcycle Industry Council |1990~, the industry trade association, the average product life of an ATV is about seven years. Since the data used in the regression analysis were based on 1985-87 model year ATVs, the average remaining product life of these ATVs was about 5.2 years.(12) At a 5 percent discount rate, the present value of the $120 injury cost differential over a 5.2 year period is about $560.

Thus, the implicit market valuation of the risk differential, based on the regression analysis, is generally consistent with the direct estimate of the injury cost differential, based on the Safety Commission's Injury Cost Model. The direct estimate of the injury cost differential must be viewed as approximate. Nevertheless, the general consistency of the two risk differential valuations, and consequently the fact that there is no evidence that consumers overreacted to the new risk information, suggests that implicit information in the consent decree improved the efficiency of the market for used ATVs. It does this by enabling consumers to better account for the relative risks of three- and four-wheeled vehicles.

Provision of Risk Information as a Safety Strategy

A stop-sale is, of course, an extreme method of conveying risk information to the consuming public and, as described by Oi |1973~, has welfare implications of its own. Nevertheless, the statistical results suggest that providing risk information to the public can alter consumer market behavior and may in some circumstances constitute an effective safety strategy by enabling consumers to make more informed product choices.

Although conceptually the provision of information has always had considerable economic appeal, the efficacy of information in affecting consumer behavior has frequently been questioned, as in Adler and Pittle |1984~.(13) However, there is a small but growing literature on the effects of providing new risk information on consumer behavior. In addition to the results of this analysis, Viscusi, Magat, and Huber |1986~ conducted experiments with warning labels and found that consumers responded in a manner that was broadly consistent with rational economic behavior. Behavior was affected by the level of the risk indicated, the amount of risk information, and the expected economic benefits of safety precautions. In another study, Viscusi and O'Conner |1984~ found chemical labeling in the workplace affected workers' risk perceptions and led to the expected demands for compensating wage differentials. These studies all suggest that providing new risk information can be an important ingredient in safety policy.

VII. SUMMARY

Previous studies have demonstrated a systematic relationship between well-publicized recalls (as well as product liability suits and other regulatory violations) and capital market losses for the firms involved. These studies focused on investor reaction to recalls but did not measure directly consumer market response to new risk information.

This paper examines explicitly consumer market reaction to risk information implicit in a well-publicized stop-sale of new three-wheeled ATVs. The analysis shows a substantial reduction in the relative market price of three-wheeled ATVs in the market for used ATVs, a result that is consistent with the expected impact of new risk information on the demand for these vehicles. Moreover, the results suggest that providing new risk information can improve market efficiency and can function as an important safety tool.

GREGORY B. RODGERS -- Directorate for Economic Analysis, U.S. Consumer Product Safety Commission, Washington, D.C., 20207. The views expressed are my own and do not necessarily reflect the position of the CPSC or other members of its staff. Since the article was written in my official capacity, it is in the public domain and may be freely copied or reprinted. I would like to thank Patrick S. McCarthy, Paul H. Rubin, Fay H. Dworkin, Rodney T. Smith, and the anonymous referees for helpful comments in the preparation of this article.

1. Jarrell and Peltzman |1985~ also found that automobile recalls by the National Highway Traffic Safety Administration resulted in about a 1.4 percent reduction in stock values. Although this lower reduction was generally attributed to the fact that automobile recalls are common enough to be generally anticipated, it nonetheless exceeded the direct recall costs. However, a recent study by Hoffer, Pruitt, and Reilly |1988~ reported that Jarrell and Peltzman overestimated the affects of auto recalls.

2. This study found that the negative impact on stock prices increased with the length of the Wall Street Journal article, and was greater if the article was an initial report (i.e., the article provided new information).

3. Peltzman |1981~ and Mathios and Plummer |1989~ also found that deceptive advertising cases initiated by the Federal Trade Commission resulted in significant reductions in the net worth of affected firms.

4. A consent decree is a binding court order entered into voluntarily by the parties. Although this legal settlement will be referred to as the "consent decree," it should be noted that it actually involved several consent decrees. The preliminary consent decrees, which are of primary interest in this paper because they stopped the sale of new three-wheeled ATVs, were entered into on 30 December 1987 between the Department of Justice, representing the Consumer Product Safety Commission, and (1) American Honda Motor Co., Yamaha Motor Co., Kawasaki Motor Corp., and Suzuki Motor Corp., and (2) Polaris Industries, L.P. (U.S. District Court, |1987~). The final consent decrees were negotiated and filed on 14 March 1988 (U.S. District Court, |1988~).

5. The January-April 1988 guide provided price estimates based on sales data from the September-December 1987 period, the period prior to the consent decree; the May-August 1988 guide provided price estimates based on sales data from the January-April 1988 period, the period following the consent decree; the May-August 1989 guide provided price estimates based on sales data from the January-April 1989 time period, a year following the consent decree.

6. Based on discussions with Lenny Sims, editor, National Automobile Dealers Association appraisal guides for motorcycles and ATVs.

7. Similar interaction terms were evaluated for the other independent variables. None were significant, nor did their inclusion alter the primary relationship of interest in this paper, the relationship between the stop-sale provision of the consent decree and the price differential between three- and four-wheeled ATVs.

8. The decline in the price of four-wheeled ATVs in the post-consent decree time period is given by the sum of the coefficients for the Post-Con and Four-Wh |Mathematical Expression Omitted~ Post-Con variables.

9. As pointed out by one referee, it is possible that other factors, such as concerns about parts availability for the discontinued three-wheeled ATVs or changes in consumer tastes, also had some influence on the increase in the price differential. Tastes could have been influenced if, for example, some consumers interpreted the discontinuance of three-wheelers to mean that they were somehow an outmoded style. These factors could also have affected the further increase in the price differential in the early 1989 time frame.

10. For an industry perspective on the risk data see Heiden and Staines |1986~.

11. In a related use of a hedonic regression model, Atkinson and Halvorsen |1990~ estimated the value of a statistical life with data from the automobile market.

12. This is a weighted average based on sales estimates and product survival rate estimates provided by industry.

13. Viscusi, Magat, and Huber |1986~ pointed out that the apparently disappointing results of many information campaigns may have resulted because they were basically educational efforts that provided little new knowledge.

REFERENCES

Adler, Robert S., and R. David Pittle. "Cajolery or Command: Are Education Campaigns an Adequate Substitute for Regulation?" Yale Journal of Regulation, Winter 1984, 159-94.

Atkinson, Scott E., and Robert Halvorsen. "The Valuation of Risks to Life: Evidence from the Market for Automobiles." Review of Economics and Statistics, February 1990, 133-6.

Deppa, Roy W., and Joseph A. Hauser. "ATV Lateral Stability and the Limits of Control." Paper presented at the 1989 SAE Government/Industry Meetings, SAE #891108, Washington, D.C., May 1989.

Fulcher, Debra, and Linda Reiff. "Update of All-Terrain Vehicle Deaths and Injuries." Washington, D.C.: U.S. Consumer Product Safety Commission, February 1991.

Hoffer, George E., Stephen W. Pruitt, and Robert J. Reilly. "The Impact of Product Recalls on the Wealth of Sellers: A Reexamination." Journal of Political Economy, June 1988, 663-70.

Heiden, Edward J., and John C. Staines. "An Analysis of ATV-Related Injuries: Trends and Causal Factors." Washington, D.C.: Heiden Associates, Inc., September 1986.

Jarrell, Gregg, and Sam Peltzman. "The Impact of Product Recalls on the Wealth of Sellers." Journal of Political Economy, June 1985, 512-36.

Market Facts, Inc. "All-Terrain Vehicle Market Study." Contract CPSC-P-87-1280. Washington, D.C.: Market Facts, Inc., 1988.

Mathios, Alan, and Mark Plummer. "The Regulation of Advertising by the Federal Trade Commission: Capital Market Effects." Research in Law and Economics 12, 1989, 77-93.

Motorcycle Industry Council, Inc. 1990 Motorcycle Statistical Annual. Irvine: Motorcycle Industry Council, Inc., 1990.

National Automobile Dealers Association. Motorcycle/Snowmobile/ATV/Personal Watercraft Appraisal Guide. Costa Mesa: National Automobile Dealers Association, various issues 1988-89.

Oi, Walter Y. "The Economics of Product Safety." Bell Journal of Economics, Spring 1973, 3-22.

Peltzman, Sam. "The Effects of FTC Advertising Regulation." Journal of Law and Economics, December 1981, 403-48.

Rodgers, Gregory B., and Paul H. Rubin. "Cost-Benefit Analysis of All-Terrain Vehicles at the CPSC." Risk Analysis, March 1989, 63-9.

Rosen, Sherman. "Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition." Journal of Political Economy, February 1974, 35-55.

Rubin, Paul H., R. Dennis Murphy, and Gregg Jarrell. "Risky Products, Risky Stocks." Regulation, no. 1, 1988, 35-9.

Rubinfeld, Daniel L., and Gregory B. Rodgers. "Evaluating the Injury Risk Associated with All-Terrain Vehicles: An Application of Bayes' Rule." Journal of Risk and Uncertainty, May 1992, 145-58.

Technology and Economics, Inc. "The Consumer Product Safety Commission Injury Cost Model." Contract CPSC-C-78-0091. Cambridge: Technology and Economics, Inc., July 1980.

U.S. Consumer Product Safety Commission. "Report of the All-Terrain Vehicle Task Force to the CPSC." Washington, D.C.: Consumer Product Safety Commission, September 1986.

U.S. District Court for the District of Columbia. "United States of America vs. American Honda Motor Co., Inc. et al: Preliminary Consent Decree." Civil action No. 87-3525. Washington, D.C., 30 December 1987.

-------. "United States of America vs. American Honda Motor Co., Inc. et al: Final Consent Decree." Civil action No. 87-3525. Washington, D.C., 14 March 1988.

Viscusi, W. Kip, and Charles O'Connor. "Adaptive Responses to Chemical Labeling: Are Workers Bayesian Decision Makers?" American Economic Review, December 1984, 942-56.

Viscusi, W. Kip and Joni Hersch. "The Market Response to Product Safety Litigation." Journal of Regulatory Economics, September 1990, 215-30.

Viscusi, W. Kip, Wesley A. Magat, and Joel Huber. "Informational Regulation of Consumer Health Risks: An Empirical Evaluation of Hazard Warnings." Rand Journal of Economics, Autumn 1986, 351-65.
COPYRIGHT 1993 Western Economic Association International
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Rodgers, Gregory B.
Publication:Economic Inquiry
Date:Jan 1, 1993
Words:5087
Previous Article:Does industrial diversity always reduce unemployment? Evidence from the great depression and after.
Next Article:Uncertain effects of money and the link between the inflation rate and inflation uncertainty.
Topics:

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters