The AEI-Brookings Joint Center for Regulatory Studies held a conference on November 9, 1999, to discuss a recent study by James T. Hamilton and W. Kip Viscusi on the Environmental Protection Agency's Superfund program. The study analyzes the benefits and costs of the Superfund program and recommends reforms to increase its efficiency and effectiveness.

Robert W. Hahn, the director of the AEI-Brookings Joint Center for Regulatory Studies, opened the conference. James T. Hamilton of Duke University presented the results of the study, which John A. Hird of the University of Massachusetts and Don R. Clay, a former assistant EPA administrator currently affiliated with Koch Industries, Inc., critiqued.

Robert W. Hahn, Joint Center

In 1980 the Environmental Protection Agency launched its massive Superfund program, authorized by the Comprehensive Environmental Response, Compensation, and Liability Act to provide federal money for the cleanup of contaminated hazardous waste sites. Both scholars and journalists have been widely criticized the program because it is not cost-effective; some experts estimate that 90 percent of insurance expenditures and 20 percent of corporate expenditures are legal fees or other transaction costs unrelated to clean-up activities. Mr. Hahn noted that widespread agreement across the political spectrum exists that "Superfund is broken and needs to be fixed."

According to Mr. Hahn, a steady stream of work documents the problems with the Superfund program. The study by Mr. Hamilton and Mr. Viscusi offers the most comprehensive analysis of the program to date and proposes some provocative solutions to the program's problems.

James T. Hamilton, Duke University

Mr. Hamilton first described the study, which used original data on the cleanup of 130 hazardous waste sites to examine the extent that political decisions and quantitative risk assessments influence cleanup and remediation decisions. The study shows that the most effective 5 percent of cleanup expenditures eliminate over 99 percent of the risk. The mean cost per cancer case averted at the sample 130 sites is $11.7 billion, and the median cost is $418 million. Hamilton also pointed out that the cost of cleanup is high because regulators chose target risk levels based on political variables and risk perception biases rather than on defensible risk assessments.

Mr. Hamilton asserted that the Superfund program is an extremely inefficient way to reduce risks at designated hazardous waste sites for two reasons. First, the EPA exaggerates the risk at Superfund sites. Mr. Hamilton explained that the EPA uses several conservative risk scenarios and parameter assumptions to assess the risk, a practiced that compounds the conservatism of the estimates. The EPA, for example, uses upper-bound estimates for the length of time people are exposed to a risk, the mean concentration of the chemical, and the maximum concentration level. Second, political factors influence the EPA's cleanup decisions. Mr. Hamilton noted that the EPA spends more to avert cancer cases in counties with high voter turnouts and states with more environmentalists. Moreover, high voter turnout is most strongly associated with low-risk sites.

Mr. Hamilton also argued that minority communities are exposed to higher risks than white communities. According to Mr. Hamilton, the average number of polluted sites within one mile of individuals living around the Superfund sites is 11.4 for minority residents and 6.3 for white residents. Also, the higher the percentage of minorities living within one mile of polluted sites, the less the EPA spends to avoid a case of cancer.

Mr. Hamilton concluded by recommending four ways to improve the efficiency and increase the fairness of the Superfund program. First, the EPA should provide the public with accurate information about the characteristics of the population exposed to the risk, the use of society's resources, and the agency's response to risk at different sites. Second, the EPA should conduct a formal benefit-cost analysis that highlights differences between current and future risks, as well as exposed populations at different sites, instead of focusing on only individual risks. Third, the EPA should conduct sensitivity analyses of risk assessment parameters. Finally, the agency should include the cost per cancer case averted in the risk assessment.

John A. Hird, University of Massachusetts

Mr. Hird expressed his concern that, despite widespread agreement regarding the inefficiency of the Superfund program, little has been done in the past twenty years to reform it. He argued that reform efforts have languished primarily because the general public's perception of the health risk from toxic waste sites usually differs from that of experts. Experts are concerned with how to reduce risks effectively. The general public, on the other hand, is often concerned with a "diverse and sometimes ambiguous set of concerns about the nature and distribution of risk." Those concerns include the source of risks, the concentration of risks, and whether risks are natural or man-made. To reconcile the differences between experts and the general public, Mr. Hird suggested the formation of new institutions that would involve the public in risk management decisions. He contended that the discussion of risk policy should be broadened to incorporate the public's legitimate concerns and to recognize the opportunity costs of all public and private decisions. Otherwise, even excellent analyses like the Hamilton-Viscusi study could fall on deaf legislative ears in the face of popular opposition.

Don R. Clay, Koch Industries, Inc.

Mr. Clay generally agreed with Hamilton and Viscusi's findings on the high costs and ineffectiveness of the Superfund program. He noted, however, that reducing cancer risk was never the program's main objective. In addition, he criticized the Hamilton-Viscusi study for ignoring the successes of the program, such as restoring property values and deterring careless disposal of hazardous waste. Mr. Clay also noted that since 1991, the last year of data used in the Hamilton-Viscusi study, the EPA has addressed many of the cost inefficiencies reported by the authors. The agency has established a review board to evaluate periodically the effectiveness of the program and has also stopped adding new Superfund sites. Finally, Mr. Clay asserted that Congress is unlikely to reauthorize the Superfund program, so it will eventually focus on emergency response and long-term site maintenance. He suggested that agencies apply the Hamilton-Viscusi recommendations for the Superfund program to other federal programs, which have equally or more complex risk assessment procedures than Superfund.