Redesigning CERCLA Liability

An Analysis of the Issues

Executive Summary

Superfund, the federal hazardous substance cleanup program, is one of the most hotly debated pieces of environmental legislation. It empowers the Environmental Protection Agency to identify high-priority contaminated sites and clean them up with the financial help of those responsible for the contamination. Since its enactment in 1980, it has been criticized for:

  • Relying on inaccurate risk assessments that overstate environmental risk;
  • Requiring more cleanup than may be reasonable, leading to costs out of proportion to benefits;
  • Placing responsibility for cleanup on innocent parties, unfairly distributing cleanup costs, and imposing large liability where there is no fault;
  • And, as a result of all these, encouraging long, bitter, and expensive litigation over what should be a Superfund site, who should pay for its cleanup, and how much they should pay.

Environmental scientists and policymakers (including the Clinton Administration) now universally recognize that Superfund reform is in order. Superfund problems were all debated during reauthorization hearings in the 103rd Congress. The cost side of the problem—hitting people, as it did, in the pocketbook—attracted most of the attention. Three types of liability that exist under Superfund were considered in reform bills:

  • Strict liability, under which a party is responsible for cleanup, even though it may not have acted negligently;
  • Joint and several liability, under which a party can be liable for the whole cleanup, even though it may have only been responsible for a small part of the overall contamination; and,
  • Retroactive liability, under which a party is responsible for cleanup, even though the contamination may have happened long before Superfund took effect.

Many people believe that these forms of liability are unfair for several reasons. First, nearly everyone who has touched a hazardous substance between its generation and its disposal can be held responsible for its cleanup. Second, the standards of proof for the EPA are less stringent than those required under common law. Third, the burden of proof is shifted to potentially responsible parties to show that they are not liable. According to others, however, draconian liability is necessary so that people take proper care with hazardous substances.

Superfund reform efforts in the 103rd Congress failed, but it is virtually certain that industry and environmental groups will tackle the issue again in the 104th Congress. The question of who should pay (and how much) for cleaning up hazardous substances is also important on the state level, since many states have adopted their own hazardous substance cleanup programs to supplement the federal program.

This report examines ideas for reforming Superfund liability. It looks at the three forms of liability and asks how efficiently alternative proposals reduce risks, allocate costs, and allocate risks among potentially responsible parties. A good liability rule should encourage people to prevent pollution and clean it up when it happens. However, it should keep in mind that there is such a thing as “too much cleanup” or “too much prevention.” Liability rules should make people bear the consequences of their actions by paying to clean up (to a reasonable level) the contamination that they've caused. However, to punish people for unforeseeable, non-negligent contamination that occurred long ago would have no deterrent effects. In addition, liability rules should be structured so that the allocation of costs is fair and risks can be spread for those parties who cannot easily absorb large risks. In some cases, though, a tradeoff among these goals may be necessary.

This report concludes that:

(1) In cases of retroactive application:

  • Strict liability should be eliminated; and,
  • A negligence-based rule should be adopted for both care and mitigation.

(2) In cases of prospective application:

  • Strict liability should be retained;
  • Joint and several liability should be eliminated, except when the parties have a contractual relationship (as do generators and site owners, or buyers and sellers of land).

This Study's Materials





;