Every day, Americans are exposed to hundreds of chemicals in the air we breathe, the water we drink, and the products we use. The vast majority of these chemicals have never been tested for safety. Many have been shown to cause serious health harms, ranging from cancer to autoimmune illness to IQ loss. They also have disproportionate effects on some of the most vulnerable populations in our society, such as children, minorities, and industrial workers.
The law that is supposed to protect Americans from dangerous chemical exposures – the Toxic Substances Control Act (TSCA) – was long considered a dead letter after EPA’s failure to ban asbestos, an extremely hazardous carcinogen. The agency issued a ban in 1989, but it was promptly struck down by the U.S. Court of Appeals for the Fifth Circuit in Corrosion Proof Fittings v. EPA. Following the Fifth Circuit decision in 1991, the agency never again sought to exercise its authority under TSCA to prohibit the use of a chemical already on the market.
For the next several decades, EPA officials, environmental groups, and members of Congress debated the reasons that EPA’s asbestos regulation did not survive judicial review and whether TSCA should be amended in response. According to certain agency staff and some environmental organizations, TSCA’s requirement that EPA perform a cost-benefit analysis to justify the asbestos ban made it impossible for the regulation to stand up to judicial scrutiny. As a result, when Congress had a once in a generation opportunity to amend TSCA in 2016, Democrats made removal of cost-benefit analysis from determinations about whether a substance poses an unreasonable risk their top priority in negotiations with Republicans. To earn Republican support for changing the law, Democrats agreed to include sweeping preemption of state toxic chemical controls, which had been a major goal of the chemical industry.
Yet a historical examination of the asbestos episode reveals that the accepted story of EPA’s failure is wrong. Archival documents, many of which have never been viewed by those outside EPA, demonstrate that it was not cost-benefit considerations that doomed the asbestos regulation. Instead, disagreements with the Office of Management and Budget over whether EPA or OSHA should address asbestos and EPA’s refusal to fully quantify harms and monetize benefits were largely responsible for the problems with issuing the regulation and marshalling enough support to withstand judicial review. In fact, EPA could have justified the ban on cost-benefit grounds if the agency had quantified and monetized the benefits using information available to it.
This paper makes several arguments based on how science and economics were used – and misused – to justify the asbestos ban. The first is that Democrats in Congress struck the wrong bargain with Republicans in the 2016 amendments to TSCA and failed to fix the underlying problems with toxics regulation. As a result, the same challenges that plagued the asbestos ban have continued to frustrate EPA efforts since Congress amended the law. Other methodological approaches for deciding when to regulate require similar informational inputs, expertise, and value judgments. The paper therefore suggests that environmental scholarship and advocacy should focus greater attention on how underlying analytical assumptions shape all methods for deciding whether or not to regulate. Finally, and more poignantly, this paper serves as an example of why accurate, well-sourced history is essential for understanding how administrative agencies function and what lessons we should draw from their successes and failures. Had Democrats been aware of what actually transpired within EPA and the executive branch, they would have been much better equipped to enact meaningful TSCA reform.
Environmental Law | Environmental Policy | Law and Economics | Legal History
Date of this Version
Working Paper Citation
Rothschild, Rachel, "Unreasonable Risk: The Failure to Ban Asbestos and the Future of Toxic Substances Regulation" (2022). Law & Economics Working Papers. 236.