This article analyzes a comprehensive sample of over 350 chemicals tested for carcinogenicity to assess the determinants of the probability of regulation. Controlling for differences in the risk potency and noncancer risks, synthetic chemicals have a significantly higher probability of regulation overall: this is due to the greater likelihood of U.S. Food and Drug Administration (FDA) regulation. Measures of risk potency increase the probability of regulation by the U.S. Environmental Protection Agency (EPA), have a somewhat weaker positive effect on regulation by the U.S. Occupational Safety and Health Administration (OSHA), and decrease the likelihood of regulation by the FDA. The overall regulatory pattern is one in which the FDA targets synthetic chemicals and chemicals that pose relatively minor cancer risk. The EPA particularly performed more sensibly than many critics have suggested.
1 Much of this fear of new chemicals stems in part from society's overreaction to new risks that are introduced. This phenomenon is designated the "reference risk" effect by Viscusi, Magat, and Huber [1987] and the "status quo bias" by Samuelson and Zeckhauser [1988]. For a review of risk regulation policy trade-offs more generally, see Morrall [1986].