Ethyl Corp. v. Environmental Protection Agency

The ordinary meaning of “endanger” is to expose someone to threatened harm, not to prove that injury has already occurred. A statute aimed at danger is therefore precautionary: its purpose is to permit action before the anticipated injury materializes.

The Clean Air Act’s structure reinforced this reading. Congress used different language in § 108, which requires an actual “adverse effect” before a pollutant is listed for national ambient-air regulation. By using “will endanger” in § 211, Congress authorized EPA to act at an earlier, preventive stage.

Endangerment depends on both the probability and the severity of harm. A comparatively lower probability can justify regulation when the feared harm is severe, while a higher probability may justify action against a less severe harm. EPA’s finding of a significant risk of widespread lead poisoning, coupled with its phased rather than immediate prohibition, fell comfortably within that standard.

Congress did not require the formal factual findings that the House version of the bill would have demanded. The enacted statute instead required EPA to consider all relevant medical and scientific information. That choice recognized that environmental-health regulation often requires judgments about risk, uncertainty, and developing scientific knowledge.

EPA could not regulate on mere hunches. But it could use scientific evidence, expert judgment, theoretical projections, preliminary data, and reasonable inferences to make a rational risk assessment where definitive proof would be impossible without waiting for the threatened harm to occur.

People absorb lead from multiple sources, including food, water, paint, dust, and air. Once lead enters the bloodstream, its source does not matter biologically; the health risk depends on the aggregate body burden.

Petitioners’ proposed requirement that automotive lead be dangerous “in and of itself” would make § 211 largely ineffective. No individual source of lead, other than perhaps paint in some cases, may independently produce a harmful level, even though multiple sources together plainly can.

The statute did not say that an additive’s emissions must be the sole cause of endangerment. EPA could therefore determine that gasoline lead significantly raised total exposure and regulate it because it was a substantial source that was comparatively amenable to reduction.

EPA also satisfied its duty to consider alternatives under § 202. That provision concerns emissions standards for new vehicles and could not effectively control lead emissions from the existing fleet of vehicles using leaded gasoline.

Section 211(c)(2)(C) expressly requires a finding about substitute fuels before EPA prohibits a fuel or additive. The court did not definitively decide whether the same requirement applied to a mere control, because EPA had in fact addressed the issue.

EPA explained that reducing lead would increase use of aromatic hydrocarbons, which could contribute to emissions of some carcinogenic compounds. But it reasonably concluded that the substitution would not create an equal or greater danger because automobile controls would continue to reduce those emissions overall, and automobiles represented only a limited share of the relevant ambient pollutants.

This was a predictive and policy-laden judgment rather than an adjudicative fact susceptible to conclusive proof. EPA’s explanation was sufficient to show the fundamental rationality of its conclusion.

Under arbitrary-and-capricious review, the court had to study the complex record carefully enough to determine whether EPA considered the relevant factors and acted rationally. But the court could not substitute its own scientific judgment for that of the expert agency or demand a preponderance, much less certainty, from informal rulemaking.

EPA reasonably treated blood-lead levels above 40 micrograms per 100 grams of blood as warning signs of undue absorption and potential harm. The agency acknowledged that 40 micrograms was not a bright line between health and disease, but could prudently use it as a preventive benchmark.

The evidence supported EPA’s conclusion that some adults and many children had elevated blood-lead levels. EPA reasonably gave weight to studies of persons such as traffic-exposed workers, drivers, and urban residents, while recognizing the limitations of broader population studies whose results were obscured by differences in diet, climate, and other lead sources.

EPA had a rational basis for finding that airborne lead made a significant contribution to body lead. Its conclusion rested on mutually reinforcing theoretical calculations, epidemiological evidence showing urban-suburban and proximity-related differences, and clinical studies showing that controlled exposure to airborne lead increased blood-lead levels.

EPA also reasonably relied on the dustfall hypothesis as additional support for the regulations. Evidence showed high lead concentrations in urban dust, widespread pica among preschool children, absorption of ingested lead, and correlations between children’s lead levels and proximity to traffic or lead-contaminated environments. Lead paint remained the principal source of acute childhood poisoning, but that fact did not negate the additive risk from automotive lead dust.

The court emphasized that no single study had to be conclusive. In a frontier area of environmental medicine, the cumulative force of suggestive studies, evaluated through an expert agency’s reasoned assessment of risk, could supply the rational basis required by the APA.

EPA had provided multiple notice-and-comment opportunities over nearly three years. When it materially revised its analysis after earlier comments, it reproposed the regulations and opened a further comment period. The final regulations relied on the same basic health theory presented in the reproposal.

The third health document primarily updated the scientific record and responded to comments. The agency was not required to start a new formal comment cycle every time it received new studies, reconsidered criticisms, or adopted suggestions made during rulemaking.

The key studies challenged by petitioners had been placed in the public file, provided to Ethyl through Freedom of Information Act disclosures, or otherwise made available before the final decision. Several had also been specifically discussed or criticized in submissions to EPA.

The final switch from leaded-pool averaging to total-pool averaging was within the range of options identified in the reproposal and responded to comments from a majority of refiners. EPA therefore complied with both the letter and the practical purpose of APA notice-and-comment rulemaking.