Portland Cement Association v. Ruckelshaus

The court declined to decide whether EPA, as an environmental-protection agency, is broadly exempt from NEPA. NEPA's text and legislative history did not conclusively resolve that broader question, and later congressional action supplied ambiguous rather than decisive evidence of the original Congress's intent.

For § 111 specifically, the Clean Air Act required EPA to act on demanding statutory deadlines. Requiring a detailed NEPA statement, interagency circulation, and evaluation of comments could frustrate Congress's instruction that EPA quickly identify source categories and issue performance standards.

Section 111 also requires much of the substantive environmental inquiry that NEPA seeks. EPA must select the best demonstrated emission-reduction system while taking account of cost, which includes counterproductive environmental effects as well as economic burdens. EPA must therefore explain the environmental considerations supporting its proposed rule, make supporting materials available, receive comments, and provide a record subject to direct judicial review.

This functional-equivalence approach preserves public, agency, congressional, and judicial scrutiny without creating a separate NEPA process that would duplicate the § 111 rulemaking and invite incremental delay. On remand, however, EPA had to address environmental concerns raised by the manufacturers, including whether stricter air controls could produce increased water pollution or other adverse effects.

Section 111 requires EPA to take account of the costs of emission reduction, but it does not compel a numerical comparison of every control cost against a quantified improvement in ambient air quality. Such quantification is often difficult or impossible, and imposing it as a precondition to action would conflict with the statute's expedited rulemaking timetable.

EPA had sufficiently considered general industry costs at this stage. Its background materials and a financial-impact study estimated capital and operating costs of pollution-control equipment and concluded that the added costs could generally be passed through without materially affecting competition from construction substitutes. The manufacturers did not directly undermine those findings.

The record nonetheless raised a serious question whether the standards effectively excluded wet-process plants or the less expensive electrostatic precipitator technology. Because a remand was independently required, EPA was directed to consider evidence that the standards could unduly restrict cement supply, particular cement qualities, regional availability, or lower-cost production.

EPA did not unlawfully discriminate by setting a stricter particulate limit for cement plants than for steam-generating power plants and incinerators. Section 111 calls for industry-specific standards based on the best adequately demonstrated system available for the particular source category, not a uniform emissions number for unrelated industries.

Comparisons with industries producing substitute products can be relevant to economic cost. But EPA had considered competition from construction substitutes, and no claim before the court showed that EPA had ignored or irrationally assessed that competitive effect.

New-source performance standards regulate the future rather than merely codify the performance of existing facilities. Therefore, the phrase 'adequately demonstrated' does not require proof that every operating cement plant can already comply with the standard.

Congress did not demand technology in routine current use, nor did it authorize purely theoretical or experimental controls. EPA may extrapolate from existing technology to what will be available for installation in new plants, but the scope of that projection depends on the available lead time and must remain within the bounds of reasonableness rather than 'crystal ball' speculation.

Because the challenged standards applied immediately to newly constructed or modified plants, EPA's latitude for technological projection was narrower than it would be for a standard taking effect years later. Any prediction therefore required reliable assumptions, meaningful evidence, and a reasoned response to significant objections.

Meaningful notice and comment requires regulated parties to know the material basis for a proposed rule. EPA repeatedly said that its standard rested on source testing, yet it did not timely disclose the locations, detailed methods, and underlying data for the Dragon dry-process and Oregon wet-process kiln tests.

The court recognized that the Clean Air Act's short deadlines might make complete disclosure difficult at the proposal stage. But material information should be disclosed as it becomes available, and affected parties must be allowed to comment, even after issuance of the rule when necessary. Rulemaking cannot fairly rest on data known in critical detail only to the agency.

The court's earlier remand allowed the manufacturers to submit engineer Ralph Striker's detailed analysis of alleged errors in the Dragon test. EPA merely placed his analysis in the record and offered no written response. Given the court's prior determination that the submission had at least potential significance, EPA had a duty to address it.

Striker asserted, using chemical and flow-rate calculations, that EPA's reported gas-volume and kiln-feed figures could not both be correct and that the test's reported emissions of 0.202 pounds per ton might actually have been approximately 0.404 pounds per ton. The court was not equipped to decide that technical dispute itself, but it required EPA to explain whether and why the criticism was unsound or immaterial.

EPA also had to address other potentially material concerns: the limited number of successful tests; the unsupported extrapolation from a dry-process baghouse test to wet-process plants; sampling periods far shorter than the regulatory two-hour average; whether the Dragon kiln was operating at maximum production; and possible deviations from normal measurement procedures.

EPA's appellate counsel could not cure the defective explanation by disclaiming reliance on tests and literature that EPA itself had invoked when proposing and promulgating the rule. Courts review the agency's stated reasons, not post hoc rationalizations offered by counsel. If EPA relied on technical literature on remand, it had to identify the specific findings it considered significant and permit informed comment.

The Oregon tests also required further consideration because they were conducted only during normal operations and excluded interruptions caused by baghouse pressure problems, leaking bags, and dust-removal-equipment failures. EPA could consider its proposed malfunction, startup, and shutdown provisions, but it could not simply defer the statutory question of available technology to unguided enforcement discretion.

Opacity observation may be a useful, inexpensive screening or enforcement tool. But once EPA makes opacity itself an enforceable performance standard, rather than merely a trigger for further testing, the measure must satisfy § 111 and be sufficiently objective to give manufacturers fair notice of compliance.

The manufacturers supplied evidence that even trained smoke inspectors varied substantially when estimating plume opacity. Although that evidence did not conclusively establish that all opacity measurements were invalid, it raised a serious issue about the reliability of measurements at the especially low 10% level selected by EPA.

EPA responded primarily by noting that states commonly use opacity standards, but the cited state standards generally allowed at least 20% opacity. That response did not establish that observers could reliably distinguish a compliant 10% plume from a violation. EPA therefore had to develop and explain a record showing reasonable measurement accuracy.

The court emphasized that EPA has primary responsibility and technical expertise in setting pollution standards. Judicial review is not a license to choose a different standard simply because judges might evaluate the technical evidence differently.

But review cannot be meaningless. The court must determine whether EPA considered the relevant factors, exercised reasoned discretion, and made no clear error of judgment. That task requires enough disclosure and explanation to permit courts and affected parties to test the agency's reasoning.

EPA's environmental objective was laudable, but the statute requires standards based on achievable, adequately demonstrated technology. Because the record lacked timely disclosure, a meaningful opportunity for comment, and responses to several material technical challenges, the court remanded for further proceedings consistent with its opinion rather than affirming the rules on the existing record.