Skip to main content
Find a Lawyer
Michael C. Dorf

Why the Supreme Court Decision Upholding Cost-Benefit Analysis Under the Clean Water Act Should Not be Used to Discredit Best-Practice Standards


Monday, April 6, 2009

Last week, in Entergy Corp. v. Riverkeeper Inc., the U.S. Supreme Court rejected a challenge to the Environmental Protection Agency's use of cost-benefit analysis in regulating water pollution by power plants. Writing for the Court, Justice Scalia said that the EPA acted reasonably in weighing the costs and benefits of various technologies when it promulgated regulations under Section 316(b) of the Clean Water Act. That law requires that power plants employ "the best technology available for minimizing [their] adverse environmental impact."

Most of the back-and-forth among the Justices in Entergy Corp. concerned issues of statutory construction and administrative law. Yet, as I shall explain in this column, Entergy Corp. is also a very significant case for what it may portend for so-called "best-practice standards" (a term which I shall define and explain below).

The majority in Entergy Corp. appears to treat this form of regulation as a license for regulatory agencies to promulgate lax rules and standards for industry. Yet, properly implemented, best-practice standards actually hold out the promise of stricter and more effective regulation. As the United States and the world enter a period of renewed belief in the need for regulation in a variety of domains, it would be unfortunate if best-practice standards were unfairly discredited as a cover for lax regulation.

The Statute and Regulations at Issue

Coal-fired (and other) power plants generate heat as a by-product of their activity. To prevent overheating, such plants use enormous volumes of cooling water, which they typically obtain from nearby natural bodies of water. Doing so is hazardous to the fish and other aquatic life forms that live in these bodies of water, and so Congress has, over the years, required that power plants take various measures to mitigate the harm they thereby cause.

Section 316(b) of the Clean Water Act is one such measure. For new power plants, the EPA regulations require the use of any technology at least as effective as a "closed-cycle cooling system," which re-uses the same water, thereby substantially reducing the amount of water used, and thus diminishing the impact of new power plants on fish and other aquatic life. However, because it is much cheaper to build a closed-cycle cooling system into the original design of a power plant than it is to retrofit an existing plant with such a cooling system, the EPA regulations exempt old plants from this requirement.

Under the challenged EPA regulations at issue in Entergy Corp., an old power plant will not be subject to the closed-cycle-cooling-system-or-its-equivalent requirement if it can show either that the cost of retrofitting would be significantly greater than the EPA assumed in setting the standard for new plants, or that compliance costs "would be significantly greater than the benefits of complying with the applicable performance standards." Such a showing then subjects the existing plant to a requirement to use alternative technologies that produce results "as close as practicable to" the results of closed-cycle cooling.

The key weasel-word there is "practicable." In practice, the EPA exemption for old plants means that while new power plants must use technology that reduces environmental impact by about 98 percent, old plants are subject to standards that require reductions that are only in the range of 80 to 95 percent, or in some instances, even lower.

The Statutory Construction and Administrative Law Questions in Entergy Corp.

The dissenters in Entergy Corp. thought that the EPA exemption for existing power plants was inconsistent with Section 316(b) of the Clean Water Act because that section nowhere authorizes the EPA to conduct cost-benefit analysis. That omission, the dissenters thought, was hardly accidental, because other provisions of environmental laws that Congress passed at the same time as Section 316(b), expressly authorize cost-benefit analysis. Thus, the dissenters argued, Congress did not intend the agency to balance costs and benefits.

And for good reason, the dissenters said. Cost-benefit analysis tends to understate the value of regulation because costs (such as how much money a plant must spend on harm-mitigating technology) are easy to calculate, but benefits are not. Here, for example, the dissenters noted that the EPA monetized the benefit of the regulation by placing a dollar value on the fish that would be caught for food and sport, even though those fish only amount to two percent of all the fish and shellfish that are killed by power plants. Even disregarding the value of those fish and shellfish to the creatures themselves (as Congress almost certainly did), such non-commercial species have value as part of the ecosystem as a whole. Thus, the dissenters concluded that Congress deliberately chose not to authorize the EPA to balance away the benefits of the best-available-technology rule.

The majority disagreed. Silence, Justice Scalia said for the Court, is just that: silence. The other statutory provisions that authorized cost-benefit analysis also authorized agency consideration of other factors not addressed by Section 316(b), Justice Scalia noted, and so by the if-Congress-said-it-there-but-not-here-Congress-meant-to-forbid-it logic of the dissent, one would have the absurd result that the EPA could not consider any factors in promulgating regulations under Section 316(b). Thus, the majority thought that the other statutory provisions cited by the dissent were not part of the relevant context for understanding Section 316(b) itself.

Accordingly, although the majority agreed with the dissenters that the EPA was not required to conduct cost-benefit analysis, the majority also concluded that neither was the EPA forbidden from doing so. Where a statute conferring regulatory authority on an agency does not address some question, longstanding administrative law principles state that the courts should defer to a reasonable agency regulation that does address the question. And, the Court concluded, the cost-benefit analysis performed by the EPA was reasonable.

Indeed, how could it be otherwise? Justice Scalia noted that even the environmental groups challenging the EPA regulations acknowledge that the EPA did not have to require power plants to "spend billions to save one more fish or plankton." Thus, at some point, the costs of further marginal mitigation of environmental harm outweigh the benefits, and the only real question is where to draw that line. The EPA, the majority said, could have answered that question any number of ways, but the only issue for the Court was whether the line the EPA drew was reasonable.

Needless to say, that was not the end of the debate. The dissent, in turn, thought that Section 316(b) authorized only a very narrow form of cost-benefit analysis for cases like the save-one-fish example, but not the all-things-considered cost-benefit analysis that the majority accepted. And Justice Breyer, in a separate opinion, split the difference, allowing that the EPA might be able to justify the form of cost-benefit analysis it used but also reasoning that it had not done so in the record before the Court.

The Future of Best-Practice Standards

In no small part, the Court's division in Entergy Corp. manifests differences in ideological preferences among the Justices. The conservatives in the majority are skeptical of the value of environmental regulation and thus eager to uphold a Bush-era EPA regulation that the power industry favors, while the liberals in dissent want stricter enforcement of the environmental laws.

Reading Entergy Corp. and studying the history of lax regulatory enforcement under the last Administration, environmental groups and others who seek tighter regulation of industry may thus be tempted to eschew best-practice standards, and lobby Congress to pass hard-edged rules. Yet that would be a mistake.

To understand why, we need to distinguish among design standards, performance standards, and best-practice standards. A design standard is simply a requirement that a regulated actor use a specific technology. A requirement that coal-burning power plants have smokestacks at least 100 feet high is an example.

Design standards have the virtue of clarity but the vice of rigidity. They tend to lock in the limitations of the technology available when they are adopted. Accordingly, legislators and regulators sometimes adopt performance standards, which require that a regulated actor achieve some specific objective, but leave the choice of means to the regulated actors. A requirement that drinking water (supplied by water companies or municipalities) contain no more than fifteen parts per billion of lead is an example of a performance standard. It does not require that the water supplier use any particular water purification technique, and it thereby encourages water suppliers to develop cheaper methods of complying.

Still, conventional performance standards share an important defect with design standards: Both are static. In setting the level of acceptable pollution or other harm, the regulator uses an existing baseline. Given existing technology, it may not be feasible to reduce the lead content of drinking water to below fifteen parts per billion. Future technology could perhaps reduce lead levels substantially without substantially greater cost, but once the EPA sets the performance standard at fifteen, regulated actors do not have an incentive to reduce lead content further. This is especially true where the regulated actors are not utilities or municipalities subject to local political pressure but multi-national corporations.

Accordingly, in recent years government has often turned to best-practice standards. Section 316(b) of the Clean Water Act is an example. It does not set any particular target but instead tells regulated actors that they must adopt the best methods available. Such best-practice standards create incentives for regulated actors to compete on regulatory compliance. A power company that develops and implements some new technology that achieves better performance along the relevant dimension (here, fewer fish and other aquatic animals killed) can thus force its competitors to play catch-up. And independent inventors who develop better technology have an automatic market for their products--because failure to adopt this new, best available technology constitutes violation of the best-practice standard.

Best-practice standards thus avoid the lock-in drawbacks of design standards and fixed performance standards. Pro-regulation liberals should therefore favor them, right?

In fact, some do, but many do not, precisely because they worry that an Administration with ties to industry will take a relaxed view of what constitutes the best available technology. As the EPA action at issue in Entergy Corp. reveals, that is no mere hypothetical problem.

Yet neither is it a reason to throw out the baby with the cooling water. For one thing, an Administration that takes its regulatory obligations seriously (as the Obama Administration appears to) can achieve much more with best-practice standards than with fixed design standards or fixed performance standards that Congress revisits only infrequently. A vigilant Administration can pay close attention to developments in the field and continually update the specific obligations entailed by a best-practice standard (giving regulated actors a reasonable period to comply, of course).

But if the worry is that pro-regulation Administrations will not always be in power, then the best solution would appear to be to layer best-practice standards on top of performance standards. Congress can both clearly instruct the agency to mandate at least a particular level of performance, and make clear that this performance standard is a floor by also mandating that the agency apply a best-practice standard if the best available technology can do better than the Congressionally-set floor. That way, even if the courts allow a regulation-hostile Administration to water down a best-practice standard, there will still be a hard-edged performance standard providing a backstop. Nothing in Entergy Corp. forbids this approach.

Michael C. Dorf, a FindLaw columnist is the Robert S. Stevens Professor of Law at Cornell University. He is the author of No Litmus Test: Law Versus Politics in the Twenty-First Century and he blogs at

Was this helpful?

Copied to clipboard