Supreme Court Misses the Reality of EPA Clean Water Act Permits

When the US Supreme Court invalidated the Environmental Protection Agency’s use of “end-result” requirements in Clean Water Act permits on March 4, the court failed to engage with how permitting actually works. The decision imposes impractical burdens on agencies and could lead to a more adversarial permit process and less effective permits.

The court reversed the US Court of Appeals for the Ninth Circuit, and held in San Francisco v. EPA the Clean Water Act doesn’t authorize “end-result” permit requirements broadly prohibiting discharges that contribute to a waterbody failing to meet water quality standards. Instead, permits must specify actions permittees must take to avoid violating those standards, the court held.

San Francisco holds National Pollutant Discharge Elimination System permits covering the operation of its combined sewer system and sewage treatment plants. The system occasionally discharges into the Pacific Ocean during heavy rain, when the combined volume of both stormwater and sewage is more than it can transport and treat. NPDES permits for such combined sewer systems typically include a variety of conditions, including pollutant-based numeric limits, termed “effluent limitations,” and other conditions that require specific practices regarding operations, monitoring, and recordkeeping, referred to generally as “narrative” requirements. Some permits also include a specific narrative requirement prohibiting discharges that contribute to a violation of applicable water quality standards in receiving waters.

It was this subset of permit conditions at issue in San Francisco and which the 5-4 majority led by Justice Samuel Alito labeled “end-result” requirements. While San Francisco’s primary argument was that all “narrative” permit conditions were unauthorized by the CWA, the court more narrowly invalidated only conditions that flatly prohibit violations contributing to water quality problems. Rather than putting the onus on permittees to determine how to avoid such violations, the majority requires EPA to draft more specific “limitations” that create “concrete plans” to follow “in order to achieve [the] desired result” of meeting water quality standards.

The majority opinion is clear to avoid one sweeping distortion of the CWA permit program. The court rejected San Francisco’s broad argument the CWA authorizes only permit limitations that met the statutory definition of “effluent limitations.” In doing so, the court avoided disrupting practical and routine permit requirements such as those related to operations, maintenance, record-keeping, or monitoring.

However, the majority’s approach to the CWA text and history represents weaker work.

Justice Amy Coney Barrett’s dissent is more compelling on the plain meaning of the word “limitation;” she even catches her colleagues committing the cardinal sin of omitting part of a quoted definition.

The majority’s interpretation of changes to enforcement sections of the CWA made in 1972 also is, as Barrett notes, tenuous. And the majority’s “everyday” analogy to implementing math standards in schools makes little sense.

Alito compares cities operating wastewater plants to students and regulators to teachers. But if student proficiency is the goal, students would represent the water bodies to which water quality standards apply, teachers would be the regulated entities, and the principal the regulator. That also comports with the way sewer systems operate in reality—like the teachers, municipalities are in the middle of this chain, not its end.

To satisfy the permit conditions imposed on them, municipalities often impose rules on people and businesses using their systems and must make judgments, both in planning and in real time, to react to local conditions. As Alito notes, the “principal’s obvious expectation would be that the teachers devise and ‘implement’ a plan” so the students meet those standards. However, that is precisely the approach the court invalidated.

The majority’s approach is also hard to square with other realities of the CWA permit system.

Alito held that permit limitations must come “from without” and can’t rely on the permittee determining itself what steps are necessary to protect water quality. The court emphasizes that EPA “do[ing] its work” and using its information-gathering tools and expertise will prevent adverse effects on water quality. But this ignores the massive resource demand being imposed on already under-resourced permitting authorities.

Moreover, the court doesn’t consider that this case involving a large city in a unique geography may not be broadly representative of the permitting process for most municipalities. Often, there is a productive relationship between permitting agencies and the public entities operating combined sewer systems—a relationship that goes unmentioned. This opinion undermines a system in which permittees make their own proposals, both during negotiations and when needed to demonstrate why minimal or no enforcement action should be taken in the event of a violation.

It’s curious the opinion focuses on hypothetical criminal penalties and high civil penalty ceilings without noting EPA’s enforcement discretion and without suggesting any municipality has ever been subjected to criminal prosecution for violating these “end-result” conditions.

As Barrett points out, this decision will push EPA to be more adversarial in using its information-gathering powers to write permits with “concrete plans” specific enough to ensure the end-result in terms of water quality protection. The permit process will become more complex and time-consuming, perhaps acceptable for a large, well-resourced permittee like San Francisco—which demonstrated a willingness to delay and litigate providing information in this very case—but potentially paralyzing for smaller systems around the country. The communities served by small sewer systems will pay through increased water rates for the legal and consulting bills this new adversarial permit system will likely bring. This may lead to delays in permit approval and more frequent penalties for unlawful discharges or noncompliance.

Alito’s majority also omits that state agencies are primarily responsible for NPDES permitting, operating under both delegated CWA authority and state statutes. Unlike the CWA, some state laws include language prohibiting “contributing” to water quality problems—the sort of language Alito found lacking in the CWA to justify “end-result” permit conditions.

Does the opinion allow state agencies with such authority to continue imposing similar conditions in permits issued under both state and federal law? The majority opinion offers nothing on that point because it largely fails to engage with how CWA permitting works in reality.

The direct effect of this decision is likely to be the deletion of a few lines in NPDES permits that contain these “end-result” requirements. However, in upending a longstanding permit practice, it’s possible this decision will lead to less protective permits due to the administrative impossibility of drafting sufficiently specific conditions in every permit, greater variation in permitting across states, and a fundamental turn toward a more formal and adversarial permitting system.

The case is San Francisco v. EPA, U.S., No. 23-753, decided 3/4/25.

This article does not necessarily reflect the opinion of Bloomberg Industry Group, Inc., the publisher of Bloomberg Law and Bloomberg Tax, or its owners.

Author Information

Robert Weinstock is clinical associate professor at Northwestern Pritzker School of Law.

Alexa Longstaff is a second-year J.D. student at Northwestern Pritzker School of Law.

Write for Us: Author Guidelines