Freedom and a Funeral for Chevron Deference

Forty years ago, perhaps unwittingly, the Supreme Court instructed lower courts to defer to reasonable agency interpretations of ambiguous federal statutes. In the years since, lower courts heeded this instruction perhaps all too well, leading to concerns courts were abdicating their responsibility to interpret laws and facilitating agency aggrandizement. Last week, in Loper Bright Enterprises v. Raimondo, the Supreme Court said enough was enough. Chief Justice John Roberts, writing for a six-justice majority, put an end to so-called Chevron deference.  

“Today, the Court places a tombstone on Chevron no one can miss,” proclaimed Associate Justice Neil Gorsuch in a concurring opinion. Yet any reports of the administrative state’s death would be premature. The Court’s decision in Loper Bright may help corral some agency excesses, but it is unlikely to prove a meaningful obstacle to expansive federal regulation.  

Chevron deference takes its name from Chevron USA  v Natural Resources Defense Council, a 1984 case in which the unanimous Supreme Court rejected an environmentalist group’s challenge to Reagan Administration regulatory reforms that sought to make it easier for firms to comply with the Clean Air Act (CAA). At issue was whether the EPA had adopted a legally permissible definition of the word “source.” Under the CAA, “stationary sources” of regulated pollutants were required to obtain permits when making facility changes that could increase emissions. But what is a “source”? Is it the smokestack or opening from which pollutants emanate? Or is it a given building, or an industrial facility as a whole? NRDC preferred the former interpretation, as had the Carter administration, but industry and the Reagan Administration preferred the latter, and the text of the CAA did not answer the question.  

The Supreme Court found the relevant text of the CAA unclear and determined that this sort of choice — whether to adopt a more stringent or a more flexible regulatory policy — was the sort of policy choice Congress had left to the EPA to make as it administered the law. 

The Chevron opinion, written by Justice John Paul Stevens, included a rather thorough examination of the CAA’s text, structure and legislative history, none of which answered the question of how “source” should be defined. In explaining why this meant the EPA should get to choose, Justice Stevens articulated a two-part test for evaluating agency interpretations of federal statutes. First, look to the text of the statute to see if it answers the question at issue. If so, follow Congress’s instructions. If not, however, then defer to any reasonable interpretation offered by the implementing agency, in this case the EPA.  

Justice Stevens did not understand his opinion to announce a new test, let alone a doctrine, and nor did his colleagues. Indeed, it took several years before Bush Administration officials and other justices seized on his language in the Chevron opinion to demand judicial deference to agency interpretations. Over time, this text took hold, and Chevron became one of the Supreme Court’s most-cited opinions and a centerpiece of modern administrative law, allowing agencies to define the scope of their own power. 

In theory, Chevron was supposed to help provide greater legal certainty, both by providing judges with a clear framework and producing greater uniformity of federal law. In practice, the doctrine appeared to serve the latter purpose, in that it fostered greater agreement across federal courts and seemed to reduce the extent to which judges’ policy preferences influenced their decisions. But it did not produce stability or predictability. To the contrary, it empowered agencies to revise and reverse their statutory interpretations to align with their policy goals, and so long as no court declared that the statute was clear, they could get away with it. Chevron itself involved a Reagan Administration reversal of Carter Administration policy, and the doctrine has been used to justify agency about-faces on the meaning of regulatory laws when the White House changes hands. 

In recent years, the Court repeatedly sought to remind lower courts that, even under Chevron, courts retained the obligation to interpret relevant statutory provisions independently, using the “traditional tools of statutory interpretation,” and to follow the clear meaning of the statute if one could be found. It was only if the law had run out, and the statute truly failed to answer the question at hand, that deference to a reasonable agency interpretation was due. As articulated by the Court, this second step should only be reached after reviewers had exhausted the “traditional tools of statutory interpretation.” Accordingly, reaching the second step should have been somewhat rare. That a statute is complicated, technical, or difficult to parse does not mean it does not have a clear meaning, just that it may take more work to uncover. 

Justice Anthony Kennedy, in one of his last opinions on the Court, lamented that lower courts had been abdicating their responsibility to resolve questions of law, providing the most “cursory analysis” of statutory text before embracing whatever position an agency had offered. The “reflexive deference” offered by many courts was “troubling,” particularly when it gave federal agencies a green light to expand the scope of their own authority and address matters never delegated to them by Congress. Chevron, in Kennedy’s view, was about allowing agencies to fill in details and exercise policy discretion, not aggrandize their own power.  

The central thrust of the Chief Justice’s opinion overturning Chevron in Loper Bright is that it is a judge’s responsibility to resolve questions of law when cases come before federal courts. As Alexander Hamilton wrote in Federalist 78, “interpretation of the laws is the proper and peculiar province of the courts.” This is not merely a matter of constitutional design. It is also explicit in the Administrative Procedure Act, which instructs reviewing courts to “decide all relevant questions of law” and “interpret constitutional and statutory provisions,” as well as to “set aside agency action” that the court determines is “not in accordance with law.” So, if an individual goes to court alleging that a given regulation exceeds the scope of the agency’s power or otherwise misapplies the relevant statutory text, the court must answer that question, and it must do so without giving the agency a trump card. 

It is one thing for courts to acknowledge that agencies often have special expertise about the matters they are entrusted to regulate, and may even have special insight into the meaning and purpose of given statutory provisions. Indeed, in some cases, agency attorneys may have drafted and proposed the statutory language Congress later enacted. But it is something else entirely to say that when there is doubt, the agency must prevail. But that is what Chevron had provided. Under step two of the Chevron doctrine, a reviewing court was instructed to uphold any “permissible” interpretation of ambiguous statutory text offered by the agency, even if not the best possible interpretation or that which the court would have adopted on its own. 

Ending Chevron sends a signal to lower courts that they need to evaluate agency assertions of regulatory authority more carefully and fulfill their independent obligation to interpret statutory text, but it hardly means an end to the administrative state. The Court is not preventing agencies from regulating. Nor is it telling lower courts to reject agency interpretations. Rather, the Court is reinforcing the point, recently made in its so-called “major questions doctrine” decisions, that agencies are only to exercise power which Congress actually delegated to them.  

Accordingly, the Court did not go so far as to say that courts are required to review all statutory interpretations de novo, without any regard for the implementing agency’s opinion. After all, agencies often know quite a bit more about the relevant subject matter and how a given statute may apply than does a reviewing judge; agency staff are often subject-matter experts, while federal judges are generalists. Nonetheless, agency opinions are just that: opinions. They may have the power to persuade, but they do not bind, absent a clear indication from Congress that the question at hand is wholly delegated to the agency’s discretion.  

The Court, in effect, is instructing lower courts to review agency interpretations of statutes as they did in the decades before Chevron, a period in which federal regulatory agencies were quite active. No doctrine instructed courts to defer reflexively to agency interpretations in the 1960s, 1970s, or most of the 1980s, but federal agencies issued myriad new rules, particularly with regard to health, safety, and environmental protection. Under a test outlined in Skidmore v. Swift & Co. (1944), courts were to consider agency views in light of “the thoroughness evident in its consideration, the validity of its reasoning, its consistency with earlier and later pronouncements, and all those factors which give it power to persuade, if lacking power to control.”  

The biggest difference between now and the pre-Chevron era is that courts back then  

were prone to accept broad, purposivist interpretations of federal statutes, and thus were more likely to approve of broad assertions of agency authority that could be said to advance Congress’ intentions, even if not provided for in the statutory text. Today, however, federal courts are more focused on statutory text and less inclined to believe that Congress has delegated broad power to an agency without express statutory warrant. After all, no statute pursues its stated purpose to the infinite degree. The scope of any delegated power will have some limit, so the question inevitably becomes whether the agency action under review is on the right side of the line.  

Another difference between now and then is that Congress used to take seriously its obligation to legislate and to update regulatory statutes with some regularity. The major environmental statutes, for example, used to be revised and reauthorized every five to ten years. Relevant provisions rarely got too obsolete or out-of-date. Today, however, such updates are a rarity. Most of the major federal environmental statutes have yet to be updated this century.  

In the absence of legislative actions, the executive branch has not been content to sit on its hands. To the contrary, successive presidential administrations have become increasingly aggressive at attempting to anchor far-reaching policy initiative in decades-old statutory text, thus pouring new wine out of old bottles. Examples include the Centers for Disease Control and Prevention’s COVID-inspired eviction moratorium, the Occupational Safety and Health Administration’s test-or-vaccinate mandate for large employers, the Environmental Protection Agency’s Clean Power Plan, and the Biden Administration’s student-loan forgiveness plan. In each case, the agencies in question decided on their course of action and then searched out statutory text sufficiently ambiguous to justify their plans.In each case, the Supreme Court found the agency’s justification wanting. The Loper Bright decision reaffirms that this is not the way that agencies are supposed to do their jobs. As the Chief Justice wrote: “At best, our intricate Chevron doctrine has been nothing more than a distraction from the question that matters: Does the statute authorize the challenged agency action?” 

Overturning Chevron should encourage lower courts to police agency assertions of power more closely, but it will not constrain the administrative state. So long as Congress persists in enacting broad, poorly written statutes and fails to amend and update those statutes as circumstances or policy priorities change, the executive branch will seek to fill the gap and courts will be left playing defense. And as before Chevron was decided, the more technical or complex the subject matter at hand, the more likely it will be that judges will defer to agency judgment. 

Loper Bright has replaced deference with respect, but it will take more than that to meaningfully constrain the administrative state.