The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


Overturning Chevron Would Not Gut the Administrative State—but It Would Strengthen the Rule of Law

Eliminating judicial deference to administrative agencies' interpretations of federal law would not destroy the administrative state, or even significantly reduce the amount of regulation. But it would have some real benefits, nonetheless.


The Chevron Corporation was a party to the famous Supreme Court precedent increasing judicial deference to administrative agency's interpretations of federal law.

Judge Brett Kavanaugh, Donald Trump's latest Supreme Court nominee, is a critic of the Chevron doctrine - the legal rule requiring federal judges to defer to executive branch agencies' "reasonable" interpretations of federal law in situations where Congress has not clearly addressed the issue in question. Neil Gorsuch, Trump's previous Supreme Court appointee, is an even more thoroughgoing Chevron skeptic. This has led some on the right to hope and many on the left to fear that Kavanaugh's confirmation will lead to the gutting of Chevron, which in turn would strike a major blow against the "administrative state." For the most part, neither the hopes nor the fears are justified.

Kavanaugh's appointment may well continue (or even accentuate) the Court's growing tendency to cut back on Chevron deference - though it is not clear that Kavanaugh's Chevron skepticism is greater than that recently expressed by Anthony Kennedy, the justice Kavanaugh would replace. But even if the Court imposes tighter limits on Chevron deference or eliminates it entirely, that would not put much of a dent in the administrative state

In discussing this issue, many people tend to forget that the Supreme Court only decided the Chevron case in 1984, and we had a large and active administrative state long before then. Somehow, the powerful agencies established in the Progressive era, the New Deal, the Great Society, and the Nixon administration managed to survive, thrive, and regulate without Chevron.

Pre-Chevron administrative agencies did enjoy the benefit of less extensive forms of judicial deference, such as "Skidmore deference." Those would likely persist even if Chevron were to be severely limited or overruled. But even if the Supreme Court were to completely eliminate judicial deference to administrative agencies' interpretations of federal law (thereby treating them the same as any other litigant), the latter would still wield enormous discretionary power. In a world where there are far more federal laws than any administration could hope to effectively enforce, they would still have broad discretion to determine which violators to go after, and how aggressively. They would also retain control over a broad array of technical questions related to enforcement rules, scientific standards for regulation, and other issues.

Even on the specific question of interpretation of statutory law, the elimination of formal deference probably would leave in place a good deal of deference in practice. Across a wide range of issues, generalist judges seeking to manage large case-loads may still give special weight to the views of supposedly expert agencies, even if they are not formally required to do so. This is especially likely to happen when it comes to questions that are highly technical and not ideologically controversial. Deference would be weakened, but far from completely eliminated.

To the extent that ending Chevron would put agencies on a tighter leash, it is far from clear that this would necessarily benefit the political right more often than the left. As my VC co-blogger (and leading administrative law scholar) Jonathan Adler points out in a New York Times article, a reduction in judicial deference could stymie deregulatory policies as readily as those that increase regulation. The Chevron decision itself deferred to a Reagan administration policy that shifted air pollution regulation in ways decried by environmentalists.

In policy areas such as immigration and drug prohibition, most conservatives - especially since the rise of Trump - actually favor more regulation than most of the left does. Pereira v. Sessions, one of the Supreme Court's recent decisions cutting back on Chevron deference strikes down a policy that sought to make it easier to deport immigrants. The same is true of then-Judge Gorsuch's most famous lower court opinion criticizing Chevron.

Ending Chevron deference would not gut the administrative state, or even systematically advance the cause of deregulation. It would, however, have some important beneficial effects. It would put an end to what then-Judge - and future liberal Supreme Court justice - Stephen Breyer, writing in 1986, called an "abdication of judicial responsibility." Neil Gorsuch expressed similar views more recently, calling Chevron "a judge-made doctrine for the abdication of the judicial duty." The Constitution gives judges, not agency bureaucrats, the power to interpret federal law in cases that come before the courts. Getting rid of Chevron would help restore the proper constitutional balance in this area.

The elimination of Chevron would also increase the stability of legal rules, and make it harder for administrations to play fast and loose with the law. As Gorsuch pointed out in a well-known opinion he wrote as a lower court judge, Chevron deference often enables an agency to "reverse its current view 180 degrees anytime based merely on the shift of political winds and still prevail [in court]." When the meaning of federal law shifts with the political agendas of succeeding administrations, that makes a mockery of the rule of law and undermines the stability that businesses, state governments, and ordinary citizens depend on to organize their affairs. A new administration should not be able to make major changes in law simply by having its agency appointees reinterpret it. Such power is even more dangerous in an age of severe polarization and widespread partisan bias.

Conservatives suspicious of liberal bureaucrats have good reason to oppose Chevron. But the same is true of liberals who believe (often correctly) that GOP administrations appoint agency heads who bend the law in order to advance partisan agendas. If you don't trust the likes of Jeff Sessions or Scott Pruitt to interpret the law objectively, you should be happy to see a judiciary that defers to them as little as possible.

NEXT: "US Court Authorizes Service By Twitter on Wikileaks"

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Chevron deference does not "abdicate" judicial responsibility except to the extent you define that to require a weaker form of deference. For those who care about original understanding, there was a strong understanding of much deference overall of governmental action before courts can overturn it. More so than is in place now.

    Deference doesn't mean abdication. If the regulation violates a constitutional provision, let's say, the courts would not just handwave it. It also requires some degree of a fit between the regulation and the underlining law. Congress can also ultimately change regulations. The flexibility provided here is part of the overall understanding of the administration state.

    It is somewhat question begging there if the "rule of law" is threatened in that respect, again because Chevron deference still leaves open some judicial review. Ultimately, it is a matter of degree. Do we trust agencies with professional experts that have overhanging oversight by people elected and appointed/confirmed by those elected with specific questions somewhat more than other situations? Agency action also includes much room of public feedback and opens internal review opportunities.


    1. Do generalist judges have the wherewithal to determine close questions of agency action? There are going to be questions along the edges here but it is unclear to me that a new normal is required. It won't "gut" the administrative state -- though certain people rather it did -- but can badly interfere with it. The reality of the situation, given all the regulations involved etc., is that not TOO much will change in the end though that might counsel us that change isn't really worth it.

      [This is a general reply and is not meant to be a direct reply.]

  2. I would agree. The problem is that agency interpretations of statutes seem often to be quite political. Someone wins an election, puts people in who share their politics, and they, by controlling the agency, control the interpretations of the relevant laws. The theory behind Chevron is that agencies have some sort of expertise in their area, so should be deferred to. But, in many cases it is the politics, and not any special agency expertise, that is driving the agencies' interpretations of the relevant laws, and why should courts give such politicized interpretations judicial difference?

    1. Because elections matter.

      1. Therein lies the problem

        1. Elections are a problem when dealing with words from the mouths of officials that, when disobeyed, land you in jail?


      2. And elections matter a great deal more when there are no effective limits on the scope of power which government may yield. Our founders attempted to solve, or at least minimize, this problem by setting limits on what government could do. Sadly, the judiciary abdicated its responsibility to enforce those limits long before the arrival of Chevron deference. Reining in Chevron deference is a start, but too little and too late, as far as I am concerned.

        1. Watching conservatives pivot from judicial minmalism to judicial maximialism has been a ride.

        2. Chevron deference isn't unconstitutional. There is no abdication by the judiciary of its constitutional responsibility. It's just an arbitrary judge-made rule for deciding a tiny class of close cases. As the OP pointed out, Chevron itself involved deference to deregulation. In that specific case, reversing Chevron means removing "limits on the scope of power which government may yield".

        3. Absolutely, tortured interpretations allowed FDR's New Deal to trample individual and economic freedom all the way to recent history such interpretations allowed Obamacare to stand.

    2. There are a range of possibilities including among experts and elections influence the results there.

      There still is "special agency expertise" involved, just a new set of experts.

      A related thing occurs when giving schools or local prosecutor offices a range of discretion, which occurs. Elections put new people in with special expertise that might do something differently. Since there is debate on how to do things among experts too.

      Elections is part of the system in place. Chevron deference isn't total deference. When politics overwhelm and there does not seem to be a reasonable grounds or it violates some constitutional provision etc., there still is room for judicial checks.

    3. Right, and once an agency determines that one time what a certain word means, reversing that understanding by the same process and degree of deliberation is considered "arbitrary and capricious".

      Carbon dioxide as a pollutant. Tobacco products as medical devices. Water bodies as all draining into navigable channels. Indecency as forever malleable (an example of the opposite tactic: an agency's deliberately failing to ever adopt a clear standard).

      1. "...reversing that understanding by the same process and degree of deliberation is considered "arbitrary and capricious"."

        Why, because you said so? Chevron deference only applies if the statute is ambiguous. The alternative rule is that whatever administration sets the regulation when the statute is passed gets to hang on to it forever, no matter what elections happen, unless and until Congress rewrites the statute. That's as "arbitrary and capricious" a rule as deference in real time.

        1. No, the alternative rule is that when someone litigates it, the court decides what the statute means, rather than accepting what the agency thinks it means.

          1. Sure. But why is the court's interpretation any more likely to be accurate than the agency's? Agencies at least can claim technical expertise and are required to go through a fairly lengthy deliberative process before issuing regulations. Judges have no technical expertise and make their interpretations based on what amounts to a single administrative hearing.

          2. Great, so for an agency with a national scope, we'll have dozens of federal district judges issue conflicting interpretations of an ambiguous statute rather than just one. That's even more arbitrary and capricious.

  3. I'm curious how much free advertising Chevron has received from Chevron deference. Exxon must be pissed.

    1. lawyers seeing the Chevron sign might be more likely to pull in for gas or service there, by the same means that anybody wants their brand name repeated a lot. Never thought of that.

      That's OK, Exxon benefits from the publicity regarding the cat fight between their tiger & Kellogg's once Exxon's Tiger Marts started selling food. I wonder if they have one of those in Princeton.

    2. It can confidently say more people know about the Exxon Valdez than Chevron deference

  4. Some people will tell you deference makes the law more reliable?that article 3 judges having the ability to overturn more nimble article 1 tribunals will make people less sure of the ground they have to stand on to comply with regs. But I think it's the other way around, because people's having the ability to understand and rely on the statutes as they come out of Congress, without fear that an administrator will later adopt a meaning that doesn't comport with the legislation, will make them more sure of the rules.

    Remember how it was threatened to be with tobacco products 20 yrs. ago? FDA asserted, after decades of saying the opposite when petitioned by outsiders, that these were medical devices containing drugs. They would've had to strain and twist a regulatory system designed for therapeutic products to fit cigarets, etc. Once so strained and twisted, the regs would've been entirely subject to administration whim in order to work at all.

    1. But I think it's the other way around, because people's having the ability to understand and rely on the statutes as they come out of Congress, without fear that an administrator will later adopt a meaning that doesn't comport with the legislation, will make them more sure of the rules.

      That statement assumes that the Courts will both overturn administrative interpretations that do not comport with the meaning of the statute and will not overturn reasonable interpretations and insert unreasonable ones of their own. Those assumptions show more faith in the courts than can reasonably be justified given their record in such matters.

  5. Obviously, the deference sets up a situation where you get unprecedented reinterpretation by agencies with substantial input and influence by the special interest groups du jour. The EPA is a great example of an agency in turn giving deference to their preferred partners to in essence rewrite statues. For example, if congress meant every possible body of water, except for some well-defined artificial waterways, to be covered by the clean water act why did it even use the term navigable? Oh wait, they used the term navigable so that the act would be constitutional under their limited powers. Kinda' begs the question, don't it.

  6. It would not gut the administrative state at all. It would just move the power from bureaucrats to judges. What is lost in this debate is the fact that Chevron deference protects inaction or a restrictive interpretation of the law as much as it does activism and broad interpretations. Overturning Chevron would allow federal judges to overrule agencies decisions to weaken or revoke regulations just as much as it would allow them to stop new ones. That fact never seems to occur to people like Somin. Somin seems to think that every judge is going to be pure and take their new power to overrule the bureaucracy as a means to restrain government power rather than a means to insist upon its expansion or at least prevent its reduction.

    I think Somin means well here and I agree with his sentiment. The problem is that Somin doesn't seem to understand how regulatory law works in practice and places far too much faith in judges. I would rather have the Bureaucrats who at some level answer to the President who then answers to the voters making the call than judges who answer to no one but themselves making the call. Ideally, Congress undoes these laws and regulations. But that is not going to happen.

  7. I think overturning Chevron is just as likely to expand the administrative state as it is to restrict it. If an agency rules that a regulation doesn't reach a certain person or activity, without Chevron, the court is free to insert its own interpretation that it does. Understand, that citizen activist groups have standing to sue over these regulations. Such suits can involve a group objecting to a narrow interpretation made by an agency just as easily as they can involve a broad one.

  8. I either don't have or won't express here an opinion whether Chevron was rightly decided (beyond the character limit in any event). But I do want to agree with Ilya that the pre-Chevron regime in practice was significantly deferential to agencies. Indeed (though I wouldn't attribute this to Ilya), in many ways agencies were better off with "only" Skidmore deference. With Chevron, an attorney defending the agency tends to get locked in to the ordering of Chevron's framework. But beginning a brief with a weak "step 1" argument because it's, well, step 1, may not be a good idea. I would rather (back when I did such things) have defended an agency with a brief arguing that the agency decision is absolutely right dammit than with a mealy mouthed, correct-enough-for-government-work brief or a suggestion that the subject is so darn complicated that a mere article III generalist couldn't possibly figure it out and should defer to an agency. Plus, arguing a case under Skidmore not only frees the government attorney from a framework that might be ill-suited to that particular case, but also allows relying on the wealth of different Skidmore era precedents (Norwegian Nitrogen, e.g.) that may be much more on point in explaining why the agency got it right.

Please to post comments