Wednesday, February 26, 2020

Justice Thomas Was Right Before and Wrong Now: Brand X and Chevron Should Not Be Overruled

by Michael C. Dorf

Dissenting from the denial of certiorari on Monday in Baldwin v. United States, Justice Thomas urged his colleagues to reconsider and overrule its 2005 decision in National Cable & Telecommunications Assn. v. Brand X Internet Services. Justice Thomas himself wrote the opinion for  the Court in Brand X. While he deserves praise for showing a willingness to change his mind, his dissent is wrongheaded. Justice Thomas was right in 2005 and wrong now.

Although the Baldwin cert petition targeted Brand X, the real quarry for Justice Thomas is the doctrine on which it relies: Chevron deference to agencies. Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., permitted the Reagan administration EPA to apply a more lenient interpretation of the term "stationary source" in the Clean Air Act than had the Carter EPA on the ground that an agency's reasonable construction of an unclear statutory term is entitled to deference by reviewing courts. Brand X applied Chevron even in a context in which the prior contrary construction was given by a court rather than by the same agency under a different administration. Part II of Justice Thomas's dissent argues that even if Chevron is right, Brand X is wrong. In a follow-up column, I'll explain why he's mistaken about that. Today I want to focus on what he says about Chevron itself.

In his Baldwin dissent, Justice Thomas says that "Chevron is in serious tension with the Constitution, the APA, and over 100 years of judicial decisions." Here I'll address his extremely dubious constitutional arguments, because if he's wrong about them, then he offers only sub-constitutional grounds for overruling Chevron. I think those arguments are also weak, but even if the sub-constitutional grounds were more substantial, that would not warrant overruling. Chevron is ultimately a default rule of statutory interpretation, which Congress can change if it so chooses. There is no need for action by the Court that would unsettle decades of administrative law jurisprudence on which Congress and the agencies have relied. That's a step Congress can take, but if it doesn't, the Court shouldn't.

To be sure, if Chevron is unconstitutional, then that cuts the other way. Not entirely. Even in constitutional law, reliance and stare decisis count for a fair bit. While it's true that the Court has never directly faced the question whether Chevron is unconstitutional, its longstanding assumption that it is constitutional is entitled to respect due to the reliance on that assumption.

In any event, we need not worry too much about whether Chevron should be retained despite being unconstitutional prima facie, because Chevron is in fact constitutional.

Why does Justice Thomas think Chevron is unconstitutional? Much of what he says in Baldwin quotes his own prior writings, but I won't bother with internal citations. His renewed call warrants a renewed examination. Let's consider the two core claims.

Justice Thomas says first that Chevron violates Article III. Courts have a constitutional duty to exercise their independent judgment to say what the law is, and so Chevron, by requiring them to accept an agency's reasonable construction of a statute that does not comport with the courts' own determination de novo requires judges to abdicate their Article III duty. Justice Thomas doesn't cite Marbury, but we can think of the point as relying on John Marshall's statement that "[I]t is emphatically the province and duty of the judicial department to say what the law is."

In addition, according to Justice Thomas, by giving over the law-declaration/interpretation role to agencies within the executive branch, Chevron violates Article II. Agencies can exercise the executive power, he argues, not the judicial power, which Chevron ostensibly gives them.

Justice Thomas's constitutional objections rely on a tendentious interpretation of Marbury. As Larry Kramer has very helpfully explained--and as one sees throughout the "Constitution outside the courts" literature--Marbury did not establish judicial exclusivity in constitutional interpretation or legal interpretation more generally. What Marbury did was to establish the co-equal ability of courts to expound the law. But the other branches retain the power and duty to say what the law is as well.

To be sure, departmentalism--the view Kramer and other popular constitutionalists expound--is not universally accepted. There are judicial supremacists. Maybe Justice Thomas is one of them.

Yet even if so, his argument against Chevron's constitutionality fails. Certainly the Article II argument must be wrong. Charged with enforcing the law, the executive must routinely determine what the law means in order to know how to enforce it. Agencies cannot possibly seek guidance from courts in advance of applying the law in every circumstance (and wouldn't receive it anyway, given the prohibition on advisory opinions). So there is nothing incompatible with Article II about the executive construing the law.

Assuming we reject departmentalism, the Article III objection initially appears stronger. We can make the most sense of it by discarding the Article II objection and saying that whatever initial power the executive branch has to construe the law, its constructions can't be conclusive on courts. And that might be a good constitutional objection if it were the only way to understand Chevron.

But it isn't the only or even the best way to make sense of Chevron. To see why, let's think about the facts of Chevron itself. At issue was the term "stationary source" in the Clean Air Act. The Carter EPA had treated each smokestack (or equivalent) within an industrial facility as a stationary source that had to comply with emissions limits, whereas the Reagan administration used the "bubble concept," which treated the whole facility as a single stationary source, so that even if a single smokestack considered separately would be deemed out of compliance under the Carter approach, there would be compliance under the Reagan approach if the facility as a whole was within the limit (which would be higher for the larger facility).

Justice Thomas reads Chevron to have given over to the EPA the power to define the statutory term "bubble concept." However, we can understand the statute differently. Consider the following three versions of the statute:
A. The EPA shall require a permit for each stationary source. 
B. The EPA shall require a permit for each stationary source, a term that is deliberately vague so that the agency can exercise its judgment about whether to treat each smokestack as a separate stationary source or to treat an industrial facility composed of multiple smokestacks as a single stationary source. 
C. The EPA shall require a permit for either individual smokestacks or for entire industrial facilities.
I take it that no one -- not even Justice Thomas -- would say that version C is unconstitutional. But C is indistinguishable from B, except for the fact that in B Congress has added the extra step of having the agency exercise its discretion via the term "stationary source." That, however, is purely a difference in wording, not substance. So if C is constitutional, as it indubitably is, then so is B. Now we come to A. Chevron says that when a statute takes the form of A, Congress should be understood to have meant B. That is, Chevron says that the use of a vague or ambiguous term in a statute conferring power on an agency should be understood as a delegation to the agency of the authority to select a reasonable interpretation of the vague or ambiguous term, in light of the agency's expertise and policy views.

Now, one can object that when Congress uses vague or ambiguous terms in statutes authorizing agency action, it doesn't really intend to delegate authority in the way I've just described. And in fact Justice Thomas lodges that objection. But that is entirely a statutory objection, not a constitutional one. If one rejects the statutory objection, then A is equivalent to B, which is equivalent to C, which is clearly constitutional. So there is no valid or even plausible constitutional objection to Chevron. Justice Thomas's constitutional argument is entirely parasitic on his statutory interpretation objection.

I think Justice Thomas is wrong about the statutory interpretation points, but even if he is right, that's for Congress to change. Just yesterday, Justice Thomas joined an opinion by Justice Alito in Hernandez v. Mesa that was replete with arguments to the effect that if Congress doesn't like the Court's implied-right-of-action jurisprudence, it can legislate a solution. The same is true here. Chevron has been the law for over 35 years, and as Justice Stevens (the author of Chevron) frequently observed, it did not break new ground, so the notion of deference to agencies is older still (despite what Justice Thomas says in Baldwin). If it ain't broke, don't fix it, and even if it is broke, leave the fix to the appropriate actor.

5 comments:

  1. Chevron is the love of my life, but I don't think I agree with you about statutory stare decisis and Chevron. Would you claim that the Court shouldn't touch its super-clear-statement approach to abrogating sovereign immunity, or Gregory v. Ashcroft, or how it does constitutional avoidance, or its frankly ridiculous mania about anti-redundancy, until Congress, I don't know, amends the Dictionary Act to say, e.g., that questions of abrogation should be decided straight-up, even if you think these default rules are wrong? I think the Court should endeavor to get its default rules right. Now, I'll grant you that statutes enacted post-Chevron were largely enacted on the assumption Chevron would apply, though Congress might not always anticipate that courts will find ambiguous what they do. But as to pre-Chevron statutes, such as the Clean Air Act itself, Congress didn't rely on Chevron in writing them and there are lots of reasons for inaction with respect to Chevron besides thinking it's right -- besides which, even if today's Congress thinks it's right, they didn't enact the Clean Air Act and the Congress that did may not have anticipated or agreed with Chevron and indeed may have written it on the premise of the muckier doctrine that preceded Chevron. So I don't ask for super-strong statutory stare decisis for Chevron, or really get the strong theory of statutory stare decisis generally, I suppose.

    ReplyDelete
  2. To be clear, I think Chevron is right, so stare decisis isn't needed, but the Court--including Justice Thomas--has said that stare decisis has extra force in statutory cases, and I think that's also sensible, so long as "extra" doesn't mean "infinite." As to the transition, I think Stevens is right that Chevron didn't break much new ground. There was deference before too. Maybe not as far back as the 19th century but when Congress wrote the Clean Air Act.

    ReplyDelete
    Replies
    1. Assuming I'm a good textualist for purposes of this question, or even someone who cares about the purposes and intentions of a statute's enacting Congress, could you explain to me why it's sensible? What does subsequent inaction in response to what we assume is a misinterpretation as an original matter really have to do with anything?

      Delete
  3. It's a comparative judgment about statutory versus constitutional cases. The bar is much higher for constitutional amendments, so failure to amend the Constitution in response to a presumptively erroneous constitutional ruling cannot be taken as Popular acquiescence in the ruling, whereas Congress's failure to override a statutory ruling--especially over the course of many years--is more of a signal of acquiescence. If you want to put that in terms of the enacting Congress, we can also imagine that there's some correlation between the enacting Congress's preferences and those of subsequent Congresses. Admittedly, that's an intentionalist inference, not a textualist one. But I think there's no such thing as a "good textualist," at least not if textualism is distinct from reasonable purposivism (cf Kagan "we're all textualists now").

    ReplyDelete
    Replies
    1. What if, to put this in raw terms, King v. Burwell had come out the other way and a Republican Congress had "acquiesced" to it? That would basically tell us only that the party that voted against a bill was happy with an interpretation of that bill that, I think it's uncontroversial to say, would thwart its purposes and greatly reduce its scope. Now, I think it would be really strange to use that sort of acquiescence as an argument for strengthened stare decisis, sans some dynamic theory that looks to currently enactable interpretations, such as Elhauge's or perhaps Eskridge's, and which I don't think is the Court's or any Justice's. If that's right, by what speakable means could the courts distinguish the sort of acquiescence that's relevant from the sort of oppositional acquiescence that's not, and might it not just be better to disregard acquiescence so as not to rope in the kind that if anything is almost a counterargument against adhering to a statutory precedent (that people who voted against a bill, or would have, like what the Court's done to it)?

      Delete