The Ideological Stakes of Overruling Chevron

 by Michael C. Dorf

On Monday, the Supreme Court granted certiorari in Loper Bright Enterprises (LBE) v. Raimondo in an order that limited the grant to question 2 of the petition, which asks: "Whether the Court should overrule Chevron or at least clarify that statutory silence concerning controversial powers expressly but narrowly granted elsewhere in the statute does not constitute an ambiguity requiring deference to the agency?"

For those of my readers who did not attend law school in the last 40 years, didn't take administrative law when they did, or took administrative law but forgot the one thing that everyone else remembers from that class, Chevron refers to the 1984 case of Chevron U.S.A., Inc. v. NRDC. It has long been cited for the proposition that, when Congress delegates power to an agency in unclear statutory language, a court reviewing administrative action will defer to a reasonable agency interpretation of the statute even if the court believes that some other interpretation is better.

For many years, Chevron was understood to involve a two-step process: First, is the statute unclear? If not, courts give no deference. If yes, then second, is the agency's interpretation reasonable? More recently, scholars have argued that there are additional steps. There is step zero, step one-and-a-half,  and as I once suggested mostly tongue in cheek, step negative one.

The Court could resolve LBE by addressing only the second half of the cert question on which it granted, but doing so would really amount to a Chevron step zero holding: it would be saying that when one reads the statute as a whole, indulging ordinary rules of statutory construction, what appears to be an ambiguity in a clause standing in isolation is not in fact ambiguous. I don't have a view about whether that proposition fairly applies to the particular statute at issue in LBE but I doubt the Court granted review in the case in order to "clarify" a proposition that is already entailed by its prior step-zero cases.

Moreover, if the Court were interested in finding a way to resolve LBE without deciding whether to overrule Chevron, it would have granted review on both questions presented in the cert petition. Because it didn't, it appears there are at least four Justices who want the opportunity to overrule Chevron. Given the druthers of the Chief Justice, this is not the sort of issue with respect to which I see him seeking to moderate the Court's direction--especially given that the Court is unlikely to pay a price with public opinion for overruling Chevron. Despite what I expect to be the best efforts of the likes of Nina Totenberg and Adam Liptak, most of the public won't comprehend the case or its stakes.

My readers aren't most of the public, however, so I'll add a few words about those stakes. Spoiler alert: I think they're modest rather than very high, but only because, the courts already have powerful tools at their disposal to undercut effective regulation, even without formally overruling Chevron.

Overruling Chevron would not necessarily be deregulatory. The Chevron case itself upheld a Reagan administration interpretation of the Clean Air Act Amendments of 1977 that was challenged by environmentalists as being too lax. The Court deemed the term "stationary source" sufficiently unclear to give the administration policy discretion and then upheld the industry-friendly exercise of that discretion as reasonable. Given that Chevron itself approved a deregulatory agency policy, its overruling wouldn't necessarily be deregulatory. At least in principle, Chevron results in deference to agencies regardless of whether those agencies are answerable to a Democratic or a Republican president.

But that's in principle. Some studies show that Republican appointees are more likely to defer to agency decisions during a Republican administration than during a Democratic one, and vice-versa. To be sure, the most comprehensive study thus far claims to find that Chevron acts as a substantial constraint on partisan judging, but I would note two important caveats.

First, substantial constraint hardly means complete constraint. The study just linked finds that even when appeals court panels employ Chevron deference, political affiliation of the judges makes a difference. For instance, liberal panels grant Chevron deference to (what the study's authors code as liberal) agency interpretations in 91 percent of cases, whereas conservative panels do so in only 66 percent of cases. Even so, conservative panels voting to uphold liberal policies 2/3 of the time does suggest, as the authors conclude, that Chevron makes a difference.

But that brings me to my second caveat. The authors used as their control sample cases in which the courts did not apply Chevron. They found that judges' ideological orientation made a bigger difference then. However, whether to apply Chevron is itself up to the judges and manipulable at Chevron step zero. Thus, some of what the authors conclude is a constraining effect of Chevron could be a product of the fact that judges who, on ideological grounds, do not wish to defer to agencies, will find some way not to apply Chevron because that's easier than declaring an agency interpretation unreasonable.

The upshot of all of that is that Chevron results in a bit more non-ideological deference to agencies than we would see without Chevron. Accordingly, overruling Chevron would make it somewhat easier for judges to simply vote their ideological druthers. At the appeals court level, that would lead to an ideologically mixed result because the personnel on the appeals courts are currently ideologically mixed. However, given the conservative super-majority on the Supreme Court, a greater ability to vote ideological druthers means a greater likelihood of right-leaning decisions in challenges to agency action from that Court. And because SCOTUS is likely to decide the most consequential administrative law cases, that effect probably dominates the mixed impact at the appeals court level. Accordingly, insofar as Chevron has been acting as a constraint, its overruling would reduce that constraint and thus lead to more ideologically conservative results in the most important cases.

That effect is likely to be modest, however, in light of two other factors. First, as noted above, even under Chevron, judges and Justices have multiple points at which they can invalidate agency action they dislike while claiming simply to be following the Chevron steps. Second, whether conceptualized as one of those steps or as an independent doctrine, the recent inflation of the major questions doctrine (MQD) gives conservative judges and Justices a powerful tool for gutting the administrative state.

Indeed, MQD is already in considerable tension with Chevron. Chevron says (more or less) that where Congress leaves an issue open, it can be understood to have delegated to an agency the discretion (to be exercised reasonably, of course) to address that issue. But MQD says that if addressing the issue would be "major" in the sense of imposing high compliance costs or involving a substantial sector of the economy, then Congress should be presumed not to have delegated the power to answer that question to the agency. Given the manipulability of what counts as a major question and the fact that SCOTUS tends to take only important cases, at least at the Supreme Court level, MQD can be used to avoid deference pretty much at will.

Finally, unlike Chevron, MQD isn't even arguably ideologically neutral. MQD says that Congress should not be presumed to have delegated power to agencies to regulate major portions of the economy. As CJ Roberts wrote for the Court last Term in West Virginia v. EPA, the MQD applies when "agencies assert[] highly consequential power . . . ." Asserting power means regulating. Choosing not to regulate or to deregulate would not typically be such an assertion of power. Accordingly, MQD is inherently deregulatory.

Bottom line: Overruling Chevron would give judges and Justices somewhat greater flexibility in choosing how to invalidate regulatory actions by administrative agencies, but it wouldn't fundamentally change very much, because the Roberts Court has already gone far down the road towards deconstruction of the administrative state.