The Volokh Conspiracy

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

Supreme Court

Does Justice Kennedy Want to Restore Chevron's Domain? (Updated)

Justice Kennedy calls for reconsidering parts of the Court's Chevron jurisprudence.


Today in Pereira v. Sessions, a nearly unanimous Court concluded that the relevant statutory text provides that a notice to appear for a removal proceeding must include the time and place of the hearing in order for the notice to trigger the "stop time" rule, which may limit eligibility of some nonpermanent residents to remain in the country. The case concerned a small, yet important, question of immigration law.

In an opinion by Justice Sotomayor, the Court concluded "the plain text, the statutory context, and common sense" all supported its conclusion. Because the majority thought the statute was clear, the federal government's contrary interpretation of the statute was not eligible for Chevron deference, a point on which Justice Alito disagreed.

Writing separately, Justice Kennedy took the opportunity to reveal his discomfort with the Court's Chevron jurisprudence and call for reconsidering Chevron from the ground up.. He wrote:

This separate writing is to note my concern with the way in which the Court's opinion in Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984), has come to be understood and applied. The application of that precedent to the question presented here by various Courts of Appeals illustrates one aspect of the problem.

The first Courts of Appeals to encounter the question concluded or assumed that the notice necessary to trigger the stop-time rule found in 8 U. S. C. §1229b(d)(1) was not "perfected" until the immigrant received all the information listed in §1229(a)(1). . . .

That emerging consensus abruptly dissolved not long after the Board of Immigration Appeals (BIA) reached a contrary interpretation of §1229b(d)(1) in Matter of Camarillo, 25 I. & N. Dec. 644 (2011). After that administrative ruling, in addition to the decision under review here, at least six Courts of Appeals, citing Chevron, concluded that §1229b(d)(1) was ambiguous and then held that the BIA's interpretation was reasonable. . . . The Court correctly concludes today that those holdings were wrong because the BIA's interpretation finds little support in the statute's text.

In according Chevron deference to the BIA's interpretation, some Courts of Appeals engaged in cursory analysis of the questions whether, applying the ordinary tools of statutory construction, Congress' intent could be discerned, 467 U. S., at 843, n. 9, and whether the BIA's interpretation was reasonable, id., at 845. In Urbina v. Holder, for example, the court stated, without any further elaboration, that "we agree with the BIA that the relevant statutory provision is ambiguous." 745 F. 3d, at 740. It then deemed reasonable the BIA's interpretation of the statute, "for the reasons the BIA gave in that case." Ibid. This analysis suggests an abdication of the Judiciary's proper role in interpreting federal statutes.

The type of reflexive deference exhibited in some of these cases is troubling. And when deference is applied to other questions of statutory interpretation, such as an agency's interpretation of the statutory provisions that concern the scope of its own authority, it is more troubling still. See Arlington v. FCC, 569 U. S. 290, 327 (2013) (ROBERTS, C. J., dissenting) ("We do not leave it to the agency to decide when it is in charge"). Given the concerns raised by some Members of this Court, see, e.g., id., at 312–328; Michigan v. EPA, 576 U. S. ___, ___ (2015) (THOMAS, J., concurring); Gutierrez-Brizuela v. Lynch, 834 F. 3d 1142, 1149–1158 (CA10 2016) (Gorsuch, J., concurring), it seems necessary and appropriate to reconsider, in an appropriate case, the premises that underlie Chevron and how courts have implemented that decision. The proper rules for interpreting statutes and determining agency jurisdiction and substantive agency powers should accord with constitutional separation-of-powers principles and the function and province of the Judiciary. See, e.g., Arlington, supra, at 312–316 (ROBERTS, C. J., dissenting).

This opinion is notable for several reasons. First, while Justice Kennedy has joined opinions in the past that adopted a constrained view of Chevron (such as Chief Justice Roberts' Arlington v. FCC dissent), Justice Kennedy has rarely expressed these sorts of concerns in his own opinions. (One exception is Gonzales v. Oregon.)

Second, I find it interesting that he not only cites concerns raised by Justice Thomas, but also by then-Judge Gorsuch on the Tenth Circuit. Justice Kennedy presumably thinks these concerns should be taken seriously.

Third, Justice Kennedy's formulation—calling to "reconsider" Chevron's "premises" and implementation—does not necessarily presage a willingness or desire to overturn Chevron. Rather, it seems to reflect a recognition that courts at all levels have not always paid enough attention to Chevron's foundation and have not always considered the extent to which Chevron deference should be limited to what Tom Merrill and Kristin Hickman call "Chevron's Domain."

Justice Kennedy's recognition that Chevron has limits—limits that are built into its foundational premises—is most welcome. Of course one would expect me to feel this way, having sided with the Arlington dissenters and written my own account of how courts should restore Chevron's domain. I also think it's a good thing any time a member of the Supreme Court starts to think more seriously about the underpinnings of contemporary administrative law—something some of the justices do not do nearly enough.

UPDATE: See also Ilya Somin's post here. Unlike Ilya, I do not believe that Chevron deference necessarily entails any abdication of judicial duty. Rather, as I explained in my symposium essay on "Restoring Chevron's Domain" and my article with Nathan Sales on statutory silences and agency jurisdiction, whether Chevron is problematic depends upon how the doctrine is conceived and how it is applied. If we understand Chevron as grounded in a delegation of authority to the agency to resolve statutory ambiguities as a statue is implemented, deferring to the agency's permissible interpretation is merely following Congress's instructions. Yet if Chevron is understood this way, it means that Court's must assure themselves both that a statute is ambiguous and that Congress has delegated such authority. Thus, it is proper for courts not to defer in cases raising constitutional concerns, presenting "major questions" (as in King v. Burwell), or concerning matters that Congress cannot be presumed to have delegated to the implementing agency.

SECOND UPDATE: Some see Justice Kennedy's stated willingness to reconsider Chevron to be quite alarming. Joshua Matz, for instance, warns of potentially dire consequences should the Court revisit Chevron. Attacks on Chevron, Matz maintains, are part of an "administrative law Lochnerism" that will free judges to invalidate agency actions that fail to conform with their laissez faire preferences.

This analysis is overwrought on many levels. Among other things, it exaggerates the centrality of Chevron's role of judicial review of agency action and fails to account for either the history of the doctrine or how it fits in with other doctrines concerning agency interpretations of regulatory statutes.

While Matz presents Chevron as a doctrine liberals should love, it's worth noting that the doctrine was born in a case upholding the Reagan Adminsitration's efforts to loosen air pollution regulations under the Clean Air Act. The doctrine rose to prominence as the Reagan and Bush administrations used to to beat back broad, purpose-driven interpretations of federal statutes used to force agencies to adopt ever-tighter regulatory measures. Throughout its history, Justice Scalia was among Chevron's most ardent defenders, and Justice Breyer was among its greatest skeptics.

Were Chevron overturned, it would not mean open season on agency actions. Federal agencies would maintain a quiver-full of deference doctrines they could deploy in defense of their regulatory initiatives, including the broad deference agencies enjoy to make policy determinations, exercise enforcement discretion, reach judgments on the meaning of scientific and technical analyses and make factual determinations. None of the doctrines in these other areas would be touched by a modification of Chevron, and it would be no less difficult to demonstrate than agency action is "arbitrary and capricious" under the APA.

In addition, were Chevron narrowed or eliminated, this would leave Skidmore deference in place—the doctrine that guided judicial review of agency statutory interpretations for the forty years before Chevron was decided in 1984 (and that the Chevron court itself thought it was following). Under Skidmore, courts are required to consider an agency's views, particularly where there is reason to believe the agency has technical or other expertise in the matter. Indeed, in Skidmore itself a lower court was reversed because it had failed to give the agency's views proper consideration.

The more complex and technical the subject matter—think EPA and FCC regulatory schemes—the less likely a switch from Chevron to Skidmore is to produce a major effect on case outcomes. In other areas, however, such as labor or immigration—areas that don't require the same amount of scientific or technical expertise and where agency opinions tend to be less consistent over time—the change could be quite significant, but not in ways that would necessarily advance conservative interests. Today's decision in Periera is a case in point. The Court's refusal to give Chevron deference worked to the benefit of immigrants, and it was the conservative dissenter (Justice Alito) who wanted to defer to the government.

As I noted above, I prefer constraining Chevron to overturning it, but I also think folks like Matz exaggerated the extent to which ditching Chevron would make a profound difference or advance conservative policy priorities.