With Judge Kavanagh’s nomination pending, there has been no shortage of speculation among commentators on how his appointment might affect the Court’s jurisprudence. Of particular interest to readers of this blog, many have noted that a Justice Kavanagh would tip the Court toward greater skepticism of Chevron deference. On and off the bench, Judge Kavanaugh has expressed concern with Chevron. Most notably, in his Story Lecture to the Heritage Foundation last February he suggested that Chevron deference gives rise to troubling indeterminacy caused by the different thresholds judges use when determining whether a statute is ambiguous.

A Justice Kavanaugh skeptical of Chevron would find company on the High Court. Justices Gorsuch and Thomas have both strongly signaled that they view Chevron deference to be unconstitutional. Judge Kavanaugh’s appointment would likely empower an already existent contingent within the Court to push for restricting the doctrine—if not do away with it altogether. Given the possibility of another Chevron-skeptic being added to the Court, it is time to consider the prospect of a Chevron-less world.

So what would a future without Chevron deference look like? In a new Essay on the Stanford Law Review Online, Deference Conservation—FOIA’s Lessons for a Chevron-less World, John Brinkerhoff and I offer one answer. Inspired by legislative efforts to abrogate Chevron through the imposition of a de novo standard of review, John and I looked at an area of administrative law where courts are already explicitly required to apply de novo review: Freedom of Information Act (FOIA) litigation. We concluded that doing away with Chevron, without more, would have limited practical effect—at least in the long run.

Our reasoning is simple: “In the FOIA context, rather than deferring to agencies on matters of statutory interpretation, courts have shifted deferential treatment to other aspects of litigation.” For example, we find that courts deferred strongly to agency assertions of fact and adopted agency-friendly legal interpretations and procedural rules. “The result,” we explain, “is that the government wins 90% of the time, but the de novo standard is formally maintained.” We posit that a similar shift would occur in the APA context were the Supreme Court (or Congress) to mandate de novo review of agency legal interpretation. Judicial deference would survive, but it would manifest differently.

The experience of FOIA litigation suggests that agency deference is not easily mandated away. Where one mode of deference is foreclosed, alternative forms becomes more pronounced. We term this dynamic “deference conservation.” I consider this phrase to be a nod to Jerry Mashaw’s posited “Law of Conservation of Administrative Discretion,” the observation that administrative law can only shift policy discretion between actors, not “squeez[e]” it out entirely. So long as the application of Congress’s statutory commands requires policy choices, courts will face institutional pressures, derived principally from agencies’ relative expertise and political accountability, to leave those decisions to the agencies—and that means deference.

Put differently, institutional interests generate “hydraulic pressures . . . before which even well settled principles of law will bend.” This Holmesian adage refers to the weight of public opinion upon judges, but—based on our Essay’s observations of FOIA litigation— I suspect that it could be applied in the context of judicial deference to executive agencies. For good or for bad, deference is a product of institutional pressures. Like steam through a pipe, such pressures can by channeled but not quelled.

Deference conservation suggests that the effects of abrogating Chevron would be limited—but that does not mean such an act would be an empty gesture. John and I posit a few potential consequences:

To begin, because deference conservation is admittedly not absolute, the mere act of abrogating Chevron might send a signal to lower courts, emboldening them to be less friendly to government litigants at the margins. Similarly, agencies anticipating such a shift might themselves become less aggressive in their statutory interpretations.

But these “softer” effects might eventually fade over time as institutional pressures build. Other changes, however, might be more permanent. For example, if instead of deferring to agencies’ interpretations, courts erected higher standing requirements to shield agencies from legal challenge, fewer merits decisions would exist in the first place. Because of these, or other potential outlets for deference conservation, the full downstream consequences of abrogating Chevron are hard to predict.

With Judge Kavanaugh’s confirmation still uncertain, it may be too soon to speculate about a post-Chevron administrative state. But the deference conservation concept presented in our Essay has consequences for today. Based on our observations from FOIA litigation, we suggest that deference to agencies by the courts has less to do with doctrinal formulations than it does with the institutional—and small “c” constitutional—structures of our administrative state.

Debates about agency deference should be less about whether, say, the major questions canon is applied at Chevron Step One or Step Zero and more about the way we are distributing authority between the three branches of the federal government. In this regard, our analysis complements Nicholas Bednar and Kristin Hickman’s, who have recently argued that the anxieties driving Chevron’s critics are caused by Congress’s practice of delegating broadly to agencies, not the niceties of the Supreme Court’s deference regimes.

This critique of the ongoing Chevron debate is particularly pertinent given Judge Kavanaugh’s nomination. As discussed above, Judge Kavanaugh has focused on the fact that judges use different criteria for determining when a statute is ambiguous and thus when Chevron is appropriately applied. But if one takes deference conservation seriously, this sort of worry mistakes symptoms for causes. Deference conservation suggests that the variances with which judges find ambiguity arises from the same constitutional tensions that neutered FOIA’s de novo review provision. Tinkering with the details of the Chevron framework seems unlikely to change the administrative state dramatically. We might see percentage point differences in agency win-rates, but not much more.

Justices Gorsuch and Thomas, however, offer critiques that sound in a different register. Both have cited Philip Hamburger’s attacks on Chevron, which are grounded in the notion that agency deference amounts to a judge’s dereliction of her Article III responsibilities. If one were to take both Hamburger’s claims and the deference conservation thesis seriously, then one would be driven to the conclusion that the maintenance of our constitutional order demands a dramatic reorganization of the administrative state’s institutional structures.

Further discussion of Hamburger’s argument will have to wait another day. But I do want to suggest one tentative thought. Once we realize that agency deference is a product of the institutional structures that Congress has crafted (primarily through the use of delegation, but also other mechanisms), a connection to Congress’s control over the federal courts’ jurisdiction comes into focus. As made dramatically clear this past Term in Patchak II, Congress has the ability—within some limits—to structure the jurisdiction of the federal courts to limit judicial review selectively in order to achieve specific policy goals. If we were to conceive of deference as a constraint of judicial review—as Hamburger does—then the fact that it is ultimately grounded in Congress’s legislative choices regarding the structure of the administrative state should give us some comfort.

Understanding deference conservation allows for a reorientation of the Chevron debate, so that the focus is more squarely on questions of institutional and constitutional structure. As our Essay concludes, “[s]uch a discussion would cut more to the heart of the issue than shadowboxing over statutory ambiguities.” This would be to the benefit of both Chevron supporters and detractors, as it allows for a conversation that squarely addresses the organization of the administrative state and its legal consequences.

Daniel B. Listwa is a J.D. Candidate at Yale Law School