As a summer clerk at the Solicitor General’s office in 1983, I worked on the brief in the famous administrative law case, Chevron v. National Resources Defense Council. I had no idea then that it would become one of the most cited cases of all time. Its key holding is that courts should defer to any reasonable interpretation by an executive agency of a statute under that agency’s jurisdiction unless Congress clearly had a specific intent to the contrary. In the now-familiar parlance, under Chevron Step One, it was the judiciary’s task to determine whether the statute’s meaning is clear, but under Chevron Step Two, it was the agency’s task to interpret anything that was not clear. Thus, Chevron handed to agencies a major domain of power—that of legal interpretation over a wide range of issues.

Today, Chevron is under fire. It has not been overruled nor do I think it is likely to be. But its domain is shrinking and will continue to get smaller. If President Trump gets another term, it may well resemble the Cheshire cat—a still-powerful symbol for a body of law without much doctrinal substance. Just as Chevron was an iconic decision marking the continuing rise of the power of the administrative state, its relative decline in importance captures the three reasons that the administrative state is being cut back. Thus, this essay will use Chevron to introduce these three factors—the return of originalism, the rise of textualism, and the greater distrust of unsupervised expertise—that are transforming administrative law and, with it, the administrative state.

The Return of Originalism

The most important influence is the rise of originalism, because important elements of the administrative state, including Chevron, are in tension with the original Constitution’s separation of powers. For instance, given that the Constitution grants Congress legislative authority, it seems anomalous that the executive can wield these powers and make its own determination of what is legal through statutory interpretation. To be sure, the question of where the legislative power ends and the executive power begins is a difficult one, but that does not mean that a line cannot be drawn. For instance, Chief Justice John Marshall captured an important distinction in the Founding era when he suggested that Congress must legislate on important subjects when they affect private rights even if the executive could be left to fill in the details.

Chevron thus potentially ran afoul of the nondelegation doctrine by allowing the agencies to make major decisions about what the law meant, not just fill in the details of intricate statutes. But in the recent case of King v. Burwell, Chief Justice Roberts made clear in a majority opinion that the Chevron doctrine would not apply to major questions, such as the one in that case (whether Affordable Care Act tax credits should be provided to those purchasing insurance under federal as well as state exchanges). While Roberts suggested that Congress would not want Chevron to apply to such major questions, the more persuasive basis of the decision is the nondelegation doctrine. Applied to major questions, Chevron, as originally decided, sustains a huge delegation of legislative power to the agency, permitting it to interpret a central element of a statute.

The renewed influence of originalism on nondelegation issues is even clearer beyond Chevron cases. For instance, in United States v. Gundy, Justice Neal Gorsuch objected, on originalist grounds, to the decision by the plurality of the Court allowing Congress to delegate to the Attorney General the determination of which sex offenders convicted before the enactment of a sex-offender registration statute would nevertheless have to register after its enactment, because the statute provided no substantial limits on his discretion. And although Gorsuch’s opinion was a dissent, Justice Samuel Alito, who gave the decisive vote for the majority, wrote in a concurrence that he too would be willing to reconsider the delegation doctrine in a more appropriate case. Justice Bret Kavanaugh did not participate in Gundy, but has since indicated that he too is open to reexamining the doctrine that is fundamental to the discretionary power of the administrative state.

The Rise of Textualism

Textualism is also playing a part in shrinking Chevron. The doctrine seems to assume an intentional method of interpretation of statutes. In Chevron, Justice John Paul Stevens stated that an agency is bound by “a clear intent of Congress,” but “otherwise can provide its own reasonable interpretation.” But if textualism is used as the method of statutory interpretation, the question becomes not whether Congress had any conscious intent on a particular issue, but rather whether the text speaks to the issue. That more encompassing inquiry can reduce the scope for agency power under Chevron Step Two by having the judiciary settle more matters under Chevron Step One.

Justice Scalia’s jurisprudence exemplifies this point. While he was a defender of Chevron, he reached Step Two less than any other justice. By using formal methods of statutory interpretation, he discovered clear meanings in the text. Indeed, his last book, Reading Law, offers scores of clarifying canons for interpreting statutory text. The kind of judges appointed by Trump thus may complete a trend that Scalia himself initiated: reducing the significance of Chevron by discovering more meaning in statutes than Chevron assumed. They are likely to extend this influence throughout the judiciary.

Textualism is reshaping administrative law more generally. In Kisor v. Wilkie, Justice Kagan cut back substantially on what is called Auer deference—the deference that the judiciary is to give to an agency’s interpretation of its own regulations. She emphasized that deference was due only to “genuinely ambiguous [regulations], even after a court has resorted to all the standard tools of interpretation.” Thus, Kagan effectively endorsed a textual approach to determining ambiguity, including applying traditional canons of interpretation. It is significant that even the more liberal justices are willing to cut back on the extent of Auer deference and encourage the judiciary to first engage in a searching textual analysis of these regulations before giving any deference to the agency. Gorsuch, writing for Justices Thomas, Alito, and Kavanaugh, would have gotten rid of any fixed domain of deference for agency interpretation of its own regulations.

Distrust of Unsupervised Expertise

Finally, confidence in the unsupervised expertise of agencies has waned since the apogee of the administrative state, and this sentiment has also reshaped Chevron. Since Chevron, the Court has held that deference should not apply to casual determinations of statutory meaning by an agency, but only to ones made under formal conditions that received the blessing of agency heads, as in notice-and-comment rulemaking.

The Court has been less willing to defer to the expertise of bureaucrats in other contexts as well. In Department of Commerce v. New York, the Court held that President Trump’s Secretary of Commerce had the authority to reject the opinions of the experts in the census office that adding a citizenship question to the census form would fail to improve, and possibly undermine, the accuracy of the Census Bureau’s count of citizens. So long as the Secretary gave any plausible reasons for the rejection of his experts’ advice, the Court concluded that his decision should be upheld because Congress had delegated administrative authority to the Secretary and not to the census experts. (The Court concluded that the reason the Secretary gave for the change was pretextual and thus invalidated his decisions, but would have upheld him had his reason been genuine.)

In Free Enterprise Fund v. Public Accounting Oversight Board, the Court recently insisted that the members of an oversight board appointed by the Securities and Exchange Commission must be removable at will by the SEC, itself a commission that the Court assumed was insulated from removal by the President except for good cause. The Court reasoned that the double insulation from removal provided by statute was inconsistent with the political control that the Constitution demanded. As the Court stated “One can have a government that functions without being ruled by functionaries, and a government that benefits from expertise without being ruled by experts. Our Constitution was adopted to enable the people to govern themselves, through their elected leaders.” Here two factors—a return to originalism (and the unitary executive it contemplates) and skepticism about unsupervised expertise—pointed in the same direction.

So long as the five-member formalist majority holds sway on the Supreme Court, these three factors will combine to move toward cabining the power of the administrative state. If President Trump is reelected and appoints more justices like Gorsuch and Kavanaugh, these trends will accelerate.