The modern administrative state creates a pervasive system of consolidated power that weakens the checks and balances carefully constructed by the United States Constitution. Acknowledging the pressing need to impose legal constraints on the administrative state, Congress enacted the Administrative Procedure Act (APA) in 1946. While the APA has remained largely untouched by Congress since then, courts have interpreted the act’s requirements in a manner that is in tension with the text and original understanding of the law.
This tension between the text and purpose of the APA on the one hand and the contemporary judicial application of the law on the other is perhaps most significant in the realm of judicial review of an agency’s legal interpretations. The APA expressly ordered the courts to review de novo any and all administrative interpretations of law. That directive is consistent with the long-standing common-law principle that courts must decide all legal issues independently. Two doctrines, conventionally known as “Chevron Deference” and “Auer Deference,” require courts to defer to administrative interpretations of law in a goodly number of cases rather than decide all issues of law de novo. Chevron requires deference to agencies’ interpretations of statutes, and Auer requires the same (if not even greater) judicial deference to agencies’ interpretations of rules. Both decisions exacerbate the problems of consolidated power created by the post–New Deal administrative state.
Each doctrine raises the same broad issue: whether administrative power is above the law, unchecked by courts exercising their constitutional responsibility to interpret and apply the law. Chevron and Auer are unsupported by basic principles of American constitutionalism. They are inconsistent with the relevant statutory law and with presumptions about Congress’s intent to delegate power. The courts also apply them inconsistently. For all of these reasons, Congress should consider amending the APA to end this deference regime.
The pitfalls of deference have become increasingly apparent to federal judges, some of whom have explicitly questioned the rationale for Chevron and Auer deference. Justice Neil Gorsuch, for instance, criticized Chevron deference harshly in an opinion written while he was a judge on the U.S. Court of Appeals for the Tenth Circuit, asserting that the doctrine “seems to have added prodigious new powers to an already titanic administrative state…. It’s an arrangement, too, that seems pretty hard to square with the Constitution of the founders’ design.” Others, including Chief Justice John Roberts, Justice Clarence Thomas, and recently retired Justice Anthony Kennedy, have echoed Justice Gorsuch’s concerns. Given the number of justices troubled by the Chevron and Auer deference rules, there is more than a small chance that the Supreme Court of the United States will eventually reconsider those decisions.
Congress, however, does not need to wait for the Supreme Court to correct the constitutional problems that Chevron and Auer deference raise. By amending the APA to affirm the judicial responsibility to interpret the law, Congress can eliminate any threat that these deference doctrines pose to basic constitutional principles.
Deference Is Inconsistent with Constitutionalism
The most fundamental threat that judicial deference poses to constitutionalism is its inconsistency with the judicial power itself. Article III of the U.S. Constitution vests “The judicial power of the United States” in the Supreme Court and the inferior federal courts that Congress chooses to establish. It empowers the courts to decide “all cases, in Law and Equity, arising under this Constitution, the Laws of the United States, and Treaties made…under their Authority.” This judicial power is a responsibility to exercise independent judgment in cases where courts must interpret the law. In one of the most famous cases decided by the Supreme Court in the early days of the Republic, Chief Justice John Marshall explained that “it is emphatically the province and duty of the judicial department to say what the law is. Those who apply the rule to particular cases must, of necessity, expound and interpret that rule.” In other words, the power to decide cases and controversies necessarily includes the power to interpret the law, since the act of applying the law entails the act of interpreting it.
That approach to judicial review was hardly a novel one. Before Parliament became a modern-day legislative body in England, the common-law courts were the predominant source of law. They developed the law of contracts, torts, property, and crime through the case-by-case adjudicatory process by which common-law decision-making became known throughout the Western world. In so doing, judges independently decided not only what the correct answer to a dispute should be, but also who should make that call: the courts. It was not up to the Crown—such as King John—or one of a king’s lieutenants—such as the Sheriff of Nottingham—to decide what was the law; that was the court’s job. In fact, the whole purpose of the Magna Carta was to make it clear that the king and the entire English government were subject to the law, a principle that today is known as the “rule of law.”
The colonists carried English law with them to the New World, and the colonial American courts followed the English common law. In fact, colonists saw the common law as their English birthright. In each case, the common law was the set of “principles, usages, and rules of action applicable to the government and security of persons and property, which do not rest for their authority upon any express and positive declaration of the will of the legislature.” Organizers of the colonies had to grant settlers not only the right or opportunity to own property, but also the rights of the Englishmen who remained behind, as reflected by guarantees in the colonial charters. Accordingly, Marshall’s declaration in Marbury that it was the responsibility of the courts to articulate the law was the direct outgrowth of the type of common-law decision-making that characterized the work of all of the courts that preceded the Supreme Court.
The power to interpret the law is thus a power clearly implied in Article III’s vesting of the judicial power in the federal courts. It is also a power that federal courts must exercise independently, without deference to the other departments of the government. The Constitution’s Framers designed the judiciary carefully to ensure its independence, protecting judges’ salaries and their tenure from interference by the other branches. While courts can give weight to the way other departments construe the laws, they have the duty to render their own judgment when interpreting and applying the law.
Throughout the first century of American history, courts understood and took this responsibility seriously. Although the record is somewhat murky, on the whole, judges refused to defer to administrative interpretations of law. As one scholar recently wrote, “there was no rule of statutory construction requiring judicial deference to executive interpretation qua executive interpretation in the early American Republic.”
Consider an issue that commonly arose during the 19th century. As Americans moved west, they sought to obtain land owned by the federal government. When the U.S. Land Office granted land patents to claimants, rival claimants frequently challenged those patents in common-law ejectment suits. In deciding these cases, judges would have to settle legal issues, such as whether the Land Office lacked jurisdiction to issue the patent. When they had to interpret the relevant statutes to settle those questions, judges did so without deference to an executive officer’s decision.
Deference Is Inconsistent with the Administrative Procedure Act
The Chevron and Auer doctrines are also inconsistent with the most directly relevant act of Congress dealing with judicial review of administrative decision-making: the Administrative Procedure Act. The APA explicitly mandates that “reviewing court[s] shall decide all relevant questions of law, interpret constitutional and statutory provisions, and determine the meaning or applicability of the terms of an agency action.” The straightforward language of the APA, in other words, provides for independent, de novo judicial review of agencies’ interpretations of statutory provisions, as well as of the meaning or applicability of an agency’s regulations. It therefore seems to forbid both Chevron and Auer deference.
Both the Administrative Procedure Act’s historical context and the legislative debate surrounding the law lend support to this reading of the statute. Judicial review of agencies’ legal interpretations was largely non-deferential throughout the 19th century, but by the 1940s, scholars had introduced confusion regarding the validity of the distinction between law and fact. The law had historically distinguished between questions of fact (“Who owns Blackacre?”) and questions of law (“What does fee simple ownership of Blackacre mean?”) and had deferred to an agency’s fact-finding decisions but not to its legal rulings. Early in the 20th century, however, legal theorists questioned the legitimacy of this distinction and sought to extend deference to both categories. As John Dickinson, a prominent Progressive legal theorist, argued:
[T]he distinction between “questions of law” and “questions of fact” really gives little help in determining how far the courts will review; and for the good reason that there is no fixed distinction. They are not two mutually exclusive kinds of questions, based upon a difference of subject matter. Matters of law grow downward into roots of fact, and matters of fact reach upward without a break, into matters of law.
As a result of that botanical view of legal decision-making, administrative law in the 1930s and 1940s shifted to a different model in which administrative interpretations of both law and fact received deference from reviewing courts. The passage of the APA in 1946 was in part a reaction against that trend and an attempt to restore the earlier distinction, preserving independent judicial review of agency legal interpretations and deferential review of agency fact-finding. The APA was the result of several years of constitutional resistance to the administrative state and was generally understood to be a first step in putting limits on its power and discretion. On the floor of the House of Representatives prior to the law’s passage, Francis E. Walter (D–PA) clarified that Section 706 “requires courts to determine independently all relevant questions of law, including the interpretation of constitutional or statutory provisions.” In the Senate, Patrick McCarran (D–NV), who shepherded the bill to passage, confirmed that “the principal purpose of the bill is to allow persons who are aggrieved as a result of acts of governmental agencies to appeal to the courts.”
In short, throughout the debates over provisions of the APA, there was broad agreement in favor of judicial review of administrative action and the need to preserve and restore judicial review. As University of Virginia Law Professor Aditya Bamzai summarizes, “Read against the history of the APA’s adoption, section 706 is best interpreted as an attempt…to instruct courts to review legal questions using independent judgment.” Chevron deference, therefore, is contrary to the APA’s clear statutory mandate that judges apply their own independent interpretation of the law rather than defer to an agency’s.
Chevron Rests on a Fictional Congressional Intent
An oddity about the Chevron doctrine is the date of its birth: 1984, nearly four decades after the APA became law in 1946. The APA was “a new, basic and comprehensive regulation of procedures in many agencies” that “settl[ed] long-continued and hard-fought contentions, and enact[ed] a formula upon which opposing social and political forces have come to rest.” The APA therefore was of surpassing importance to the proper performance of judicial review.
Given that, we would have expected that if agencies were still to receive deference despite the quite explicit text of Section 706 to the contrary, the lower federal courts would have said something to that effect long before 1984. But they did not. In the decades after 1946, the Supreme Court decided a host of landmark cases involving the proper scope of judicial review under the APA, such as Citizens to Preserve Overton Park v. Volpe and Vermont Yankee Nuclear Power Corp. v. Natural Resources Defense Council. Yet none of those decisions hinted at the rule later created in Chevron. That raises the question of whether this is yet another example of the dog that did not bark.
That judicial deference to agencies’ legal interpretations runs contrary to both the text of the Constitution and the text of the APA certainly helps to explain why the Chevron doctrine did not emerge for decades and why, when it did, it arose in a case that did not involve the APA. Chevron involved the Clean Air Act’s regulation of “stationary sources” of pollution. The opinion, as is now widely understood, never intended to set forth a doctrine or lay down new principles regarding judicial review of agency legal interpretation. The Chevron doctrine, in other words, emerged not from the APA or the Constitution, or even from the Chevron opinion itself, but as a result of a struggle in lower courts during the mid-1980s “that converted a narrow Clean Air Act case about imaginary bubbles over factories into a generalized doctrine of administrative law.” In other words, the Supreme Court did not deliberately create the Chevron deference doctrine. The Court stumbled into it unwittingly.
Given such a shaky foundation, it should not be surprising that the rationale for the Chevron doctrine is still unclear. The Chevron opinion itself suggested many reasons for judicial deference to administrative interpretations of law, including the complexity of the issue, the thoroughness of the agency’s reasoning, agencies’ subject-matter expertise, agencies’ political accountability, and Congress’s intention to delegate interpretive power to the Administration. Supporters of a broad reading of Chevron, however, argue that deference is grounded in congressional intent. Because Congress delegated the power to the agency to interpret the law, by leaving an interpretive gap in the statute, the argument goes, Congress must have intended that the agency would fill in the gap. Deference to the agency’s interpretation, therefore, is merely a grant of deference to Congress, which intended that the agency fill in the statutory gap.
This presumption of congressional intent does not accord with reality. Ambiguity in a statute can be the result of a number of factors, and Congress’s intent to grant an agency interpretive power is among the least likely. As political scientists have explained—and the Supreme Court has acknowledged—Congress is a multitudinous body composed of many different representatives and different interests, from which it is typically impossible to derive a singular intent. Consequently, few scholars believe that ambiguity in statutes passed by Congress indicates any intent to grant interpretative power to administrative agencies.
Deference Undermines the Rule of Law
Chevron, therefore, rests on no constitutional, historical, or statutory ground. The Court justified its rule on the basis of a fictional congressional intent—what judges would call a “lie” if someone else uttered it—that is widely rejected. Perhaps that is why courts frequently ignore or circumvent the doctrine in important cases. Ironically, the fact that courts apply the doctrine so inconsistently undermines the final argument in favor of judicial deference: that it reduces judicial interference in the administrative process and thereby promotes stability and the rule of law. In practice, the inconsistent application of judicial deference doctrines actually undermines stability and predictability.
As one scholar has explained, the courts do not follow Chevron consistently. Just deciding whether Chevron applies has befuddled the courts. Consider what D.C. Circuit Judge Brett Kavanaugh had to say on this point:
[W]hen the text of the statute is ambiguous rather than clear, judges may resort to a variety of canons of construction. These ambiguity-dependent canons include: (1) in cases of textual ambiguity, avoid interpretations raising constitutional questions; (2) if there is textual ambiguity, rely on the legislative history; and (3) in cases of textual ambiguity, defer to an executive agency’s reasonable interpretation of a statute, also known as Chevron deference.
Here is the problem. And it is a major problem. All of these canons depend on a problematic threshold question. Courts may resort to the canons only if the statute is not clear but rather is ambiguous. But how do courts know when a statute is clear or ambiguous? In other words, how much clarity is sufficient to call a statute clear and end the case there without triggering the ambiguity-dependent canons?
Unfortunately, there is often no good or predictable way for judges to determine whether statutory text contains “enough” ambiguity to cross the line where courts may resort to the constitutional avoidance doctrine, legislative history, or Chevron deference.
In my experience, judges often go back and forth arguing over this exact point. One judge will say, “The statute is clear; that should be the end of it. Case over.” The other judge will respond, “I think the text is ambiguous,” meaning that one or another canon of construction should be employed to decide the case. Neither judge can convince the other. And that’s because there is no objectively right answer.
In recent high-profile cases involving agencies’ interpretations of statutes, the Supreme Court either has ignored Chevron altogether, as in King v. Burwell (a decision upholding the constitutionality of the Affordable Care Act), or has substituted its own reading for that of the agency in spite of Chevron’s application, as in Michigan v. EPA (a decision invalidating an Environmental Protection Agency regulation for mercury emissions from power plants).
This judicial sidestepping of deference has been ongoing for years. In 2000, for example, the Court decided FDA v. Brown & Williamson Tobacco Corp., a case involving the FDA’s interpretation of the federal Food, Drug, and Cosmetic Act. The Court did not defer to the FDA but construed the statute itself, concluding that the FDA’s regulation of tobacco products was unauthorized. Moreover, in a separate line of cases, the Court has whittled away the scope of agency interpretations that qualify for Chevron deference in the first place.
The muddle that Chevron and its progeny have created has led one writer to conclude “that courts’ unfettered discretion to decide whether to follow Chevron’s framework results in arbitrary and unpredictable decisions about Chevron’s applicability.” Consequently, “exceptions to Chevron have begun to swallow the rule.” Given the confusion that currently reigns in this area of the law, the best way to restore stability and predictability is for Congress to settle matters by revising the law.
In short, the inconsistent application of Chevron prevents the rule from being justified on the ground that it creates stability and predictability. In addition, both Chevron deference and Auer deference promote a combination of powers that allow administrative agencies to both make and interpret law. This combination of powers also undermines the rule of law.
Auer, which requires courts to defer to administrative agencies’ interpretations of their own regulations rather than interpretations of congressional statutes, most obviously creates these problems. Scholars have long noted the constitutional problems raised by Auer, suggesting that separation of powers considerations foreclose combining the power to make and interpret law in any one agency. More recently, several members of the Supreme Court have echoed these concerns. In Perez v. Mortgage Bankers Association, decided in 2015, Justice Antonin Scalia wrote that “there are weighty reasons to deny a lawgiver the power to write ambiguous laws and then be the judge of what the ambiguity means.” His solution was to “abandon Auer and apply the [APA] as written” so that an agency could still be “free to interpret its own regulations…but courts will decide—with no deference to the agency—whether that interpretation is correct.”
While many have identified the clear separation of powers concerns raised by uniting lawmaking and interpretive power in administrative agencies, supporters of Chevron will maintain that this criticism is beside the point. After all, in cases involving administrative interpretations of statutes, the argument would go, the lawmaker is Congress. But the Supreme Court has applied Chevron in cases where Congress did not expressly empower an agency to define a statutory term or resolve a problem by applying the law to the facts—such as Chevron itself. Atop that, administrative agencies actually participate in a good deal of statute writing as well. One recent study found that “agencies provide technical drafting assistance on the vast majority of the proposed legislation that directly affects them and on most legislation that gets enacted.” Therefore, even in the context of administrative interpretations of statutes, the same body is often the lawmaker and the interpreter, raising the same issues that are raised in the context of Auer.
All of the reasons typically offered in favor of judicial deference to agencies’ legal interpretations are unpersuasive. Neither the Constitution nor the Administrative Procedure Act offers support for judicial deference. There is no long-standing historical support for deference. The notion that Congress intends to grant agencies the power to interpret ambiguous statutory provisions is widely considered to be—to put it politely—a fiction. In addition, deference fails to promote stability, consistency, or the rule of law.
The Consequences of Deference
Does it really matter whether or not Chevron deference exists? Given the inconsistent application of Chevron by the Supreme Court, we may wonder whether anything would change if Congress eliminated it. If administrative agencies have subject-matter expertise, perhaps judges would continue to defer to them in the absence of Chevron deference, and this reform would be much ado about nothing.
Empirical work, however, indicates that Chevron matters immensely to two specific actors: judges below the Supreme Court and administrators themselves. One recent survey of agency officials from several different departments indicated not only that they know about deference doctrines, but that agencies can be more aggressive in their interpretations of law if they will receive deference. Administrators admit, in other words, that the extent to which they receive deference affects their behavior. This confirms what a former attorney in the Office of Legal Counsel has written: “I can confirm from my own experience the accuracy of Chief Justice John Roberts’s observation in his dissenting opinion in City of Arlington v. FCC: ‘Chevron is a powerful weapon in an agency’s regulatory arsenal.’”
In the federal judiciary as well, deference doctrines matter immensely in terms of the win rates for administrative agencies in litigation. In the federal courts of appeals, administrative agencies prevail in cases where Chevron applies much more frequently than they do when it does not apply, with a 25-percentage-point difference in their win rate. Although the Supreme Court applies Chevron inconsistently, the courts of appeals appear to be more deferential in their application of the doctrine.
This confirms what administrative officials admit: Judicial deference to their legal interpretations matters, and it enhances the power of administrators. Congress should eliminate these deference doctrines to restore balance between agencies and the people those agencies regulate.
The Alternative to Deference
Defenders of Chevron and Auer often reply that doing away with these deference doctrines would produce two harmful consequences. First, it would undermine political accountability by injecting the judiciary into decisions about policy that are better left to the political branches. Second, they argue, it would introduce chaos by substituting for the clear deference rules of Chevron and Auer an unpredictable array of factors from which courts could choose in deciding whether to uphold agency decisions. Both of these objections raise the question of what the alternative to deference might look like in practice.
Fortunately, because eliminating Chevron and Auer deference would merely return us to administrative law doctrines that prevailed for more than a century, we have ample historical practice to aid us in anticipating the consequences of doing so. That practice suggests that courts would continue to grant significant weight to agencies’ interpretations of law. Judges understand that administrators have expertise in the subjects within their jurisdiction, and they consider administrators’ justification for their interpretations when reviewing an agency action. Indeed, some have suggested that Chevron is best understood as part of “administrative common law,” based not on any command of Congress at all, but rather on a sensible respect for the opinions of knowledgeable experts on the matters they know best.
Ending these deference doctrines, in other words, would not result in freewheeling judicial policymaking in which judges aggressively substitute their own judgments for that of agencies. Judges would continue to respect and weigh the arguments of experts without abdicating their responsibility to interpret the law. Moreover, there are reasons to suspect that federal judges may be reluctant to take on the responsibility for interpreting the vague statutes that are emblematic of the modern administrative state. Judges are likely to continue to give weight to the opinions of experts in administrative agencies (as they should) and are also likely to pause before filling in the statutory gaps that Congress leaves in most modern regulatory statutes.
The Chevron and Auer deference doctrines raise serious constitutional concerns, are inconsistent with the Administrative Procedure Act, and have little basis in American legal history. They threaten to disrupt the separation of powers by transferring judicial power to executive officials under the fictional presumption that Congress intends to give administrative agencies power to interpret the law whenever it creates a statutory gap. Judges do not apply the Chevron framework consistently, and this leads to confusion about the level of deference that agencies actually receive in practice.
For all of these reasons, it is time for the misguided deference regime to end. Some scholars speculate that the Supreme Court might limit or overturn both Chevron and Auer in the near future, given the reservations expressed by many justices in recent years. But Congress does not have to wait for the Supreme Court to revisit Chevron and Auer. It can and should put an end to these deference doctrines itself.
—Joseph Postell is a Visiting Fellow in the B. Kenneth Simon Center for Principles and Politics, of the Institute for Constitutional Government, at The Heritage Foundation. He is also an Associate Professor of Political Science at the University of Colorado at Colorado Springs. Paul J. Larkin, Jr., is the John, Barbara, and Victoria Rumpel Senior Legal Research Fellow in the Edwin Meese III Center for Legal and Judicial Studies, of the Institute for Constitutional Government, at The Heritage Foundation.