A Revolution in Administrative Law

Richard Epstein*

Richard Epstein

Richard Epstein

One of the most vital, but technical, items on the Republican agenda is not likely to get its fair share of public attention. On the first day of the new Congress, Republican Representative Bob Goodlatte of Virginia introduced two pieces of legislation that could fundamentally alter the structure of American administrative law for years to come.

The first bill, H.R. 26, the Regulations From the Executive In Need of Scrutiny (REINS) Act, languished in three previous successive Congresses, but it has now cleared the House by a vote of 237-187. If enacted, this bill will give Congress a final say on regulations with an estimated cost over $100 million through a mandatory up-or-down vote before they go into effect. More importantly for the day-to-day operation of administrative law is Goodlatte’s other bill, H.R. 5, the Regulatory Accountability Act, which the congressman claims will “wipe out abusive regulation—freeing Americans to innovate and prosper once more.”

This last claim is a tall order for any one piece of legislation. But there is no underestimating the effect H.R. 5 will have in shaping the agendas that Scott Pruitt will bring to the Environmental Protection Agency, Betsy DeVos to the Department of Education, Andy Puzdur to the Department of Labor, and Rick Perry to the Department of Energy—each of whom will administer a complex set of statutes that authorize their agency actions while delimiting their authority. In the Obama administration, there were profound political clashes between his progressive administrators, who constantly sought to extend the scope of their authority in all of these areas, and their opponents both on and off the bench.

The great challenge of H.R. 5 is to rethink a system of administrative law that was ushered in by the passage of the Administrative Procedure Act of 1946 in the aftermath of World War II. The APA has a quasi-constitutional status because it seeks to rationalize judicial oversight of administrative agencies that received vastly greater power after the New Deal constitutional revolution. This expansion came to fruition during the October 1936 Supreme Court term, during which the Court first increased the scope of congressional power by allowing it to regulate in virtually any area of economic or social importance, and second, weakened the individual rights afforded to private property and to economic liberties in the pre-New Deal Constitution. A decade later, Congress passed the APA to cut back on the aggressive implementation of legislation through regulation and enforcement actions instituted by the new generation of New Deal agencies. These transformative laws dealt with such core matters as labor, agriculture, communications, and securities law.

One essential provision of the APA, Section 706, regulates the interaction between the agencies and the courts. It states in part: “To the extent necessary to decision and when presented, the reviewing court 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.” In addition, Section 706 provides that the reviewing court shall “hold unlawful and set aside agency action, findings, and conclusions found to be—(A) arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.”

Under the APA’s sensible original design, a reviewing court stands in relation to the administrative agency much as an appellate court stands in relation to a trial court. On all questions of law, the appellate court makes its own interpretation of the relevant legal materials in deciding the meaning and scope of a particular provision. On questions of fact, an appellate court tends to defer to the trial court decision. Finally, an important class of cases deals with mixed questions of law and fact—e.g., whether the basic facts establish negligence by the defendant.

This time-tested regime reflects the application of the principle of comparative advantage by the different layers involved in judicial administration. Triers of fact often have to examine many witnesses and documents, for which they develop the appropriate expertise on such key issues as credibility and weight. Giving them latitude speeds up the process of adjudication, reduces the burden on appellate judges, and prevents the introduction of a fresh round of factual errors by appellate judges. At the other extreme, reading statutes and regulations is what appellate courts do well, and these questions can be resolved by standard techniques of interpretation that carry over from one substantive field to another. Courts typically apply an intermediate standard of judicial review to those difficult mixed questions of law and fact.    

For many years, the APA more or less kept true to its original conception, which treated administrative agencies as if they functioned like trial courts. By the 1970s, however, it became clear that the courts, most notably the liberal Circuit Court for the District of Columbia, had adopted an activist role in the overall process, particularly in their comprehensive review of the factual record. That Court was slapped down hard in 1978 by a young Justice Rehnquist in Vermont Yankee v. National Resources Defense Council, Inc. for introducing added layers of complexity, especially in dealing with the approval of new nuclear power plants, into the relatively lean structure of the APA.

But the truly great transformation of American administrative law took place in two major decisions of the 1980s, both of which changed administrative law for the worse, by inverting the relationship of review of facts and review of law. The first became subject to a hard-look doctrine, and the second to extensive judicial deference.

On the former, in the 1983 Supreme Court decision Motor Vehicle Manufacturers Association v. State Farm Mutual Automobile Insurance Co., the Court, speaking through Justice Byron White, read the arbitrary and capricious standard strictly to require the National Highway Transportation Safety Administration to give a “hard look” to the NHTSA’s decision to postpone the implementation of passive restraint devices in automobiles. The Court insisted, among other things, that “an agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.”

Under this standard any complex factual determination becomes fair game for an intensive review, which could be used not only to block new regulations but also to block efforts to roll back existing regulations on the ground that they were too onerous. The correct position is one that, as Section 706 suggests, shows some degree of deference to agency decisions. Otherwise no factual determination can be put into place, which is one reason why the nuclear power industry in the United States has been stopped dead in its tracks by judicial oversight.

In my view, the Goodlatte legislation goes in exactly the wrong direction when it proposes a grotesquely convoluted system of factual review under the grandiose claim that the law “restores to the people the true right to be heard by Washington’s regulators.” But “the people” also includes every special interest group that wishes to obstruct sensible reform, as in State Farm itself, as well as those, like the new cadre of Trump administrators, who have a more deregulatory frame of mind. There are many tasks of administration that need to be done and the Goodlatte labyrinth could bring the entire administrative process to a halt, killing off good projects as well as bad ones. No system of appellate review can neutralize incompetent regulators without hamstringing competent ones in the bargain. This portion of the legislation should be withdrawn and reengineered.

Nonetheless, H.R. 5 is on the mark with a far shorter and far more coherent provision that it styles the Separation of Powers Restoration Act (SOPRA), which in 2016 had been sensibly introduced as a freestanding bill. Its key provision reads that any court reviewing administrative action shall “decide de novo all relevant questions of law, including the interpretation of constitutional and statutory provisions, and rules made by agencies.” “De novo” review means that the reviewing court gives no deference to the legal opinions of either the parties or lower court judges and administrators.

This compact and straightforward provision, which should be promptly enacted, takes aim at two of the most misguided decisions of administrative law that instructed courts to take a deferential stance toward agency actions interpreting the key statutes and regulations they administer. The first of these cases, Chevron USA Inc. v. NRDC (1984), written by Justice John Paul Stevens, insisted that in all ambiguous cases, reviewing courts should defer to an agency interpretation of its governing statute. Auer v. Robbins (1997), written by the late Justice Antonin Scalia, similarly held that for an agency’s “own regulations, [its] interpretation of it is, under our jurisprudence, controlling unless ‘plainly erroneous or inconsistent with the regulation.’”

The current regime of deference enjoys strong bipartisan support even though it is plainly inconsistent with Section 706, which provided for the de novo review called for in SOPRA. The change is long overdue. It is difficult to describe in a short space the enormous doctrinal refinements of both Chevron and Auer. No one knows quite when any given statute or regulation becomes ambiguous, and there is a futile debate as to whether the same level of deference should be afforded to memoranda and opinion letters as to regulations issued after the more extensive notice and comment procedures the APA calls for in dealing with major regulations. Scholars thus talk about Chevron step zero, step one, and step two in an effort to decide just how much deference is required in any given context.

But the implicit assumption that all agencies are neutral and dispassionate enforcers of their own statutes is falsified by the day-to-day actions of agencies. Their heads have strong political agendas to advance on key issues such as labor, education, and environment, all flash points in the Obama administration. The acceptance of high levels of deference lets agencies make hash out of statutes, which is what happened, for example, in the Department of Labor’s ruling that the statutory prohibitions against sex discrimination, passed decades before, meant to apply to modern gender identity cases that no one had even imagined at the time. Deferring to agencies invites huge flip-flops with the change of administration, given their radically different views of how these various relationships could be shaped. It also allows agencies like the Federal Communications Commission to expand their jurisdiction with dubious interpretive strategies, which often upsets the balance initially designed by Congress.

That potent combination of intellectual incoherence and institutional instability makes it hard for private businesses and organizations to plan for the future. Concerns for their plans could be effectively moderated if these legal issues were simply decided by courts as they have been for centuries.

The fruits of this confusion will become evident the day after the Trump inauguration when major agencies reverse field, an action requiring as little justification as the process that adopted these (dubious) rules in the first place. It is often said about administrative agencies that only they possess the expertise to make sense out of arcane statutory and regulatory language. Nothing could be further from the truth. Most of the really important language consists of words like discrimination and pollution that judges can understand and apply as well as anyone else. The great danger of entrusting these issues to administrative agencies lies in their inherent bias. The independent agencies are in most cases run by commissions of five members, which often divide three-to-two, with the President appointing a chairman from his own party. Bias in administration is not some rare Black Swan event, but is, as standard public choice theory predicts, a common occurrence.

This is why the Democrats in Congress are so fiercely opposed to the key Trump appointments mentioned above. Now that they are out of power, the Democrats, who once praised executive and administrative power, will sound the rule of law trumpet loudly. And they should be heard on this point. H.R. 5 is important now because it can be used to restrain the abuse of discretion of Republican administrators, just as it can be used to restrain the abuse of Democratic ones.

For the moment at least, the proposed revolution in administrative law will have its greatest impact on the regulatory misadventures of the highly partisan Obama agencies. But in the long run, we need SOPRA to make sure that Republican administrators do not return the favor by hiding behind Chevron and Auer deference in order to impose, by regulation and interpretation, their own dubious decisions. Whether one deals with constitutions, statutes, or regulations, the only safe guide is to try to use text, context, structure, and purpose to determine the correct interpretation of disputed provisions. Gimmicks like Chevron and Auer deference only muddy the waters. In all cases, de novo review of questions of law is a matter of the highest importance for constitutional and institutional safeguards of the rule of law. We need it today.

*Considered one of the most influential thinkers in legal academia, Richard Epstein is known for his research and writings on a broad range of constitutional, economic, historical, and philosophical subjects.