Judicial deference to the executive branch is uniquely inappropriate when the government seeks to deport or otherwise deprive people of physical liberty.
Richard A. Epstein, a self-described “classical liberal,” may best be characterized as a man of the 19th century—and, hence, a man of my heart.
One of my first essays in zee Englisch langvitch was a book review of Professor Epstein’s Takings, published in Irving Kristol’s Public Interest—respectful, but gently critical. We’d never met but a few months later I introduce myself at a Federalist Society gig and he says, “Oh so you’re the character who wrote that? Let me explain this to you.” Which he proceeded to do until I surrendered of sheer exhaustion. So began an enduring, rewarding friendship. I submit this review of my dear friend and mentor’s latest book, The Dubious Morality of Modern Administrative Law, with confidence that he’ll explain this to me as well.
The book title, of course, invokes Lon L. Fuller’s famous account of The Morality of Law—in here-relevant part, an explication of the minimum conditions a legal system must satisfy, at least most of the time, to be called “legal” in a moral or rule-of-law sense. Such a system, Professor Fuller argued, must rest on rules, instead of ad-hoc decisions. The rules must be publicized; prospective (not retroactive); understandable; consistent; observable; and reasonably stable over time. And there must be a rough congruence between the rules and actual administration.
In a 2018 law review article, Harvard Law Professors Cass R. Sunstein and Adrian Vermeule brought those standards to bear on administrative law. They concluded that modern administrative law satisfies Fullerean standards most of the time, though not unfailingly; that the federal courts have over time superimposed judicial doctrines that are not easily traced to positive legal norms and are best understood as a kind of institutionalized Fullerean gut check (my term, not theirs); and that conservative critiques of “the administrative state” appear in their best light if they are understood as reflecting Fullerean sensibilities, as distinct from supposedly hard but eventually elusive constitutional rules. Professor Epstein’s book is a sustained, multi-pronged, and to my mind compelling response to the Sunstein-Vermeule argument.[*]
Lon Fuller, Professor Epstein argues, omitted crucial rule-of-law conditions, especially the need for an impartial judge. At variance with Fuller, moreover, and on a Hayekian note, Professor Epstein argues that formal rule of law constraints work best in the context of a classical-liberal regime that rests on property rights, freedom of contract, and protection against uncompensated takings. Once those substantive commitments go by the boards, procedural rule-of-law requirements are bound to give way as well and, for that matter, may not be worth very much. Professor Epstein does not simply assert that point; he demonstrates it. The famous (or infamous) Chenery litigation of the 1940s, for example, eventually generated the bedrock principle that the choice between rulemaking and adjudication is left to the agency’s virtually unreviewable discretion, meaning that an agency can impose novel rules and liabilities by means of adjudication. That decision prompted a spirited Fullerean dissent by Justice Robert Jackson. Surprisingly, though, but to my mind convincingly, Professor Epstein argues that this Chenery II rule may actually be right—in a fashion. The SEC, he notes, retroactively imposed never-before-heard-or-thought-of “fiduciary duties” upon corporate officers and shareholders. That radical departure from private orderings and common-law conventions “was wholly unwarranted whether done by common law judicial decision or by administrative regulation” (italics in original).
At the same time, Professor Epstein inveighs against the contemporary academic administrative law debate—not just Professors Sunstein and Vermeule but in fact most, if not quite all, participants in that debate. The “morality of administrative law,” he writes, survives only until you look at the real world and discover that it violates every Fullerean constraint. The defenders of the current order, and for that matter many of its critics, suffer from “a single systematic error—the high level of abstraction […] [T]he constant exchanges in scholarly journals about applicable standards [of judicial review] taken out of context do nothing to inform the debate.” Here, too, Professor Epstein is not content to just assert the point but demonstrates it. For example, he shows that the abstract non-delegation doctrine—delegation is okay so long as Congress articulates an “intelligible principle”—meant one reasonably limited thing in the pre-New Deal tariff cases, and a very different thing in the “anything goes” cases of the post-New Deal era.
In the course of this discussion, Professor Epstein makes a few surprising and perhaps doubtful concessions to the modern administrative state. For example, he embraces the ultimate “anything goes” delegation decision in Yakus v. United States (1944), which held that the government may criminalize and prosecute wholly private conduct (the sale of meat at market-clearing prices) without ever giving the accused a chance to contest the administrative rule under which he is convicted. (In fairness, Professor Epstein concedes that “much can be said” on the other side, as indeed it has.) That said, Professor Epstein’s general argument strikes me as very importantly right. I, too, am convinced that formal rule-of-law constraints will have little purchase under a government that recognizes no substantive limits to its ambition. (Professor Epstein passionately inveighs against the sheer insanity, and tyranny, of the Obama administration’s Title IX “guidelines” and “Dear Colleague” letters. There is simply no way of micro-managing students’ dating behavior or high school bathroom arrangements within Fullerean constraints.) And, like Professor Epstein, I have had it with the angels on Chevron’s pinhead. “The proper study of administrative law,” Professor Epstein writes, “must connect abstract doctrines with the concrete realities of individual cases, their facts, and the applicable statutory and regulatory frameworks.” Amen.
For all that, I suspect that some readers—especially readers who are persuaded that the morality of the administrative state is, well, dubious—may be left wondering where, at the end of the day, Professor Epstein intends to take his stand in the raging administrative law debate.
The current debate over administrative law, Professor Epstein writes in his opening paragraphs, “is a continuation of the debate over the legitimacy of the constitutional revolution of [the New Deal].” Some of the time, Professor Epstein writes in that constitutional key. Once the substantive limits on the general governments’ powers give way, for example, the non-delegation doctrine is a “third-best,” ineffectual option. Most of the time, though, Professor Epstein writes in an Administrative Law mode. “Forget all the fancy Supreme Court gloss that has gutted the APA,” he concludes, “and then return the law to its original design, meaning, and structure. The constitutional questions will then largely take care of themselves.”
Professor Epstein has elaborated his enduring constitutional qualms elsewhere. And I understand the attraction of the “return to the APA” argument—“APA originalism,” as it has come to be called: it seems to ground an attack on contemporary administrative law, without opening the author to demagogic “Constitution in Exile”-style attacks. The difficulty is that APA originalism runs up against the constitutionalist position at every turn.
The APA says that the “administrative process” is an adequate substitute for regular legal procedures in an independent court and that it will have to do no matter what or how much is at stake for regulated parties. It so enshrines the very premises that Professor Epstein resists and contests. And while I am no great admirer of the Jackson Pollock canvas that the courts have plastered over the APA, Professors Sunstein and Vermeule are correct that some post-APA judicial doctrines—for example, elaborate notice-and-comment rulemaking requirements—have made administrative law more lawful in Fuller’s sense. Those doctrines reflect a judicial gut sense that agencies invested with vast discretionary powers need some legal oversight. That intuition, in turn, is tied to fateful changes in administrative practice that, a bit curiously, go largely unmentioned in Professor Epstein’s book. The APA was designed for agencies to organize industry cartels, principally by means of adjudication—the FCC, the SEC and so on through the alphabet. That is nobody’s favorite way of running an economy but it was relatively harmless in rule-of-law terms.
For example, you could make do with the truly minimal APA requirements for a “Notice of Proposed Rulemaking” because the terms had long been negotiated over Bourbons at the Willard Hotel. Modern-day agencies and their regulatory rulemaking apparatus are nothing like that. Judge-made administrative common law is a response to that reality, and it may not be enough. For example, the EPA can redirect vast private investments by simply announcing that it might impose this or that rule. We have no legal rule to forestall such maneuvers, and you won’t find one in the APA.
Professor Epstein’s APA originalism rests on the notion that it mandates de novo judicial review on all questions of law, as opposed to Chevron’s deference regime and its associated canons. That is not obviously correct, but let’s say it is: how far does that get you? What’s a “stationary source” under the Clean Air Act—every pot or boiler, or the entire industrial facility (called a “bubble”)? You can look to the Clean Air Act and mobilize any textualist canon you want: law just gives out. Now what? Professor Epstein, if I understand him correctly, argues that the Court should have waived through the EPA’s bubble program because it made obvious sense. In substance, I agree. But that is not a legal argument, is it? And it is in tension with Professor Epstein’s forcefully articulated Fullerean precept that courts ought to be suspicious of sudden changes in agency policy: that’s what the Court was looking at in Chevron.
Finally: Professor Epstein argues that administrative law went astray when the D.C. Circuit and then the Supreme Court adopted deference doctrines—Chevron and its associated canons—on questions of law and “hard look” review, meaning close scrutiny of an agency’s policy decisions, on matters of fact: that’s upside down. But that abstract proposition leads nowhere. Putting aside that “hard look review” is largely a mirage, the insistence on de novo judicial review collapses when nominally “textualist” judges and Justices mobilize and manipulate interpretive canons to explain that night is day and black is white and so therefore, it’s the agency’s call.
Richard Epstein knows this. Way back when, he writes:
judges did not spend much time worrying about the inherent ambiguity of ordinary language. They tended to decide cases without any overwrought methodological apparatus. […] The lower the level of intellectual angst, the more accurate and reliable will be the [courts] interpretations given.
Like I said: a man of the 19th century, and a man of my heart. How to recover the innocence and wisdom of an earlier age is the crucial question, and the ground on which we will have to stand.
[*] Professors Sunstein and Vermeule reprise and elaborate their principal argument in a forthcoming book entitled Law & Leviathan: Redeeming the Administrative State. I have flipped through the page proofs: due undoubtedly to unfortunate timing, the book contains no reply to Professor Epstein’s critique. That is a shame, because the exchange—spirited, contentious, but mercifully free from cheap polemics on either side—is a conversation worth having.