Law's Province


Review of Law’s Abnegation: From Law’s Empire to the Administrative State, by Adrian Vermeule

Cambridge: Harvard University Press, 2016

Let me hasten to the bottom line. Adrian Vermeule’s Law’s Abnegation is a stunning achievement – a must-read for any jurist, scholar, or practitioner of administrative law or indeed of public law more generally. This slim volume suffers from only one pervasive defect, repetitiveness: it could have been slimmer without losing any of its analytical power or coverage. Extenuating this minor defect is Vermeule’s writing style, which is pellucid, matter-of-fact, elegant and forceful.

His argument, rigorously developed in six chapters, is utterly convincing (at least to this erstwhile teacher of administrative law), and can be easily stated: the arc of administrative law bends toward judicial deference to agencies. Vermeule is agnostic as to the starting point of this arc: “Jerry Mashaw’s latest comprehensive treatment of the development of the U.S. administrative state pushes it farther back than many would have imagined possible, and. . .the picture of the heroic common-law judges standing against the encroachments of royal prerogative is entirely misleading.” (217) But he has no doubt that it will continue indefinitely. The dynamics powering this arc imply that there is no stopping point along the way. In their supervision of the administrative state, he assures us, judicial review of the probing, intrusive sort ordained by (some of) the language of the Supreme Court’s 1983 State Farm decision is in steady retreat.

To Vermeule, this retreat is welcome news. I am struck by the fact that his book highlights the endless tradeoffs in public law generally and administrative law in particular, yet he does not mention (unless I missed it) even a single downside to this trend. Indeed, he excoriates scholars like Philip Hamburger and Gary Lawson who celebrate the good old days of judicial supremacy both because of their false historical accounts and because of their extreme criticisms of the modern administrative state.

His overarching academic target, however, is the “puzzle” of Ronald Dworkin’s “near-total silence about the administrative state and administrative law.” (219) Only by ignoring them, Vermeule observes, could Dworkin place courts at the center of law’s empire. (Dworkin’s focus on constitutional law might, in a more generous reading, excuse this omission.) To Vermeule, however, Dworkin’s notion of law as integrity – his “romantic[ization] of courts” was a case of “willful blindness” to the importance of deference, which was evident even when Dworkin wrote. (6, 9) Dworkin’s law, Vermeule asserts, is not on the throne but “lying more or less quietly” beneath it. (22)

The book’s master theme is indicated by its title. Vermeule insists on distinguishing between law’s “abnegation” – by which he means its rational, self-conscious decision to limit the courts’ authority to review agency decisions stringently, and its “abdication” – by which he means courts’ refusal to exercise the supervisory review powers over agencies that they legitimately possess and have a duty to exercise. This distinction is a bit strained and overwrought, for they are essentially synonyms.[1] Still, Vermeule is clear about what he means by it – a two-part explanation. First, reviewing courts’ strong tendency to allow agencies to set substantive policies, determine relevant facts, interpret ambiguous legal standards, and even fix the bounds of their own jurisdictions is anything but grudging; law, he says, has not been shoved aside but instead deposed itself. Nor is the eagerness of the agencies to seize this authority a coup d’état. Instead, Dworkin’s imperial courts have made themselves “roi faineants” by willingly ceding power to the agencies. (Sounds like an abdication to me!)

Second, the courts have relinquished their power “for valid lawyerly reasons” (as Vermeule puts it in his very first sentence). In an arresting metaphor (repeated throughout the book), he sees administrative law “working itself pure” through a reasoned, principled consistency. This is how it happens:

. . . [J]udges and lawyers value logical consistency. The main mechanism behind law’s abnegation has been a commitment to reasoned consistency on the part of the legal profession. Once law started down the path of deference, for legitimate legal reasons. . . abnegation became even more difficult to resist, purely from an internal standpoint. . . Consistency is never obligatory, and inconsistencies persist in local settings, but the demands of consistency exert a constant steady pressure that tells over time. Courts that start to defer late evolve principles that entail deference on many fronts, such as the principle that agencies enjoy discretion to allocate resources across cases and programs; in turn, those principles themselves become part of law’s deep fabric, and increase the impetus for deference in other settings. The result is a self-reinforcing feedback process in which abnegation becomes not a betrayal of law, but the logical consequence of law’s own internal commitments. (2-3)

The author’s historical and analytical baseline is Chief Justice Hughes’ opinion in Crowell v. Benson in 1932, just before the advent of the New Deal. Hughes sought heroically to design a doctrinal solution to the constitutional problems posed by an administrative state in which judges were charged, as Hughes saw it, with the duty to maintain an appropriate balance between courts and the newly-proliferating agencies. After specifying the elements of the Crowell solution – judicial interpretation of all statutes, judicial responsibility for the adjudication of public and private rights, preoccupation with formal adjudication rather than rulemaking, judicial control of “jurisdictional facts,” and so forth -- Vermeule shows how each of them has collapsed, acceding to agency authority to make these determinations subject only to clear errors and review for mere rationality.

Thumbing his nose at the “elegiac, wistful, lamented” (217) paeans by Lawson and Hamburger to the old order (indeed, to an ancien regime based on common law judging), Vermeule applauds the Supreme Court’s 2013 decision in City of Arlington v. FCC marking what he sees as the almost total surrender of the Crowell regime to the administrative state’s compelling demands for judicial deference. And far from violating our constitutional principles of separation of powers and limited government, he constantly reminds us, this fusion of functions in the agencies was authorized, indeed championed, by the legislative and executive branches duly exercising their traditional powers under Articles I and II. This, then, is no seizure of power rightfully held by the constitutional actors; it is “separation of powers without idolatry.” (57). In this view, those who argue otherwise – including Jeremy Waldron and the late Justice Scalia, as well as Lawson and Hamburger – have no constitutional or legal-historical leg to stand on. They are “inebriated by an excessive intake of principle.” (72)

Is Vermeule engaged in a normative project here? His claim that he will “not pronounce on. . .the optimal allocation of authority between courts and administrators in the modern state” (8) is a bit misleading. Without saying which allocations are “optimal,” strictly speaking, his book is replete with assertions that this or that doctrinal abnegation is highly desirable. Much the same is true of his claim that the book is not “about” law but from “within” it. Given his clear doctrinal preferences, this seems like a distinction without a difference.

Indeed, the great bulk of his book is relentlessly normative. In a series of chapters, Vermeule explains and justifies the main developments of modern administrative law regarding judicial deference to agency procedures and decisions – especially substantive due process and rationality review. In these areas, his readings of Supreme Court (and some D.C. Circuit) caselaw – Chevron, Mead, Auer, and others -- are invariably careful, nuanced, balanced, shrewd, and (to this reader) convincing. (Whether a Justice Gorsuch, whose criticism of Chevron deference seems even greater than the late Justice Scalia's, would attempt to reverse this trend, remains to be seen.) His analysis of the issues is always functional, asking what advantages courts bring to the table. Because of “their limited information, episodic perspective, and time constraints. . .the marginal benefits of judicial intervention in procedural design may be low or even negative, as the intensity of intervention increases.” (115) Will such deference allow the agency to police itself and thus be a judge in its own case? He approvingly quotes the Court’s decision in City of Arlington: “The fox-in-the-henhouse syndrome is to be avoided not by establishing an arbitrary and undefinable category of agency decisionmaking that is accorded no deference. . . Where Congress has established a clear line, the agency cannot go beyond it, and where Congress has established an ambiguous line, the agency can go no further than the ambiguity will fairly allow.” Short of this, Vermeule insists, no free-floating judicial oversight is warranted. (123)

This is particularly true in areas of uncertainty -- whether legal, factual, normative, or policy. In what is the most interesting and novel (at least for lawyers) discussion, Vermeule defends the notion that agencies must often make “rationally arbitrary decisions” (126) in the face of situations in which “first-order reasons” – reasons grounded in reliable information about past, present, and future that support reasonable, determinate predictions –“have run out.” (134) He usefully distinguished five types of uncertainty that can produce such situations. “’Brute’ or epistemological uncertainty arises from the sheer cost of acquiring facts about the world.” (135) “Strategic” uncertainty, the province of game theory, arises from the interdependence of actors’ choices, situations in which multiple equilibria are not only possible but probable. This can produce “model uncertainty” where “it is irreducibly unclear whether [a] model raises, reduces, or has no effect upon [the outcome being analyzed]. (139)

Fourth, there is uncertainty about how optimistic or pessimistic an analysis should be. Here, Vermeule’s discussion is particularly important, given widespread support for the “precautionary principle” (or what he calls the “Dismal Theorem”). (141) He convincingly shows that – generally speaking -- the precautionary principle is itself irrational when profound uncertainties exist about the future on all sides of the issue. Finally, uncertainty often exists about how much further information to gather, given the agency’s irreducible ignorance about the marginal costs and benefits of further inquiry.

Agencies, then, are often and increasingly at the “uncertainty frontier” (152), where reviewing courts must accept “second-order reasons” grounded in these indeterminate uncertainties, at least where the agency explains why those reasons apply in the particular case. In such cases, rationally arbitrary reasons are the best that any agency can provide. Courts, which are ill-equipped to improve on those reasons and cannot reasonably demand that the agency do better under these circumstances, should accept them instead of demanding what would be impossible or, what is much the same, impossibly costly. “It does not follow. . . that any decisions that are rationally arbitrary, in the decision-theory sense, must also count as arbitrary in the legal sense.” (150, emphasis original)

Vermeule then devotes a chapter specifying (somewhat repetitively) what he means by the “thin rationality review” that he (and the Supreme Court, as he shows) endorses, and why State Farm and other “hard look” decisions are outliers. Most valuable are his discussion of the tradeoffs that agencies face in making policy at the uncertainty frontier, where they must eschew optimization in favor of “satisficing” (in Herbert Simon’s famous formulation). These include the “means-variance” tradeoff, the speed-accuracy tradeoff, and that arising from asymmetric error costs. “Thin rationality,” in his view, should satisfy reviewing courts: “arbitrary and capricious” in APA review “is best understood as unreasoned agency action” (emphasis original) (188), a decision that fails to give adequate second-order reasons for its first-order inadequacies. The Court’s Baltimore Gas decision, not State Farm (they were decided in the same Term), articulated the proper approach. And he maintains that the Court’s 2015 decision in Michigan v. EPA requiring the agency to consider costs and benefits under a specific provision of the Clean Air Act is to the same effect, standing “only for the unobjectionable proposition that rationality requires consideration of both ‘the advantages and the disadvantages of agency decisions,” not a “formalized quantified and monetized cost-benefit analysis.” (178)

Where does this leave the courts? Vermeule’s review of the caselaw (listed in an appendix) supports his belief that the Supreme Court has sent sound signals to the lower courts:

At the level of the Supreme Court the law has decided to abnegate authority to agencies under conditions of uncertainty, and rightly so. The same is largely true of lower courts, but the Supreme Court’s message has not been received in full by its subordinates. Some lower courts—especially panels of the D.C. Circuit—sometimes demand the unobtainable, overturning agency decisions for lack of first-order reasons that could not be supplied no matter how the agency had decided. Such decisions are holdovers, excreta of the traditional legal mind, which are not central to law as it now stands, and which will in an (sic) event disappear as the law increasingly works pure its own abnegation. (154)

In his final chapter – pointedly entitled “Conclusion: Law on the Margin” (209) – Vermeule explores the implications of thin rationality review for lawyers. His verdict is clear: the professional expertise of judges and lawyers is marginalized relative to that off engineers, scientists, statisticians, economists -- and for good reasons. (Oddly, he does not mention Holmes’ strikingly similar prediction 120 years earlier).[2] Chenery, Chevron, and Mead, he argues, signal a welcome reallocation of authority from lawyers to technical specialists. Vermeule again: “The cases featured in so many textbooks on Legislation and Statutory Interpretation. . . are child’s play compared to any average EPA case that the Court hears nowadays. As the administrative state takes on more functions, the complexity and stakes of its decisions rise, and judges become increasingly worried about their marginal value added.” (213) (Although he does not say it -- perhaps out of compassion -- the same might be said of law professors!).

Returning to Crowell v. Benson and Hughes’ insistence that legal questions must be decided by the courts, Vermeule finds that notion quaint. “It [is] impossible to disentangle legal questions from policy-making decisions, at least as to the complex regulatory statutes that predominate in this modern state. It [has] become clear in turn that agencies, at least as compared to courts, [are] better positioned both to make ultimate value choices relevant to regulatory questions (such as the precise rate of trade-off between burdens on regulated firms and protection of public health) and also to determine facts, causation, and the likely consequences of alternative interpretations.” (214) What, then, is left for reviewing courts in such cases, beyond testing second-order rationality? Not much. Whether this adequately protects rule-of-law values will depend in part on agencies’ ability and motivation to neuter even this thin rationality review by presenting dubious, tactical, or politicized second-order justifications for their rules. Vermeule’s treatment of this concern – that it is no different than other pretextual moves by agencies that can be flushed out in various ways -- is much too glancing and superficial to lay such realistic fears to rest. It is perhaps the most unsatisfactory aspect of this otherwise powerful, remarkable book.

Posted on 6 February 2017

PETER SCHUCK, an emeritus Yale Law School professor, is the author, most recently, of One Nation Undecided: Clear Thinking about Five Hard Issues That Divide Us, (Princeton University Press, March 2017) and Why Government Fails So Often, and How It Can Do Better(Princeton University Press, 2014).



[2] Oliver Wendell Holmes, Jr., “The Path of the Law,” 10 Harv. L. Rev. 457, 469 (1897).