Design for Liberty: Private Property, Public Administration, and the Rule of Law
By Richard A. Epstein
(Harvard University Press, 233 pp., $29.95)

Richard Epstein is the same as he ever was. Part erudite scholar of Roman law and the common law, part provocative intellectual who promotes a view that he calls “classical liberalism,” Epstein is relentlessly true to himself, and this gives his works a unity of tone and content that both pleases and distresses. After shifting early in his career from a natural-rights justification for classical liberalism to a utilitarian justification, the Epstein program has shown no signs of internal struggle or evolution.

And it is not as though there is nothing with which to struggle. Many scholars have offered withering critiques of the Epstein program, but there is little sign that the arguments of the critics have been heard and considered. Epstein’s latest book targets the administrative state as the enemy of classical liberalism, and argues that the administrative state is inconsistent with the rule of law, but Epstein fails to come to grips with objections that have been made many times to arguments of this sort. Indeed, James Landis, a New Dealer and dean of the Harvard Law School, addressed similar arguments in 1938 and offered root-and-branch objections about which Epstein says barely a word. Is there a word to describe a thesis that was refuted before it was written? Prefuted, perhaps? If there is such a thing, Epstein’s book has been prefuted.

In Epstein’s worldview, classical liberals have to fight a two-front war against both libertarians and progressives. On the right flank are strict libertarians who insist that natural rights bar the state from undertaking any program of coercive taxation. On the left flank are progressives who favor expansive schemes of taxation, redistribution, and regulation. (These are Epstein’s conceptions, not necessarily those of the libertarians and progressives he sees himself as battling.)

As against the libertarians, Epstein argues that coercive taxation and regulation are permissible for mutual advantage—that is, to overcome problems of collective action that prevent individuals and markets from supplying public goods at desirable levels. Property holders may be forced to accept compensation “in kind,” or in other implicit goods, for the forced surrender of their rights, but only if they are left at least as well off by the transaction as they would be in its absence. Thus state schemes of anti-monopoly regulation, mutual insurance, and defense are generally permissible, subject to the mutual-advantage constraint.


THUS FAR, EPSTEIN'S arguments amount to a conventional welfare-economic view that emphasizes the role of the state in supplying public goods. More dramatic is the fight on Epstein’s left flank, where he wants to indict the administrative state. The administrative state is the enormous apparatus of modern government, both in the United States and in the other democracies, that is constituted by a menagerie of administrative agencies, boards, and commissions. In the United States, everyone has heard of the Environmental Protection Agency, the Food and Drug Administration, and a few others, but there are in fact hundreds of such agencies that act under the authority of statutory delegations from Congress, taking a staggering variety of forms and pursuing a staggering array of regulatory, redistributive, and managerial aims. In many cases, the delegating statutes combine legislative, executive, and judicial functions within agencies; they give agencies the power to make rules, enforce the rules, and judge whether the rules have been violated. This combination of functions makes a certain brand of political liberal, one who has taken very seriously some famous passages from Montesquieu and the Federalist about the separation of powers, recoil in horror.

Epstein makes the usual liberal noises about the separation of powers, but the main thread of his indictment derives from Friedrich Hayek, his intellectual lodestar. Following The Road to Serfdom, Hayek’s cri de coeur, Epstein worries that the administrative state is inconsistent with the rule of law. The claim is puzzling at first glance, because the agencies derive their powers from enacted law, rather than tyrannical fiat. To understand the Hayekian concern, we need to invoke a jurisprudential distinction between “thin” versions and “thick” versions of the rule of law.

The thin rule of law is simply the requirement that coercive governmental power be exercised in accordance with duly enacted legal instruments, and the agencies are fine on that score, because they trace their authority to delegating statutes. But the thick rule of law is more ambitious, and requires that governmental authority be exercised in accordance with a set of criteria elaborated by the legal theorist Lon L. Fuller—criteria of generality, prospectivity, clarity, and neutrality. Hayekians such as Epstein emphasize that the thick rule of law promotes the predictability of official action, and this in turn helps private actors to plan their affairs.

Epstein thus sees the administrative state as a repository of vast official discretion, antithetical to generality and predictability. The resulting political risks include official favoritism and dirigisme, “rent-seeking” or lobbying by firms seeking benefits funded by taxpayers or seeking to impose regulations that will disadvantage competitors, and ham-handed regulatory intervention that needlessly blocks improvements that private parties could otherwise have made through contractual arrangements. The common problem in all this is that the agencies exercise excessive discretion under vague delegating statutes that permit them to issue ad hoc commands rather than general rules of law.

Epstein’s main prescription is to restore key “procedural and structural” features of the classical liberal regime, of which the most important are “unbiased decisionmaking” and “judicial review of administrative actions on matters of fact and law.” The phrases are loaded. What Epstein means is that the law should not permit the combination of rule-making, enforcement, and judging functions within administrative agencies, and should reject the so-called “Chevron doctrine,” under which courts defer to agencies’ interpretations of the laws they administer so long as those interpretations are reasonable. Epstein also wants a revival of constitutional restrictions on the delegation of power to administrative agencies, an idea that the Supreme Court abandoned in the 1930s. As applied to current controversies, Epstein seemingly wants the courts to narrow or perhaps to invalidate—he is not clear about this—both the Obamacare statute and the Dodd-Frank law that attempts to strengthen financial regulation. Whatever the details, and whatever the context, more review of agencies by judges is the watchword.


NONE OF THESE diagnoses or prescriptions is exactly novel. Critics of the administrative state have been complaining about excessive delegation and arguing for more vigorous judicial review at least since Albert Venn Dicey wrote in the late nineteenth and early twentieth centuries. Works with titles such as The New Despotism argued relentlessly that the burgeoning administrative state, observable both in Britain and the United States, would eventually undo the rule of law altogether; the critics hoped that judges would constrain administrative discretion while enforcing the general and neutral legal rules of the pre-existing common law. Against this backdrop, Hayek himself was a latecomer to the party, and Epstein’s latter-day Hayekianism is old hat by now.

Since the debate is so old, it should not be surprising that the rule-of-law critique of the administrative state is subject to well-known rejoinders. It is not as though judges lack discretion of their own: a vast body of literature in political science demonstrates conclusively that in the hard cases that reach appellate courts, judges often rule on predictably ideological and partisan grounds rather than strictly legal ones. (The literature is mostly about American judges, but recent work in the British context has begun to explore whether the same finding holds true there.) The Epsteinian and Hayekian worry that agency policy-making will be “captured” by rent-seeking firms—special interests with lobbyists—is fatally non-comparative as well, because such firms can also seek rents through the litigation process, hiring expensive lawyers, and bringing wave after wave of suits in the hopes of establishing favorable precedents. Epstein falls into the nirvana fallacy—the tendency to juxtapose a jaundiced picture of agency policymaking to an idealized vision of commonlaw courts announcing and enforcing general and neutral rules. Real courts may be captured by the forces that resist regulation or else use their discretion to indulge their ideological proclivities.

Moreover, Hayek’s intuition that only general rules make decisions predictable was simplistic. Ad hoc administrative decision-making can promote legal certainty and predictability as much or more than general rules enforced by courts. As long ago as 1924, Roscoe Pound observed that general rules are inevitably open-textured, and are thus dependent on the uncertain interpretation of judges in future litigation. As Pound put it:

Especially in the complicated economic organization of today the law cannot say to the business man, well, you guess; you employ a lawyer by the year to give you the best guess that he can, and then as the result of litigation we will tell you five years afterwards whether your guess as to the conduct of your business was the correct one or not.... Our administrative commissions are nothing but traffic officers, as it were, with signals to tell us when to cross and when not to cross, and where to cross.

The largest point, however, stems from James Landis’s work in 1938. Landis’s great theme was substitute safeguards: even if the administrative state has slipped off the traditional constraints of the separation of powers, there are new political and sociological constraints against official abuses. Landis defended the combination of functions in agencies by observing, “The fact that there is this fusion of prosecution and adjudication in a single administrative agency does not imply the absence of all checks. It implies simply the absence of the traditional check.” Prominent among the substitute safeguards identified by Landis were professionalism and expertise. Agency administrators, even if politically appointed, would tend to possess or develop a mission orientation that would insulate them to at least some degree from the more sordid forms of political temptation and influence. Many have debated whether that claim is correct, but from Epstein’s book one would be hard-pressed to discern that it had ever been made.

That is the sense in which Landis prefuted Epstein. One may believe, as I do, that the rule-of-law critique of the administrative state falls flat, or one may believe the opposite. Yet Epstein never really comes to grips with the standard rejoinders, not in this book anyway. Instead he primarily reiterates the first-level Hayekian arguments in setting after setting. The result is a book that will please the faithful but not persuade the doubtful, or break any new ground. The debate stands where it was.

Adrian Vermeule is a John H. Watson Professor of Law at Harvard Law School. This article appeared in the March 15, 2012 edition of the magazine.