Courts, Agencies, and Constitutionalism: Further Thoughts

I’m very thankful to the Liberty Law site for offering me this opportunity to discuss some fascinating and important issues surrounding our modern administrative state. And I appreciate both Professor Lawson’s and Professor Seidenfeld’s responses to my essay, as well as Professor Greve’s contributions to our debate.

Administrative Law Orthodoxy: Perpetuating the Status Quo

Professor Seidenfeld’s central proposition is that I am guilty of a heresy. I depart, he argues, from the “premises shared…by the general community of scholars who study the administrative state.” Specifically, his view is that my argument “rests on strictly libertarian assumptions.” My libertarian assumption, he alleges, is “that the proper job of courts is to reduce the size of the administrative state.” Right thinking administrative law scholars, by contrast, “believe that the role of courts should be to ensure that government operates deliberatively and transparently to perform the tasks that the political branches have assigned it by statute.”

Since the essay adopts unacceptable premises, Professor Seidenfeld refuses “to be hemmed in by Postell’s untenable assumption” and does not offer “a direct response to his polemics.”

Professor Seidefeld’s accusation of heresy is partly correct: I depart from the assumptions generally accepted by many administrative law scholars. Administrative law scholars tend to use administrative law as a means of tinkering with the administrative state – smoothing out its rough edges but keeping the edifice essentially intact. I’m more interested in considering alternatives to the administrative state itself.

However, I believe that Professor Seidenfeld is wrong about what my alternative assumptions actually are, and therefore he is wrong to refuse to consider them. My central goal is not to advance libertarianism. I am more interested in preserving sound institutions than I am in limiting regulation. In other words, my main goal is not to eliminate regulation but to “ensure that government operates deliberatively and transparently.”

Thus, read carefully, my argument actually adopts the exact premises that Professor Seidenfeld chides me for shirking. My goal is not to use polemics to eliminate regulation and produce a minimal state. It is to have regulation that is produced through deliberative, transparent means. New means, to be sure, but in pursuit of the goals embraced by orthodox administrative law scholars.

Professor Seidenfeld argues that this is a fool’s errand: “[t]he notion that the courts through common law actions can provide the regulation necessary to maintain even a minimalist conception of the modern state is ludicrous.” An often-repeated assertion, but I’m skeptical. Scholars such as William Novak, Brian Balogh, and Jerry Mashaw (and too many others to mention) have chronicled the vast scope of regulation in nineteenth-century America and found no “minimalist state.” In my view the evidence clearly indicates we can have plenty of regulation within our constitutional system. Whether we use common law actions, clear statutes with clear prohibitions to be enforced by agencies and adjudicated in independent courts, or all of the above, we can find methods for producing regulations without scrapping the rule of law.

Furthermore, my hesitation about administrative justice has roots in the anti-bureaucratic wing of the Progressive movement of the early 20th-century, a movement few would characterize as libertarian. Woodrow Wilson (at least in 1912), Louis Brandeis, Robert La Follette, and other influential progressive reformers were just as wary of administrative justice as I am today. (By contrast, the railroads were the chief advocates of a strong commission during the creation of the ICC.)

(For the record: I favor national ambient air quality standards. I favor workplace safety standards. I favor food, drug, and consumer product safety regulations. I favor banking regulations. But I am interested in thinking creatively about how we might produce these regulations in a way that does not subvert representative democracy and the separation of powers. That such an approach is dismissed as heretical speaks volumes about “mainstream administrative law scholars.”)

Reforming Administrative Law through Congress

Since Professor Lawson neither “shirks from” being labeled “quixotic” or “bizarre,” he is less willing to reject my argument out-of-hand. He does present important and challenging problems for my argument.

Professor Lawson’s chief objection is that history (the pre-administrative state arrangement I outlined) and political science (the extent to which the President holds agencies accountable) are not the “real reason[s] why rule-of-law advocates should be contemplating increased judicial involvement in administrative law.” Rather, “a case against extreme judicial deference to agencies should focus on law rather than political science.” Professor Lawson then proceeds to make the legal case against extreme judicial deference.

Broadly, he argues that judicial deference to agencies is a function of the “legislative command” of Congress. On questions of fact and policy, courts are commanded by Congress to defer to agencies. Opponents of deference, therefore, have three options: 1) deliberately misinterpret the law, 2) change the law, 3) find laws that command deference to be unconstitutional. Professor Lawson chooses the third option since “Congress cannot tell the courts what process of reasoning to employ in deciding cases.”

This is an intriguing suggestion, and one to which I will defer for purposes of the argument. But I am inclined to think that Professor Lawson dismisses the second option too hastily. The political climate, he argues, is not conducive to passing statutes that will establish a wider scope of review in various areas of administrative law: procedural review, substantive review, review of agency legal interpretations, availability of judicial review, and the like. This very well may be true. But I think it is important to remember that liberals, not conservatives, are responsible for expanding judicial review in these areas over the past 40 years, and not simply on the basis of political advantage (i.e., conservatives controlled the presidency and the public interest groups were mostly on their side).

In short, perhaps many on the left would find increased judicial review of agencies appealing, in spite of the immediate political calculus. (As an aside, I’m with Professor Greve that the “calculus of who will own the agencies and the courts respectively” should not play a role in determining the institutional arrangements, and my proposals would apply regardless of who controls which branches of government.) And in any event, introducing thoughtful reforms that adjust the scope and standard of judicial review certainly allows for more appealing and interesting arguments than simply attempting to eliminate major federal programs. This strategy allows us to start talking about procedural and institutional reforms, rather than laissez-faire (unless, of course, those arguments are misinterpreted to mean laissez-faire, which apparently seems likely).

All of this is to say, I think, that while the legal reasons against extreme judicial deference are valid, the administrative state is ultimately a function of politics as well as law, and reform of the administrative state cannot be an entirely legal project, to be carried out through reform of administrative law doctrine. Political reforms – i.e., reforms forged through a political, legislative process – will inevitably play a significant role, and political arguments are informed by historical and political arguments as well as legal arguments.

What Exactly am I Talking About?

To leave the discussion at such an abstract level, Professors Lawson and Greve both rightly assert, is not helpful. Professor Greve states the basic question: “what exactly do we want courts to do?” Professor Lawson explains that I need “to spell out the mechanism(s) by which that reconsideration [of judicial deference] will take place.” In a 3000-word opening salvo, it simply was not possible to be very specific, but once the right principles are established, I acknowledge that it is just as important to get down to brass tacks.

Professor Lawson suggests that my “chief target” is the Chevron doctrine, which (roughly) requires judicial deference to any reasonable agency interpretation of an unclear statutory provision. Chevron probably needs to go. Professor Lawson supports my opposition to Chevron, once again suggesting that I don’t need history or political science to object to Chevron since the legal arguments are compelling. While I agree that there is no clear legal (i.e., statutory) authority for the Chevron doctrine, it certainly would not be difficult to take the political route and propose positively eliminating the doctrine by statute. Since many legal thinkers on the left have made Chevron a prime target in recent years, once again this may be a feasible political project.

Chevron has to do with how courts will review agencies’ legal interpretations within the existing administrative state. Professor Greve notes that my argument points in a very different direction – namely, a rethinking of the administrative state from the ground up. I suggested that we consider two alternative models that allow for regulation but with a very different institutional structure: common-law regulation and the German Rechtstaat. As Professor Greve thankfully explains, the Rechtstaat “has nothing to do with small government,” but is primarily concerned with “lawful government – which can be very, very big.” I agree that if we adopted common-law regulation or the Rechtstaat model, we would still have a large government, but we would have a more lawful government.

Professor Greve knows exponentially more than I do about the Rechtstaat, but he draws a helpful distinction between public matters and private rights that I think is instructive. In the Prussian model, independent courts were established for review when actual, individual rights are violated. At the same time, however, administrative courts were established for the adjudication of public matters through an administrative process. With regard to the resolution of public matters, procedures were severely limited, but with independent courts there was de novo review of questions of law and fact.

As a variation on this theme, I would propose serious consideration of Professor Rappaport’s suggestion on this site that certain kinds of adjudications be moved from agencies into independent courts. A helpful rule (which Professor Lawson has explained clearly in his administrative law textbook) is this: if the agency is making a determination that takes life, liberty, or property, private rights are implicated and the decision has to be made by an independent court. Conversely, if the agency is making a determination involving “public matters” an administrative tribunal is fine. But even in this latter case, it is essential to keep the investigators, enforcers, and judges separated as much as possible. (Such distinctions between different types of agency actions incorporate JV DeLong’s thoughtful comment that different agencies are different, and that no “one size fits all” solution to this question is wise.)

As is probably apparent by now, I don’t exactly know what exactly needs to be done to (in Professor Greve’s words) “re-constitutionalize the administrative state.” But I am convinced that the administrative state needs to be re-constitutionalized and that such a project will involve going “outside the domain of administrative law.” And the most likely forum for conjuring up principled and thoughtful reforms is one that brings together scholars who take constitutionalism seriously. So I thank the Liberty Fund for facilitating this exchange.