Elizabeth Magill · June 2009
77 GEO. WASH. L. REV. 859 (2009)
Discretion is at the center of most accounts of bureaucracy. It is no mystery why this is so. While agencies are hemmed in by statutes, the President, and courts, they still possess enormous discretion that they can exercise in ways that matter to the parties who have a stake in what they do. For many social scientists who study bureaucracy, that discretion is just a fact about the world that bureaucrats inhabit. Legal scholars have tended to take a more normative view. A few have celebrated agency discretion as making space for the exercise of expert judgment, but the dominant modern approach tends to be skeptical of that discretion. Scholars working within this latter tradition exhort those with supervisory power to tame that discretion as a matter of law, politics, or constitutional command.
Strangely absent from these accounts is a ubiquitous phenomenon: administrative agencies routinely “self-regulate.” That is, they limit their options when no source of authority requires them to do so. They voluntarily constrain their discretion. They adopt rules, guidelines, and interpretations that substantively limit their options—limiting either the range of outcomes they can reach or the rationales that can be used to defend their choices. They also limit their procedural freedom by committing to afford additional procedures, such as hearings, notices, and appeals, that are not required by any source of authority. The idea that an agent can have an interest in voluntarily limiting his own options is hardly novel. Indeed, it is an ancient proposition. Jon Elster put it simply in introducing his important work on the subject: “sometimes less is more.” Individuals, firms, governments, and, yes, of course, bureaucracies—especially bureaucracies—will sometimes have an interest in voluntarily limiting their options.
And yet, what this Article calls self-regulation is not now a category that exists in the study of administrative agencies. Some scholars of the administrative state, it is true, have argued that, when statutes delegate policymaking authority to agencies and those statutes are, in their crucial details, vague, agencies should be required to limit their discretion as a matter of constitutional or statutory command or administrative common law. But, other than urging agencies to limit their discretion, this literature does not actually pay much attention to self-regulation. What, precisely, is self-regulation? When do agencies engage in it and what can they accomplish when they do so? What are the implications of that behavior for various debates that we engage in about the administrative state?
The aim of this Article is to create the category of self-regulation and to persuade students of the administrative state that it has been a mistake to ignore it. To do that, this Article will identify in Part I the key features of self-regulation, outline in Part II what an agency can accomplish by self-regulating, and demonstrate in Part III the implications of serious study of these voluntary agency constraints for important debates about the administrative state.
The first step is to understand the key features of self-regulation. Doing so requires a working definition of self-regulation and a preliminary descriptive account of the types of self-regulation in which agencies engage. Self-regulation is defined here as an agency action to limit its own discretion when no source of authority (such as a statute) requires the agency to act. Paradigmatic examples include enforcement guidelines and “extra” procedures—that is, procedures that the relevant law would not require the agency to provide.
Understanding self-regulation also requires an understanding of the consequences of self-regulation—whether and, if so, how self-regulatory measures limit an agency’s options. Grasping the consequences of self-regulatory measures is the key to understanding what an agency can accomplish by self-regulating, and hence why it might do so. But it is no easy task to capture the consequences of self-regulation precisely. Understanding how binding an act of self-regulation can be depends on a host of factors, including having a clear picture of a less-than-clear corner of administrative law: under what circumstances will a court force an agency to follow its own self-regulatory measures? Administrative law doctrines actually allow agencies to make a binding commitment to their self-regulation because the law promises that, under certain conditions, a court will enforce self-regulatory measures against agencies if and when they violate them.
Having identified the essential features of the category of self-regulation, the Article then examines what an agency can accomplish through self-regulation. Agencies often need to control policy implementation by subordinates; they may wish to limit their own options in order to induce reliance by outside parties; they may hope to protect their own autonomous policy choices—either from being changed by a future administration or from being overridden by a political principal today. Self-regulation is a way for an agency to achieve all of these objectives. Self-regulation also allows the agency to produce certain collective goods, such as information and reputation, that the agency needs to do its job but may be underproduced unless affirmative steps are taken to assure their production.
The Article then turns to the implications of agency self-regulation. Part III shows why the study of agency self-regulation can change the way we understand agencies and the need for and utility and wisdom of various controls on their behavior. Understanding when, why, and how agencies self-regulate would round out our understanding of the options an agency has at its disposal to achieve its objectives, inform our assessment of the strengths and weaknesses of the institutions that monitor and control agencies (courts, the President, and Congress), and fruitfully inform the normative debate over delegation. Because self-regulation as it is defined here is a voluntary action by the agency, studying it may also be a particularly useful way to make progress on one of the most vexing questions for those of us who study bureaucracy: what makes an agency tick? Study of self-regulation may help us evaluate competing accounts of agency incentives. Part III is not intended to exhaustively explore the implications of self-regulation; it is instead aimed at demonstrating how many important debates about governance would proceed differently if we fully incorporated self-regulation into our understanding.
This Article kicks off a now-annual issue of the George Washington Law Review devoted to administrative law. Thanks to the Law Review, we have the luxury of stepping back and taking stock of where we are in our understanding of our field. It is thus a perfect occasion to point out that we students of the administrative state are missing something. Self-regulation is a feature of the landscape that agencies inhabit and it demands our attention. Or so this Article will argue.
