Multiple agencies are all the rage in administrative law. As Professors Jody Freeman and Jim Rossi note, the traditional focus in administrative law has been on investigating how individual agencies function, and how interactions with the White House, Congress, and the courts shape (for better or for worse) their decisionmaking.1 In contrast, the newer scholarship looks at how multiple agencies interact. Some of that scholarship focuses on particular areas of law, calling on policymakers to consider using a combination of multiple agencies, rather than one single agency, to solve particular policy problems.2 Other scholarship is more cross-cutting, trying to identify larger patterns that run across substantive areas, describing how the existence of multiple agencies and their interactions might shape agency decisionmaking and what we might (or might not) want to do about it.3 Freeman and Rossi’s piece, though it draws on a particular case study, has a broader ambition and makes an important and useful contribution to this second category.
This renewed focus on multiple agencies makes a lot of sense, in part because multiple-agency decisionmaking is a fundamental problem. There always will be questions about how to balance between two extremes: should we lump decisionmaking for different issues together within a single agency, providing greater potential for coordination but at the risk of having different decisionmaking processes interfere with each other or of losing the potential benefits of specialization; or should we separate decisionmaking for different issues into multiple organizational units, perhaps making decisionmaking for each individual issue more efficient but at the risk of having agencies get in each other’s way when issues interact?4 So long as there are multiple things for the government to do, there will always be a question about what organizational structure will allow it to be most successful in dealing with the interactions among those different goals.
Take environmental law, for instance. It has been regularly drawn on for examples of how multiple agencies interact and might be used for policymaking,5 perhaps because it presents an extreme example of the problem of many interacting goals. Environmental problems regularly involve externalities, the unintended and unaccounted-for impacts of other socially productive activities; regulating those externalities requires balancing the costs and benefits of the externality-causing activity (for example, oil and gas production) and the costs and benefits of the activities or resources impacted by the externality (for example, wildlife harmed by oil and gas development, which might in turn support recreational hunting). A key question is whether you want to manage the externality-causing activity separately from the externality, or together. The cross-cutting issues present in environmental law mean that there are lots of opportunities to think about using multiple agencies as a way to manage environmental law and policy challenges.
The move in administrative law to consider multiple agencies, rather than individual agencies, as the key unit of analysis (whether descriptive or prescriptive) is therefore important and necessary. It also continues a history in administrative law of building on prior work in political science and economics: just like prior waves of administrative law that built off of public choice theory or positive political theory,6 the multiple agency literature has built off of some key articles from outside the legal academy.7
Despite the prior useful work, including Freeman and Rossi’s piece, there is lots of new terrain still to explore here. In this brief Response, I can only touch on a couple of important questions that still require fuller development in the literature, but there are surely many more to consider.
One key question to answer is: when do we want agencies to have overlapping responsibilities? Different scholars have come to very different answers to this question.8 The answer will likely be highly contingent on many different factors, such as: the relationship between the overlapping agencies (are they collaborative or competitive, do they have complementary goals, or are their goals in tension with each other); the policy area (environmental law might require different structures than securities law); the internal dynamics of the agencies involved (different agencies may have different cultures and professional backgrounds); the political context (different political pressures may shape how agencies act and react to each other); and more.9
Consider relationships among agencies: Two of the kinds of coordination that Freeman and Rossi’s article develops (interagency agreements and joint policymaking) and their leading case study (the Environmental Protection Agency (EPA)—National Highway Traffic Safety Administration (NHTSA) joint rulemaking)10 are examples of collaboration among agencies — in other words, agencies working together to achieve a common goal. This is an important kind of relationship, and a common one (as they point out). But it’s not the only one. Freeman’s prior work has highlighted an example of a different kind of relationship among multiple agencies, the “agency as lobbyist,” in which one agency with one goal uses administrative comments and consultation (among other tools) to attempt to change another agency’s position as that second agency pursues a different, somewhat conflicting goal.11 Professor Anne Joseph O’Connell’s work examined the possibility that competition among agencies might produce better counterterrorism intelligence.12 My work has analyzed an even more combative relationship — the “agency as regulator” — in which one agency retains a veto power over another agency’s decisionmaking.13 Each of these relationships may have very different dynamics14 — for instance, the feasibility and possible impacts of agency capture by interest groups likely varies significantly for each of them.15
A second important set of questions is: What do agencies think about overlapping responsibility? How do agencies respond to each other? When do agencies want to collaborate with each other, and when do they want to compete? The answers to this set of questions are extremely important whether we want to simply understand how multiple agencies interact, or whether we want to design agency structures in order to achieve particular goals.
In answering that question, most of the current literature, including Freeman and Rossi’s article, has generally focused on the external forces that shape agency behavior, such as congressional oversight, budgetary pressures, White House supervision, and the threat of judicial review.16
But understanding how agencies respond to each other will also depend on looking inside agencies, at the internal forces that drive how their employees and leaders respond to outside forces.17 For instance, when will agencies want to collaborate or conduct shared rulemakings on their own without significant external pressure, and when will they need to be pressed and prodded into action? The answers to these questions will be very important to a White House staffer who has to know when she has to spend limited time and energy closely monitoring an agency to make sure it collaborates, or when she can leave the agency to its own devices because, once it has been directed to collaborate, it will want to do so.18 The answers to these questions will also be important for institutional and legal designers who will want to know when they should resort to the (expensive) tools of litigation or the threat of litigation to force agencies to work together, and when they are better off relying on informal efforts among agencies to accomplish coordination and collaboration.
Likewise, there may be situations where it might be desirable to have one agency abdicate its shared responsibility in favor of another agency.19 Again, external forces might encourage or discourage an agency from abdicating, but internal forces might be just as important: Does the dominant professional culture (for example, lawyers, engineers, scientists) within the agency support or resist expansion into a new area of expertise? Does the agency and its employees have a strong orientation around a particular mission,20 and is that mission consistent with expansion or withdrawal from a new policy arena?
Answering these kinds of questions will require understanding how bureaucracies function — and again, there is a rich political science literature that will be helpful to legal scholars as we pursue these questions.21 Unlike prior political science literature that administrative law has drawn upon, this literature may be less mathematical, and more qualitative.22 Research on these questions (whether by legal scholars or political scientists) will also require a lot more empirical research or understanding of how agencies function, and what motivates bureaucrats and political appointees.23
There are some excellent examples of empirical work in administrative law focusing on individual agencies and how they interact with external institutions (such as courts),24 though perhaps not enough of them.25 The addition of questions about multiple agencies in administrative law adds another important set of empirical questions for us to answer in order to understand how agencies really do respond to each other. There is already some of that work to draw on now,26 but certainly more is needed. There is a whole new research agenda out there for administrative law — it’s about time that we followed Freeman and Rossi in pursuing it.
[1] Jody Freeman & Jim Rossi, Agency Coordination in Shared Regulatory Space, 125 Harv. L. Rev. 1131, 1135 (2012). For classics in the field, see for example Jerry L. Mashaw & David L. Harfst, The Struggle for Auto Safety (1990); R. Shep Melnick, Regulation and the Courts (1983); Richard B. Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1667 (1975).
[2] See generally Eric Biber, The Problem of Environmental Monitoring, 83 U. Colo. L. Rev. 1, 66—79 (2011); Jody Freeman & Daniel A. Farber, Modular Environmental Regulation, 54 Duke L.J. 795 (2005); Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314 (2006); Anne Joseph O’Connell, The Architecture of Smart Intelligence: Structuring and Overseeing Agencies in the Post-9/11 World, 94 Calif. L. Rev. 1655 (2006).
[3] See generally Eric Biber, Too Many Things to Do: How to Deal with the Dysfunctions of Multiple-Goal Agencies, 33 Harv. Envtl. L. Rev. 1 (2009); Keith Bradley, The Design of Agency Interactions, 111 Colum. L. Rev. 745 (2011); J.R. DeShazo & Jody Freeman, Public Agencies as Lobbyists, 105 Colum. L. Rev. 2217 (2005); Jacob E. Gersen, Overlapping and Underlapping Jurisdiction in Administrative Law, 2006 Sup. Ct. Rev. 201; Jason Marisam, Duplicative Delegations, 63 Admin. L. Rev. 181 (2011).
[4] See, e.g., David A. Weisbach & Jacob Nussim, The Integration of Tax and Spending Programs, 113 Yale L.J. 955, 985—91 (2004) (developing the trade-off between consolidation and splitting of functions across agencies).
[5] See the examples drawn upon in Marisam, supra note 3, at 10—11, 22, 29—31; see also Biber, supra note 3; Bradley, supra note 3, at 766—70; DeShazo & Freeman, supra note 3.
[6] See generally Daniel A. Farber & Philip P. Frickey, Law and Public Choice (1991) (providing an overview of how public choice theory has influenced legal scholarship); Matthew D. McCubbins et al., Administrative Procedures as Instruments of Political Control, 3 J.L. Econ. & Org. 243 (1987) (seminal political science article in the area of positive political theory, which seeks to provide a descriptive account of how governance structures are created based on the incentives of different political actors); L. & Contemp. Probs. 1, 1—139 (2004) (law review special issue devoted to exploring implications of positive political theory for administrative law).
[7] See generally Michael M. Ting, A Strategic Theory of Bureaucratic Redundancy, 47 Am. J. Pol. Sci. 274 (2003).
[8] Compare, e.g., William W. Buzbee, Recognizing the Regulatory Commons: A Theory of Regulatory Gaps, 89 Iowa L. Rev. 1 (2003) (arguing that overlapping regulatory jurisdiction may lead to free-riding and underregulation by agencies), and Marisam, supra note 3, at 34—42 (critiquing redundancy), with Freeman & Rossi, supra note 1, at 1138—45, 1151—55 (noting benefits of redundancy), Gersen, supra note 3, at 213 (same), and O’Connell, supra note 2, at 1678—90 (same).
[9] For an overview of some of the issues, see Gersen, supra note 3, at 208—11.
[10] Freeman & Rossi, supra note 1, at 1169—73.
[11] DeShazo & Freeman, supra note 3. The example that DeShazo and Freeman draw on is the role fish and wildlife agencies played in making the Federal Energy Regulatory Commission’s hydroelectric dam licensing process take greater consideration of the negative impacts of dams on fish populations. This type of relationship falls within what Freeman and Rossi term “consultation” in their article. See Freeman & Rossi, supra note 1, at 1157—58.
[12] See O’Connell, supra note 2.
[13] See Biber, supra note 3, at 45—58. The Fish and Wildlife Service’s (FWS) ability to effectively veto proposed actions by other federal agencies under the Endangered Species Act is an example of this kind of relationship. This type of relationship would also fall within what Freeman and Rossi term “consultation.”
[14] All three of these examples are also better understood as points on a continuum rather than fundamentally different categories.
[15] For instance, if an interest group wanted to stop an activity, then capturing the veto-holding agency in the “agency as regulator” relationship would be necessary and sufficient; but where two agencies have overlapping jurisdiction and each can independently act without the other, to stop an activity the interest group must capture both agencies.
[16] See Dara Kay Cohen et al., Crisis Bureaucracy: Homeland Security and the Political Design of Legal Mandates, 59 Stan. L. Rev. 673, 700—02 (2006) (noting this pattern in the relevant scholarship on multiple agencies).
[17] For examples of administrative law literature that does look at internal forces in the context of multiple agencies, see Biber, supra note 3, at 17—30, 41—60; and DeShazo & Freeman, supra note 3, at 2239—41.
[18] Agencies may even collaborate voluntarily, as Freeman and Rossi note. Freeman & Rossi, supra note 1, at 1161—73.
[19] See Marisam, supra note 3, at 239—40.
[20] For discussion of the concept of agency mission, see James Q. Wilson, Bureaucracy 101, 157—58, 371—72 (1989).
[21] See, e.g., John Brehm & Scott Gates, Working, Shirking, and Sabotage (1997); Marissa Martino Golden, What Motivates Bureaucrats? (2000); Wilson, supra note 21.
[22] Compare Ting, supra note 7 (an example of a highly quantitative article), with Wilson, supra note 21 (primarily relying on qualitative data and analysis).
[23] For an example of the kinds of questions that will need to be answered to understand how both external and internal forces shape agency interactions, see O’Connell, supra note 2, at 1688.
[24] See, e.g., Mashaw & Harfst, supra note 1; Melnick, supra note 1. Others have done pathbreaking work collecting large-scale datasets that will be invaluable for understanding how agencies function either on their own or in coordination with other agencies. See, e.g., Jacob E. Gersen & Anne Joseph O’Connell, Hiding in Plain Sight? Timing and Transparency in the Administrative State, 76 U. Chi. L. Rev. 1157 (2009); Anne Joseph O’Connell, Political Cycles of Rulemaking: An Empirical Portrait of the Modern Administrative State, 94 Va. L. Rev. 889 (2008); Anne Joseph O’Connell, Vacant Offices: Delays in Staffing Top Agency Positions, 82 S. Cal. L. Rev. 913 (2009).
[25] See, e.g., Cary Coglianese, Empirical Analysis and Administrative Law, 2002 U. Ill. L. Rev. 1111, 1137 (calling for more empirical work in administrative law).
[26] See, e.g., Jonathan B. Bendor, Parallel Systems (1985); Robert F. Durant, When the Government Regulates Itself (1985); Wilson, supra note 21; Eugene Bardach, Turf Barriers to Interagency Collaboration, in The State of Public Management 168 (Donald F. Kettl & H. Brinton Milward, eds., 1996). There is also exciting theoretical work exploring the implications of interagency interactions. See, e.g., Mathias Dewatripont & Jean Tirole, Advocates, 107 J. Pol. Econ. 1 (1999); Ting, supra note 7; Sean Gailmard & John W. Patty, Stovepiping (Jan. 20, 2012) (unpublished manuscript) (on file with the Harvard Law School Library).