The Yale Law Journal


Can Interagency Dialogue Serve as the New Separation of Powers?

03 Nov 2006

In Internal Separation of Powers, an essay recently published in this Journal, Neal Katyal adds his own distinctive twist to the debates about the growth of presidential power by suggesting reforms within the executive branch that would cabin executive discretion without violating the Constitution’s commitment to a unitary executive. I commend him for taking separation of powers principles seriously, rather than excoriating them as an obstruction or as a pretext for presidential aggrandizement. At the same time, one can view the same landscape and draw strikingly different conclusions about both the scope of the problem and the appropriate remedies.

I. Congress as a Check on Executive Power

Professor Katyal’s laments what he sees as the “demise of the congressional checking function.”According to Katyal, the collapse of the nondelegation doctrine and the legislative veto, as well as courts’ willingness to defer to the executive on foreign affairs, have left Congress with “a paucity of weapons that check the President.” These problems are exacerbated by the growth of the federal bureaucracy and are most acute when the same party controls both the White House and Capitol Hill. To compensate for what he sees as Congress’s abdication of its role, Professor Katyal proposes a series of structural reforms within the executive branch, such as overlapping agency jurisdiction and interagency consultation, greater professionalization of the civil service, the establishment of an independent Director of Adjudication to resolve interagency disputes, and the appointment of agency ombudsmen.

But has the balance of power actually tipped so decisively away from Congress and toward the executive? To be sure, the federal executive branch has increased in size and strength over the last hundred years. But the fourfold increase in the number of federal executive employees since the 1930s was driven in part by an increase in population, which more than doubled over the same period. That growth was also the result of the shift in power from the states to the federal government, as evidenced by the nearly threefold increase in the number of congressional employees and the nineteenfold increase in the number of federal judicial employees over the same period. When considered in the context of the enormous increase of all federal power over the last seventy-five years, the executive branch’s rapid growth seems less ominous than appears at first glance.

The modern Congress has also developed new tools to restrain the larger executive branch. Since at least the 1970s, Congress has increasingly relied on less formal controls on agency behavior, conducting more oversight hearings and investigations, enacting appropriations riders and other legislative provisions detailing how executive officials should administer the law, including language in floor statements and committee reports to influence agency behavior, writing agency officials about the resolution of particular cases, and extracting concessions during administrators’ confirmation hearings. Use of these practices has remained widespread even during periods of single-party government. Though Professor Katyal notes the importance of such tools when he calls for increased reporting requirements and minority oversight hearings during periods of single-party government, he places too little emphasis on them.

Moreover, Congress has already begun to employ the type of multi-agency consultations proposed by Professor Katyal. For example, the Telecommunications Act of 1996 requires that the Federal Communications Commission consult with the Attorney General before allowing former “baby Bells” to sell long-distance services within their home regions. The drug statutes similarly require the Attorney General to seek the opinion of the Secretary of Health and Human Services before adding a drug to the federal government’s list of controlled substances. This is in addition to routine statutory limits on agency jurisdiction, as well as laws, such as the Administrative Procedure Act of 1946 and the National Environmental Policy Act of 1969, which have long restricted the exercise of executive discretion.

This is not to say that Professor Katyal’s assessment of the current balance of power between the branches is necessarily wrong. My point is rather that we should also consider the developments that tend to counterbalance the ones he highlights.

II. Bureaucracy as an Alternative Check on Executive Power

Professor Katyal’s primary solution to Congress’s perceived inability to oversee the executive branch is bureaucracy. In his eyes, bureaucrats are more than just purveyors of red tape. They are repositories of expertise and the embodiment of a “tradition-bound professionalism” that frames decisions in a longer-term perspective that transcends transient political exigencies.

This is just one of many visions of bureaucracy, and the conclusions that one draws about the consequences of empowering the bureaucracy depend heavily on the theory one espouses. Consider first the seminal work of Marver Bernstein,who characterizes agencies as going through a life cycle. During its early stages, an agency aggressively pursues the public interest. Over time, the agency loses vitality and becomes increasingly aligned with the industry it is charged with regulating—an outcome also predicted by “capture theory.”On this view, whether interagency consultation would promote expertise and a longer-term perspective would depend on the life-cycle stage of each agency involved in the decision.

William Niskanen posits that bureaucrats can best be understood as empire builders intent on maximizing their budgets. The result is systematic overproduction of government services, with executive officials tending to become advocates for the programs they represent and resistant to presidential control. Interestingly, Niskanen proposes competition among agencies, not just for consultation, but for the provision of government services. The benefits result not from expertise or a longer-term perspective, but rather from the discipline provided by competition for funding, which Niskanen regarded as far more effective than presidential oversight in restraining agency behavior.

The determinative role played by the particular theory of bureaucracy used to analyze the problem is most clearly illustrated by comparing Professor Katyal’s proposal to one offered by Lawrence Lessig and Cass Sunstein. Although Lessig and Sunstein share Professor Katyal’s belief that the executive branch now exercises greater control over policymaking, they reject Katyal’s progressive faith in bureaucratic expertise, which they regard as discredited and somewhat quaint, in favor of the view that administrative judgments are inherently political. As a result, they believe that the increasing involvement of the federal bureaucracy in policymaking justifies strengthening presidential control over executive agencies. Although I do not agree with all of Lessig and Sunstein’s analysis, I find it quite striking that the differences in their underlying theory of bureaucracy cause them to reach conclusions that diverge sharply from Professor Katyal’s even though they begin from a similar starting point.

III. The Costs of Partitioning Executive Authority

As Professor Katyal himself frequently notes, the primary cost of his proposal is the loss of executive energy and accountability. He claims that these losses are justified by the greater experimentation and the more complete presentation of the issues when multiple agencies are involved in a decision.

I would not be so quick to dismiss the cost. The nature of the presidency is best understood in light of the Framers’ frustration with the weakness of the plural executive under the Articles of Confederation. As a result, the Constitution consciously established a unitary executive in order to promote energy and accountability. Globalization and the accelerating pace of technological change have arguably made the need for strong central authority all the more acute.

The following example illustrates the potential pitfalls of dividing authority among too many agencies. Until the Harding Administration, the President lacked the power to develop an integrated federal budget. Instead, each agency submitted its own proposal, which the Secretary of the Treasury assembled into a “Book of Estimates.” Predictably, each agency generally favored increasing its own funding, and budgetary initiatives lacked coordination.

A 1921 statute centralized the budget process in a new White House Bureau headed by a director who answered only to the President. Given the fundamental connection between funding and policy prioritization, this change helped make federal policy more coherent. The fact that during its first year the Bureau of the Budget saved over $1 billion underscores that budgeting is a task better suited to unitary rather than collective decisionmaking.


Some of the positions taken by the current Administration make the growth of executive power seem like a new and dangerous problem. In reality, however, the legislative-executive balance of power has ebbed and flowed over time. The strong presidency of Abraham Lincoln was preceded and followed by periods of legislative domination. Similarly, Watergate spawned a period of strong Congresses—during which many innovations upon which Professor Katyal relies, such as independent counsels and inspectors general, reached their apogee—which was followed in turn by a return to presidential ascendancy under Ronald Reagan. And even during periods of single-party government, such as the Franklin Delano Roosevelt, Kennedy, Johnson, and Carter Administrations, as well as portions of the Truman and Eisenhower Administrations, members of Congress influenced administrative agencies and distanced themselves from unpopular executive actions.History therefore suggests that politics, and not institutional structure, best explains the shifts in power between Congress and the President.

Professor Katyal himself seems to recognize, at least implicitly, this lesson of history when he notes that even if Congress were to create a Director of Adjudication to oversee competing agency claims, the President would nevertheless remain free to disregard the results. In the final analysis, even Katyal recognizes that structural solutions can only do so much and that the ultimate protections against presidential power are political rather than legal.

Christopher S. Yoo is Professor of Law at Vanderbilt University. He is the co-author of a book on the history of the unitary executive that is forthcoming from the Yale University Press.

Preferred Citation: Christopher S. Yoo, Can Interagency Dialogue Serve as the New Separation of Powers?, 116 Yale L.J. Pocket Part 131 (2006),

[Editor's Note: Can Interagency Dialogue Serve as the New Separation of Powers? is a Response to Neal Kumar Katyal, Internal Separation of Powers: Checking Today’s Most Dangerous Branch from Within, 115 Yale L.J. 2314 (2006).]