Reflections: Improving the Interagency Process
February 23, 2016
As I mentioned in the first essay, CSIS played a substantial role in laying the foundation for the landmark Goldwater-Nichols legislation. Last December, we decided we needed to assemble leading defense intellectuals to help support the rising interest in defense reform. We recently polled that group and found that the greatest frustration and perceived need for reform is in the interagency coordination process. For as long as I have worked in Washington, I have heard complaints about the interagency process. The complaining became turbocharged because of frustration with the wars we fought during the last 12 years. I had a more intimate understanding of the interagency process for the brief period I was deputy secretary of defense. These reflections represent both an engaged and a detached perspective.
People complain about the interagency process for various reasons. One group decries the excessive time demanded by meetings that rehash the issues yet another time, with no forward movement in decisionmaking or concrete results. Others complain that the process is obsessed with formulating broad policy guidelines but rarely deals with policy implementation. Yet other critics argue that the focus is too granular, grinding through details that rightly should be left to lower level staff charged with managing details, leaving too little time for reflection on genuinely strategic issues.
Almost everyone uses grand concepts like “whole of government,” appealing to an obvious point that modern, complex life requires actions from a range of departments and bureaus. Most often, the appeal for whole-of-government approaches is a not so hidden critique directed against other bureaus and agencies.
Criticism of the interagency process has spiked during the Obama presidency. I often have senior government officials lament spending four to five hours a day in National Security Council (NSC) meetings in the White House. When I was deputy secretary of defense, I would estimate I spent 10 hours a week in White House meetings. No doubt this increase is caused by the astounding ballooning in the size of the NSC staff—now over 450 strong. (Zbig Brzezinski told me recently he had 60 staff and thought that to be a bit too large.)
The interagency process sits on top of a constitutional fault line in American governance.
Despite the pathologies of the current interagency process, there is a much deeper issue that underlies the problem. The interagency process sits on top of a constitutional fault line in American governance.
There is no question that the Congress has the recognized authority to supervise the work of the departments of the executive branch. Senior political appointees must be confirmed by the Senate. They depend on appropriations provided only by the Congress (a power deeply eroded when Congress can’t muster support for anything more than continuing resolutions). Congress has subpoena power to compel testimony from executive branch officers. These are established powers and authorities given to the Congress.
At the same time, the Supreme Court has judged that the president has the right of privacy for his deliberations. The Congress cannot force the president to testify on anything. Deliberations in the president’s office are legally privileged vis-à-vis congressional oversight. To be sure, there has been a continuing struggle to define the boundaries of the president’s office. Does it apply only to people working in the executive mansion? To the Executive Office of the President, which includes major institutions such as the National Security Council, the Office of Management and Budget (OMB) and the Office of the U.S. Trade Representative?
With the exception of a small number of very senior people—such as the director of OMB or the U.S. trade representative, most of these officers are not confirmed by the Senate and cannot automatically be compelled to testify. The Supreme Court has generally shied away from defining the border line between the privacy privilege of the presidency and the oversight rights of the Congress.
We can’t solve the constitutional problem underlying weaknesses in the interagency process. But we can encourage presidents to strengthen their own hand.
The interagency process lies at the crux of this constitutional fault line. The Congress can compel the secretary of defense to testify but not about deliberations within the NSC process.
Much of the modern national security establishment in the White House is less than 50 years old, but it has now become large and elaborate. In an era with instant communications, American presidents have increasingly pulled decisionmaking into the White House, with cabinet officers assigned to an outer ring. The Obama administration has famously pushed cabinet officers to the dimmer periphery, concentrating the (positive) light intensely on the White House. In fairness, President Obama has had to deal with feckless opposition during much of his tenure, which reinforces the desire to find a quiet, private space for collaboration.
So, how do we improve the interagency process? Can Congress legislate it? Not in my judgment. If I were still back in the Defense Department, I would fiercely resist congressional pressures to compromise the presidential privilege of privacy in decision-making. If I were back on the staff of the Senate Armed Services Committee, I would continue to find ways to put pressure on the Defense Department to behave better (meaning, follow my preferences) in the interagency process.
This is a problem that cannot be resolved, in my mind. It is implicit in our form of constitutional government, which is a presidential democracy, not a parliamentary or Westminster democracy. Thus, we are doomed to argue endlessly about improving the interagency process, knowing that it will always reflect the preferences of the current incumbent of the executive mansion.
But I do have one suggestion: Cap the staff of the NSC at 50 people! You can find 50 talented individuals, but you can’t find 450. (75 staff are permanently authorized, but in recent years presidents have ballooned the staff using detailees seconded from cabinet departments, which contributes to the micromanagement in my view.) Presidents do need expert advice immediately within their orbit. 50 talented individuals would be able to focus on genuinely strategic matters. A battalion size of largely average staffers will focus endlessly on administrative details that should be rightly left in cabinet departments.
If a president doesn’t trust his cabinet secretaries and makes the NSC a private national security staff to compete with Defense and State Departments instead of coordinating the work of cabinet departments, then the Congress has a right to insist on confirming senior NSC officials and controlling its budget.
Presidents make themselves weaker by having larger NSC staffs. We can’t solve the constitutional problem underlying weaknesses in the interagency process. But we can encourage presidents to strengthen their own hand.