ACTION MEMORANDUM
TO: The President of the United States
CC: National Security Advisor Gen. James L. Jones
FROM: Michael Singh
DATE: November 1, 2009
SUBJECT: Making the NSC Work
The need to reform the U.S. national security decision and implementation process is widely acknowledged in Washington. Everyone who has participated in the process at a high enough level understands that whenever a problem spills over the competency boundary of any single bureaucracy (as more and more problems seem to do these days), the policy outcome often ends up reflecting less than the sum of its governmental parts. The U.S. government struggles to integrate its assets and perspectives, frequently failing to achieve the unity of effort, let alone unity of command, needed to manage complex problems. This is not to claim that organizational deficits mark the only way things go wrong in U.S. foreign policy. But it’s one way they go wrong, and not a trivial one.
Acknowledging the need for process reform is one thing; actually achieving it something else. At the heart of the challenge is a specific conundrum. As I.M. Destler observed more than a quarter century ago, the problem with the national security process, and the solution to that problem, lie with the President. From a President’s style and key personnel flow a cascade of repercussions for the entire national security system. This was true before 1947 as well as since.1 It matters whether a President wants to control foreign policy himself, à la Wilson or Kennedy, whether he delegates that control to a strong Secretary of State, à la Harding or Ford, or whether he partially delegates it to a National Security Advisor (NSA), à la Nixon between 1969 and April 1973. The origins of presidential inclinations in this regard are manifold, ranging from personal experience to a desire to emulate or eschew the qualities of predecessors. These inclinations can lead the national security process at the highest level to be harmonious or disastrous.
Recognizing Destler’s paradox, most national security reform proposals in recent decades have at once recognized the central role of the President while struggling to accommodate it. These proposals have generally sought either to guide the President’s preferences or the National Security Advisor’s approach to his or her duties, or to suggest structural or legislative changes to their offices that would implicitly constrain them. Since both approaches would reduce the historical flexibility inherent in the President’s role, neither has gotten very far. Sweeping structural change proposed by outsiders is politically impractical because Presidents come into office with their own ideas about how to manage their duties. Changes occur when a new President himself wants them, and those that endure do so when they prove useful enough to appeal to successors.
This pocket history begs the question: Is there a way to reform the national security process independent of the important but long-range task of perfecting the national security apparatus’s constituent parts and their means of interaction? In other words, is it possible to reform the top of the process, lodged mainly in the National Security Council, without reforming the entire interagency? The answer is “yes”, and a guiding principle informs it: Officials should focus less on where the buck stops and more on the path it takes to get there. Your National Security Advisor, General Jones, has expressed his desire for an National Security Council that is “very agile”, “very flat”, and that serves you “in a very efficient way.” That’s the right idea, but we’re not there yet.
Three Shortcomings in the Current System
Veterans of the U.S. national security process could be forgiven for sometimes wondering why it exists at all, given its tendency to produce hours of meetings and mountains of paper that seem to have little bearing on actual decision outcomes. The process is supposed to bring together disparate parts of the Executive Branch to forge coherent policies. It aims generally at protective inclusivity, the reasoning being that all views on an issue need to be heard to protect ourselves from error.
The problem with protective inclusivity is that it tends to produce so many multi-tiered deliberations that the ponderous process outweighs its benefits. Protective inclusivity worked well enough during the Cold War, when the NSC system was born and came of age. But today U.S. engagement in global affairs is less focused on a singular great-power rivalry. The number of (often interrelated) subjects over which policymakers must deliberate has risen sharply, even as the amount of time they have to come to a decision has shrunk. We must therefore re-conceive national security decision processes to remake coordination and inclusivity from ends in themselves into means to achieve better results. Reformers should keep in mind three goals: integration, innovation and prioritization.
Integration: The need to integrate U.S. national security policy across the realms of diplomacy, defense, intelligence and other disciplines motivated, in part, the creation of the National Security Council in 1947, as well as its gradual expansion thereafter. But there is more than one way to think about integration. The current National Security Advisor, General Jones, described the Obama Administration’s approach to integration in his May 29 remarks to the Atlantic Council. According to Jones, policy today is made “starting with working groups . . . with appropriate participation from all interested parties”, and then moves to the deputies’ committee, “where, again, at a higher level we focus on the issue, on making sure that all claimants are represented.”2 Policy discussion then moves to a principals’ committee, and ultimately moves to a full NSC meeting “if the President needs to make a decision.”
This bottom-up approach resembles a dispute-resolution mechanism more than it does a strategic approach to policymaking. The lower levels of the chain develop a policy core, to which are appended disputes whose resolution exceeds the participants’ limited authority. These are both passed up the chain, with the focus of more senior meetings naturally alighting on the disputes, which are resolved on the basis of interagency deal-making as often as they are on a careful analysis of underlying assumptions and the full range of policy options. While this sort of system involves much deliberation, it is unlikely to lead to actual integration. Instead, it will more often lead to an inadequate consideration of policy alternatives, and to the over-involvement of higher-level officials in policy minutiae at the expense of strategic oversight and direction. Worst of all, it presents the President with a false sense of consensus. Arguments and compromises are hidden from his view, and his options are needlessly restricted.
Innovation: Innovation is not typically thought of as part of the bureaucrat’s job description, but the constant generation of fresh policy perspectives is vital to preventing stagnation and ossification in national security policy. As we have seen in recent months, innovation can be justifiably the highest priority of any incoming Administration. While policy innovation is inherently a creative task, it is not dependent simply on inspiration or luck; rather, it should be purposefully encouraged through management of the policy process.
The current national security process, however, does not encourage innovation. The relentless pace of meetings and the virtual crush of informal e-mail communications encourage an “inbox” mentality in which responding to taskings supplants creative thinking. Those charged with “big thinking”—strategic planners and their ilk—interact little with those who have day-to-day policy and operational responsibility. The result is a national security system poorly prepared to deal with contingencies, adapt to changing circumstances and incorporate new tools into its policy arsenal. Breaking out of this mentality is challenging. The press of time and the “stovepiping” of issues generally prevents U.S. officials from learning from one another or from their own history.
Prioritization: Contributing to the pressures on the national security process and its attendant lack of success in policy integration and innovation is a growing failure on the part of the U.S. government to set priorities in its engagement with the world. When everything is important, nothing is; core U.S. interests and peripheral ones receive equal time, with little to guide senior officials as to where they should direct their energy. The absence of prioritization reflects a deficit of strategic thinking in the national security system, which, if left unaddressed, will blunt the effectiveness of policy initiatives and will diffuse U.S. power in the international arena.
The manifold challenges that have beset the U.S. government in forming successful national security strategies since the end of the Cold War deserve separate treatment. Suffice it to say for present purposes that the strategic planning process has often not served policymakers well. A strategy, to be useful, must aid not only the allocation of resources but also the management of tradeoffs that is inevitable and essential in policymaking. If “to govern is to choose”, as the Duc de Levis observed, strategy should above all guide the burden of choice. Rather than the product of an isolated planning cell or a bottom-up collection of pet projects and resource requests, a strategy should be short and clear, set and enforced by the President and his top national security aides, and refined by those officials responsible for transforming it from aspiration into reality.
Four Proposals
The inability of the current national security process to integrate, innovate and properly prioritize serves as a drag on policymaking at a time when speed and alacrity are sorely needed. Fortunately, reforms that stop well short of the sweeping changes proposed by the Project on National Security Reform and others can restore these attributes to the process in significant measure. To that end, the President should consider the following four proposals.
Establish Cross-functional NSC-led Teams: To increase the effectiveness of the national security process, mid-level officials should be organized into cross-functional teams (CFTs) focusing on White House priorities. CFTs, which are commonplace in the private-sector but not in the rigidly hierarchical national security apparatus, would allow officials with similar portfolios to increase the time spent with one another, and decrease the structural constraints imposed upon their collaboration.
The aim of the CFTs would be to provide the fullest possible range of policy options for review by higher-level officials, with explicit attention given to assumptions and the basis for any disagreements between CFT members. CFTs would be led by an NSC director, and include in their membership the key regional or functional specialists from across the Executive Branch. Membership would be determined at the NSC on an issue-by-issue basis (so the Russia Team might include arms control experts, and the China team would certainly include international economists), and could change over time.
CFT members would be expected to use their time together—10–20 percent of the work week, with the remainder spent at their home agency engaged in traditional work—not only to prepare policy papers but also to innovate through brainstorming, meeting with “red cell” teams of contrarians to challenge their work, and consulting with other CFTs working on related issues to share lessons learned. Benefits would flow for integration, innovation and prioritization not simply because of greater interaction between the CFT members, but because of the trust that would develop among them (as in the case of informal coordinating mechanisms used by Presidents and their top aides).
Crucially, CFTs would not be drawn from a separate corps of “national security professionals”, as has been recommended by the Hart-Rudman Commission, the PNSR project and others. There may be a role for such a corps in reducing interagency friction, but not at the level we are discussing here. The power of CFTs would lie in the mingling of specialists rather than the creation of a new class of generalists. The right model for a CFT is a joint venture, not a bureaucratic merger. Retaining agency affiliations is also vital to attracting quality staff—an agency promises a rewarding and varied career, something that professional service in a national security corps cannot as readily do.
Invert the Interagency Pyramid: Rather than taking a progressively narrower view of policymaking as an issue moves up a chain, the national security process should ensure that successive levels of the system take an increasingly broader view. CFTs would aid that goal, since they would provide senior officials not with a consensus policy and associated unresolved disputes, but a fuller treatment of options and the assumptions underlying them.
The senior-level group, however, should be expected to do more than simply refine the CFT’s work. It must exercise strategic stewardship over policy. In other words, it must ensure that the policy proposals received from the CFT support overall U.S. strategy. Ideally, this would mean that the group would consider not only the policy recommendations regarding the topic at hand, but also recommendations from other related CFTs. (For example, it would examine the product of an Afghanistan CFT alongside that of both a Pakistan CFT and perhaps a transnational crime CFT.) The resulting overall policy should be a tapestry rather than a patchwork, with connections among subject matters explained and understood.
By implication, this inversion of the interagency pyramid precludes a pervasive special envoy system. To provide adequate strategic direction to national security policy, officials attending senior-level meetings must have a broad enough portfolio to grasp the big picture, and must be insulated from conflicts of interest and parochial concerns. The appointment of a special envoy should be the exception rather than the rule, and should not stem simply from the desire to emphasize the importance of an issue. When necessary, envoys should participate in senior-level meetings alongside the relevant regional officials, not in place of them.
Integrate Strategic Planning: Just as senior-level officials must be responsible for maintaining strategic direction in the national security process, the most senior officials—the President and his top aides—must be directly responsible for developing that strategy as their first order of national security business upon entering office.3 Top officials should personally articulate the envisioned strategy, which, after all, rests in the concatenation of decisions they themselves make. With an overarching strategy in place, select mid-level officials should flesh out how they will implement it over their areas of responsibility. This would not only provide guidance to officials further down the chain, it would help the American public to hold their leaders accountable. National security institutions should provide both continuity and coherence, but left to their own devices agencies such as State and Defense will at best provide only the former. Only sound strategic planning can enable them to also provide the latter.
Streamline the Bureaucracy: It is frequently asserted that, to function properly, the national security system (apart, perhaps, from the Defense Department) needs more resources—more people and more money. The PNSR study suggests increasing the scope both of the NSC and the State Department, and the American Academy of Diplomacy proposed a 46 percent increase in State/USAID staffing from 2010–14.4 One study even proposed a direct relationship between staffing levels and policy effectiveness, stating that “the number of people ‘on the case’ . . . determines how effectively the U.S. government can manage relations with the rest of the world.”5
It is true that too few resources can starve a decision system. But if resources are added without careful attention to the efficient functioning of individual agencies and the overall national security process, larger staffs and new bureaucracies will make things worse even as they waste money. We would be much wiser to focus on resource utilization than on total resources committed. Leaner national security institutions such as the NSC staff have generally proven more agile policymakers than the larger agencies, which are frequently gripped by inertia. Richard Holbrooke, in a pitch for government reform written nearly forty years ago, recognized this when he wrote that “size—sheer, unmanageable size—is the root problem in Washington and overseas today.”6
To achieve these qualities throughout government, two things need to happen. First, a system-wide “de-layering” needs to take place. This process, which has been effective in the private sector, brings a chief executive in closer contact with his “line managers” and promotes speed, efficiency and relevance in policymaking. Second, the number of policymaking positions should be reduced, with the goal of providing individuals with greater responsibility and compressing the distance between them and the top decision-makers. This will not only conserve resources; it will boost staff morale and retention and perhaps even allow the creation of a civilian personnel “float” for increased training without additional hiring. The result would be better, more responsive policy with potentially fewer resources, or at least with the confidence that every dollar invested in the national security apparatus was earning a healthy return.
* * *
A better national security process and a more effective, efficient national security apparatus are eminently achievable. But to attain these goals, top U.S. officials must put aside many of their instinctive responses to the system’s failings, such as developing ad hoc mechanisms to circumvent an unwieldy process; creating new bureaucracies, layers or envoys in a crude organizational fix; or throwing more resources at a poorly functioning apparatus. Instead, they should streamline national security institutions, break down bureaucratic barriers between them, and above all provide them with clear strategic direction. The basic idea is to reduce the volume of exchanges among policymakers while increasing their quality and purposefulness. In a world that is increasingly complicated, we should strive in designing our national security process for ruthless simplicity.
2A deputies’ committee is composed of relevant Deputy or Undersecretaries, whereas a principals’ committee is a meeting of cabinet Secretaries. The former is generally chaired by the Deputy National Security Advisor, the latter by the National Security Advisor. This national security system configuration arose under President George H.W. Bush and his National Security Advisor, Brent Scowcroft.
3See Paul Lettow and Tom Mahnken, “Getting Serious about Strategic Planning”, The American Interest (September/October 2009).
4“A Foreign Affairs Budget for the Future”, American Academy of Diplomacy (October 2008).
5David Shorr, Derek Chollet and Vikram Singh, “The Civilian Core of American Power”, Foreign Service Journal (December 2008), pp. 28–33.
6Richard Holbrooke, “The Machine That Fails”, Foreign Policy (Winter 1970–71), pp. 65–77.