Too Big to Succeed? Why (W)Hole-of-Government Cannot Work for U.S. Development Policy

October 05, 2010
This post originally appeared on the Huffington Post.While talking beautifully about its development plans, the Administration is really not living up to its rhetoric of elevating development to equal status with diplomacy and defense, the so-called 3Ds. If development is really such an equal partner alongside defense and diplomacy, why is USAID increasingly a minor subcomponent of the State department? The promises of making USAID the “world’s premier development agency” are ringing embarrassingly hollow. How can an agency be influential when it doesn’t even control its own budget or set its own strategic priorities? Even in areas where USAID has traditionally been very strong—disaster relief and food security, for instance—the State Department has taken over. (The Feed-the-Future initiative is effectively directed by State and, despite early promises that USAID would lead on Haitian earthquake relief and reconstruction, it was recently leaked that a State coordinator is running the show.) And it should hardly be surprising that USAID is getting its lunch eaten in the interagency when it had no head for a year and, nearly two years in, still has less than half of its top managers on the job.But what if the problems of the 3Ds aren’t really about the staff vacancies, the battle of Washington egos, and an empire-building State Department? What if the real problem is that the much-vaunted “whole-of-government” approach is fundamentally unworkable in the United States?The idea behind whole-of-government seems sensible enough: lots of federal agencies have skills and resources and experience that can be brought to bear on complex problems. If we can get everyone in the same room and all in the same boat, then the USG effort can be greater than just lots of agencies all running around doing their own thing, right? This seems especially attractive in development policy, where the United States may be involved in helping foreign countries improve health, education, agriculture, transportation, democracy, security, financial regulation, and loads more. If we want to help entrepreneurs in Liberia, why not bring in USAID, the Treasury, the Commerce Department, the US Trade Representative, and the Small Business Administration? If we’re fighting HIV/AIDS in Uganda, let’s use the expertise of HHS, CDC, NIH, and the FDA, right? We’ve now got at least 26 agencies involved in foreign aid of one kind or another. But the room is starting to look a bit crowded now. (Brody, you’re gonna need a bigger boat.)Let’s charitably assume that whole-of-government may work for the Swedes, Australians, and maybe even the British. They all use some kind of interagency apparatus to a tackle a range of issues, from fragile states to counter-terrorism to development. This is possible in some cases because the governments and budgets are small enough to be manageable. In the case of the UK, its Department for International Development is autonomous, well-staffed, has its own strategies and budget, and is led by a cabinet minister, so it can hold its own on Whitehall.But in the United States—with its sprawling federal structure and huge agency staffs and budget—just getting everyone around one table is perhaps too much to ask. The interagency process in any country is a strain. (Managing those tensions is actually what policymaking is all about.) Yet the process can become convoluted and bogged down when the scale is out of whack. Simply put: when you have too many people at the table, nothing gets done.During my short time in government (at State 2007-08) I lived through this regularly, but I think of one surreal experience often. A minor interagency decision over the trade program eligibility of Comoros (population 600,000, annual trade with the U.S. less than $3 million) became deadlocked over a minor issue by a minor agency. A seemingly trivial decision absorbed hours upon hours of staff time by dozens of government officials over many months, with the intransigence of one agency forcing the decision up the ladder until it finally reached the Secretary of State. It was a sad example of dysfunctional decision making, at root because the process (in this case, dictated by Congress) put far too many people with divergent interests in a position to block any resolution.The current confusion and holdups in U.S. development policy are usually explained, like the gossip pages, as who is elbowing whom. But the problem is not just the difficult personalities in every administration. The dynamics of interagency negotiating make this kind of deadlock all too common. Intransigence is simply built into the system. Each staffer in the interagency is sent with the same clear instructions: push our agenda and protect our equities. Understandably, no one wants to go back to their boss to say “I caved, so we lost.” The result is frequent paralysis. These problems usually can get worked out informally, through backchannels, if enough of the players know each other and can find a quiet compromise or horse-trade. But that breaks down completely if there are two dozen or more people in the room and they are strangers. The size of the crowd alone can lead to stalemate.The worrying thing is that under the current administration, this seems to be getting worse. Rather than limit the players around the table in order to get movement, Team Obama has embraced wholeheartedly broad interagency decision making. A friend who now works as a senior official for a large agency was recently needling me by proudly recounting how much better the current administration is than the last one which I worked for: “Unlike under Bush, we don’t just get three agencies to make a decision and then force everyone else to go along. We get all the agencies together to decide collectively.” “Sounds good, but doesn’t that make things impossible?” I asked. Without missing a beat: “Well, yes. I have been log jammed for over a year in the interagency on a key policy decision. We finally broke it by getting [3 federal agencies] together to finally decide.” Oh yeah, this sounds much better.If the whole-of-government model is fundamentally impractical here, how can the interagency still be made to work in practice? It seems to me this can work when there are clear lines of authority and someone influential is empowered to break deadlocks when necessary. In most cases this can only come from the White House. In fact, when the President’s National Security Council (NSC) staff calls an interagency meeting and makes known what the President wants to happen, things can get done. Usually, it is only the White House can tell agencies to do something they don’t want to do. The problem of course is that the President can only focus on so many priority issues at once. And the relatively small NSC staff (roughly 150 people versus about 19,000 at the State Department) cannot possibly coordinate and knock heads on every issue.So what to do? A first step is to abandon the naive idea that it’s best to have everyone in the room as often as possible. In fact, given the colossal nature of the U.S. federal government and accepting interagency turf wars as wholly rational, the opposite seems truer: have the minimum number of people in the room you need to make and implement a sensible decision. If the people or resources of other agencies are needed, then the NSC has to find a way to make it happen.More importantly, drop the whole-of-government fetish and figure out a sensible structure with clear strategic, budget, and reporting lines of authority. If USAID is going to be absorbed into the State Department, then just do it and stop pretending it’s not happening. Then you can have a “super-State” work with NSC to direct the other agencies to fall into line. By definition, this means U.S. development policy will be to directly serve U.S. diplomatic and security interests. Thus we are back to a 2D world, and so be it.Alternatively, if the administration truly believes in the 3Ds, it needs an agency that takes a long-term development perspective—and is able to hold its own in hand-to-hand interagency combat. At a minimum, this requires some kind of separate agency with its own budget and authorities. Such an agency should absorb most of the little development components, including the relevant staff and budgets, of the various agencies sprinkled across the executive branch. High-minded speeches and simply promising to invite USAID to occasional NSC meetings is not enough.At some point, I suspect the folly of whole-of-government will be more obvious. What comes next however is far less so. In the meantime, the gaps between rhetoric and reality are only likely to grow wider.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.