Foreign Policy Research Institute A Nation Must Think Before it Acts America’s Broken Interagency

America’s Broken Interagency

The last job I had with the Bush administration was coordinator for police training, judicial reform, and counternarcotics in Afghanistan. When I got the job, the National Security Council said, “It’s got three parts. First, you have to go to Afghanistan and try to coordinate among their agencies for police reform, judicial reform, and counternarcotics. Then you fly to Europe to coordinate with the EU on the same issue. Finally, you come back to Washington and coordinate U.S. interagency.” The last of these jobs was the most difficult one.

Afghanistan’s interagency process could best be described as “uncoordinated lack of action.” For example, in the areas of police training or counternarcotics, the ministry of the interior and the ministry of counternarcotics were supposed to coordinate their activities. The ministry of the interior would train police, the counternarcotics office was then supposed to execute the policies. Well, the ministry of the interior was run by former Mujahideen Tajiks while the ministry of counternarcotics was run by Hazaras who used to work for the Soviets. They didn’t like each other very much, they didn’t coordinate, and they didn’t talk to each other. Then, the two of them were supposed to get together and go down to Helmand and Kandahar and tell the Pashtuns how to get rid of drugs.

There was a complete lack of coordination there—either institutionalized or created on purpose by the drug lords. Even within the agencies, there was very little capacity to get anything done, so the counternarcotics ministry had an $80 million counternarcotics trust fund. Many countries who didn’t want to send troops to Afghanistan were contributing a lot of money to that fund, but Afghanistan had only spent $3 million of it after two years.

In Europe, I found a lot of coordination but no action. The Europeans are very good at coordinating—with the Policy Action Group, the Joint Coordination and Monitoring Board, etc.—but we had trouble getting them actually to act, to do something and to get their police training mission operational, to get it out to the provinces—to get it to work. So I reported back, “The Afghans are not very good at either coordination or doing anything, while the Europeans are pretty good at coordination but less so at acting.”

Then I had to work with the U.S. interagency, which was the most difficult process of all. There is action, but without coordination. In looking into, for example, prosecutor training, I found that we had three separate sets of curriculum being done by different U.S. agencies. They were inconsistent. This was the kind of lack of coordination that was going on among the agencies in Afghanistan. In another situation, where we were trying to build courthouses, we had one agency that was providing computerized equipment and another that decided they could delay providing electricity to those buildings during this time. So there was all this equipment but no electricity. One Afghan judge who had gotten to know the U.S. culture—like we were trying to get to know theirs—came up to me and asked, “How many Afghans does it take to screw in a light bulb?” I replied, “Well, I don’t know.” And he said, “Well, neither do we.”

There were more serious issues—for example, on the issue of whether to eradicate the opium crop in Helmand province. Interagency decisions had been taken to actually do this in certain of the wealthier parts of the country where doing so wouldn’t alienate poor farmers. But we got down there with the eradication force and there were fliers coming out of International Security Assistance Force saying “ISAF says not to destroy poppies.” So you have one group coming in to destroy the crop and another saying “we don’t destroy them.” It became a very serious problem.

Policy vs. Process

As the new administration gets started, we need to focus not only on reviewing policy, but also on reviewing and improving the process, which, as Janine Davidson observes, might be a little easier to reform than the substance.1 There have been three extensive bottom-up policy reviews on Afghanistan in 2008—the Afghan study group that General Jones led, the Atlantic Council Study Group, and OXFAM’s comprehensive policy study—but almost nobody is looking at the process. [Note: Since this talk, RAND Corporation has also issued an Afghan policy review paper.]

I agree with Davidson that major reorganizations are very difficult. The idea of completely rebuilding government and how it functions is not a realistic objective, especially when one is fighting two wars and doesn’t have time to wait for things to work. Davidson identifies the NSC as one specific area where improvement can be made. The NSC has the coordinating role among the interagency, but the way it was formed and staffed, it wasn’t really designed for coordinating major roles all at once. It has got a very smart group of people, but it’s under-resourced. It doesn’t have expertise in a lot of areas. At principals/deputies meetings, the action items get written up but there’s no way to monitor whether they’re really being executed unless—it’s all by back channels. There’s also no way of enforcement if one agency isn’t doing what it should—e.g., if DoD isn’t executing decisions made at the interagency that they don’t agree with.

The State Department Coordinator for Reconstruction and Stabilization (S/CRS), authorized in 2004, was supposed to be the State Department’s solution to interagency coordination, but it’s been slow getting off the ground. They appointed somebody of ambassadorial rank to coordinate all this, gave him practically no staff or money, and said “Now, everybody else in State and elsewhere, give them your staff and your money.” It’s very hard to get people to do that. The best way to get that organization operating may be to move it out of the State Department. The other agencies are not going to respect it if there is no respect within its own agency. Davidson mentions that maybe it could be moved to the White House and made part of the NSC, which might solve both problems in terms of getting NSC more expertise, more funding, more capability to monitor, improve its operations, and get more interagency buy-in.

As to funding, Davidson points out that there is a 50:1 ratio of funding for Defense Department to State Department. Therefore, in order to enable the State Department to do the things that it should be doing rather than having non-experts using Commander’s Emergency Response Program (CERP) money and the like, there’s going to have to be some parity in the funding and in the staffing. Secretary Clinton has said that she wants to do that, and it’s going to be essential to do so as State and other agencies—AID, DOJ—assume their roles in this stabilization effort in these countries. So that’s clearly got to happen. But there’s also a question of where the money comes from. A lot of the money that goes to the State Department—even when they do get money—comes from supplemental appropriations from DoD, which sends the money over to the State Department, along with its own very onerous conditions—which really continue to give DoD control over the operation. There has to be better recognition of how the funding ought to flow so that the experts can actually do the job.

The State Department has tremendous respect for the Department of Defense, but there was when I was there a perception that there wasn’t the same level of respect going to the other agencies. No organization has more competent leaders than the Department of Defense. But when you’re in the wars that we’re in now in Iraq and Afghanistan, where it’s not a traditional battlefield, you also need some followers, not just leaders. You need to recognize the civilian police training expertise of some of the people in the State Department. You need to recognize the rule of law expertise at the Department of Justice. There are other very good leaders in areas that are outside DoD’s expertise, and DoD also needs some good followers who are willing to enable the experts in the civilian area to do their jobs, because you can’t build a courthouse in Iraq or Afghanistan without some military support—not military direction or leadership, but military support.

There also is the issue of the length of tours. We frequently found that we were coordinating with multiple leaders on the military side within the space of a year. Sometimes the tours were only six months, sometimes two years, but within the tour they changed people around. For example, an idea was proposed two years ago by Defense to arm Afghan auxiliary police. It was not viewed favorably by the Department of Justice or the Department of State. We were fearful that the arms could fall in the wrong hands. There were a lot of problems with it. It was tried, it did not work, as we had predicted, and I have read in the media since I’ve left there’s a new plan to do basically the same thing by a whole new group of people at the Department of Defense. There should be a process where new military people that come in are gotten up to speed on what has and hasn’t worked in the past.

Finally, Davidson pointed out the inability to “develop” an effective counternarcotics strategy in Afghanistan. In fact, there was a process, and a strategy was developed: it’s on the website, it was approved by the deputies and the principals. It was execution that was the problem. The Department of Defense did not like certain parts of the counternarcotics strategy in Afghanistan. They made their views very clear in the principals and deputies meetings but, ultimately, consensus was reached that went against the views of the Department of Defense. DoD absolutely and completely resisted doing anything that was in that strategy, and in fact senior DoD officials told me that they simply would not execute it, despite it’s being approved by the White House.

I’ve written about the leaking of classified aspects of the strategy to the media in order to rile up our allies and encourage them to resist the strategy. We cannot have such a lack of discipline when there is an agreed-upon interagency strategy—and State and other agencies have done this too. That’s very problematic. I advocate the “one-strike-and-you’re-out” policy about leaks to the media: one strike and you’re fired, and we’ll investigate it.

That raises a related point, which is clear leadership of the interagency effort. The Obama administration has appointed a tremendous cast of heavyweights for Afghanistan and Pakistan, but it’s unclear who is running the show. George Mitchell was appointed special envoy for the Middle East, but a special representative for Afghanistan and Pakistan, Richard Holbrooke, was also appointed. I’m a special representative for the UN Office on Drugs and Crime in Mexico right now. Special representative has a very specific set of meanings and, in the context of the United States, from what I understand, they don’t think that means that Holbrooke reports to the Secretary of State, yet I doubt he’s going to be able to give direction to Secretary Gates. It’s unclear what General Jones’ role in all of this is, since he’s supposed to be the coordinator of this activity. And of course, Vice President Biden was the first person to go there. When it came out two weeks ago that perhaps President Obama would not support President Karzai, it actually was announced to the world through a leak. No one knew where the leak came from— which locus of power made this leak— there was no way to trace it down, the Karzai government was furious and, from what I understand, there was a lot of disarray in Washington as well.

The State Department’s approach reminds me of the old Judy Garland/Mickey Rooney “Andy Hardy” movies, where, whenever there’s a problem, the answer is “Let’s have a show!” The State Department seems to think whenever there is a serious policy issue, such as the disintegration of Afghanistan, “Let’s have an envoy.” But doing so undercuts the authority of the ambassadors in the country and makes it unclear what the role of the Secretary of State, the Secretary of Defense, and particularly the National Security Advisor are, and I don’t think it solves the problems. There needs to be a very quick and structured effort to streamline the interagency process to make it clear who is running the show, to prevent leaks, have clear lines of authority, and have recourse if somebody is not actually executing.

The final piece of that, of course, is if you stop leaking, the media becomes irritated and all your sources get mad at you. Another thing that undercut our activity, and both DoD and State were guilty of this, is officials’ leaking the good stuff and then sticking to bland, inane, and often inaccurate talking points when talking on the record. It’s totally counterproductive. I watched DoD go around on a series of press interviews (congressional committees are subject to this as well) and basically claim the Afghan National Army was doing great. And then there were leaks out, “Oh, we only have two of 97 units ready to go,” and then GAO comes out with a report that it’s not ready at all, and yet everybody gets stuck to the talking points. We did the same at State. We’d be given uninformative, boring, and sometimes highly-oversimplified talking points that just frustrated the media and Congress, and put pressure on everybody to leak the truth.

In addition to a strong no-leak policy, we need to empower intelligent, competent people who know the issues to go out there and tell the truth. “Here’s what’s happening. Here’s what’s going on.” You can’t leak classified information, but there’s so much information that people don’t know. For example, in my own area of the drug trade, there is an incredible amount of misinformation about the narcotics situation—about all the poor farmers who are growing poppy when, in fact, there are two UN reports that show that poppy farmers are the wealthiest people in Afghanistan. All the people who claim that eradication will drive people into the hands of the Taliban don’t seem to be aware that the only provinces that don’t have any poppy are the ones that did have an eradication program and there’s no evidence that anyone was driven into the hands of the Taliban. Just simple, objectively verifiable facts.


The interagency process clearly needs to be improved, but a focus on policy alone is not going to solve the problems. Improvement can be done in a narrow and streamlined way with some of the suggestions I have made. If we can devote attention to improving the process over the next few weeks, we’re going to get a much better result over the next four years.