July 26, 2006 | Lecture on National Security and Defense

Herding Cats: Understanding Why Government Agencies Don't Cooperate and How to Fix the Problem


(Delivered June 15, 2006)

Previous conferences in this annual series have focused on various aspects of armed conflict, includ­ing the military's role in stability operations and post-conflict activities. Those tasks are relatively simple compared to the subject of this year's conference: interagency operations, coordinating the joint effort of multiple government agencies, often with support from the ministries from different nations and non-governmental and international organizations, as well as private sector service providers-entities that often have different operational styles and practices, con­trasting organizational cultures, disparate resources, and conflicting goals and missions.

I have been asked to: (1) describe how the nature of U.S. interagency operations has changed since the end of the Cold War; (2) explain why the United States approaches interagency cooperation the way it does; and (3) suggest where there is room for improvement and what should be done to make the interagency process more effective.

Where We Are

I would like to begin with the conclusion. It is that Winston Churchill was right: Americans always do the right thing-after they have exhausted every other option. This, in fact, leads to my first point. The manner of U.S. interagency operations has changed little since the end of the Cold War. What we did well during the decades of stand-off with the Soviet Union, U.S. govern­ment agencies continue to do well. Where we muddled through-well, we continue to muddle through.

To understand the performance of the interagen­cy process, you really have to divide it into three lev­els. Let's call them policy, operations, and practice.

The highest level is policy. At the policy level, agencies in Washington reach broad agreement on what each will do to support an overall U.S. policy. Here the United States is not too bad. Interagency operations are really an invention of the Cold War. It is very difficult to look at the U.S. government in any period before in its history and point to an enduring, formal process for interagency cooperation that pro­duced anything significant. In the inter-war years between World Wars I and II, for example, the U.S. State Department refused to participate in war plan­ning or issue political guidance to the Army and Navy planners because they felt such coordination would be inappropriate and an intrusion of the mil­itary into the civilian sphere of government. That changed at the outset of the Cold War with the pas­sage of the National Security Act of 1947, which cre­ated the National Military Establishment, which would later become the Department of Defense (organizing all the services under a single federal department), the National Security Council (NSC), and the intelligence community. These entities, particularly the NSC, instituted a process of policy coordination that endures today.

At the lowest level is the practice of cooperation among individuals on the ground. Here, there are many examples in which American officials in short sleeves and soldiers in uniform work fairly well together, scratch their heads, figure things out, and get the job done to the best of their abilities uti­lizing the resources available. The embassy country teams under the leadership of the Chief of Mission are an example. All U.S. personnel stationed at an embassy, from consular officers to agricultural attachés, Immigration and Custom Enforcement agents, and military foreign affairs officers, work under unified direction.

Another good example of interagency operations in practice is the Joint Interagency Task Forces (JIATFs) that direct drug interdictions in the Caribbe­an and the Western coast of North America. They are a model of effective intelligence sharing and opera­tional coordination, not just for U.S. military and law enforcement agencies, but also for foreign govern­ments. It is not unusual for a French naval vessel to intercept drug runners headed for Europe based on information provided by the JIATF. The JIATFs are so effective that if they had twice as many planes and ships, they still would not be able to intercept all the suspicious shipments that they identify through intel­ligence gathers, fusion, and information sharing.[1]

On the ground, Americans aren't bad at inter­agency cooperation. And it is improving all the time. The military, for example, has conducted some kind of peacekeeping or post-conflict opera­tion every two years since the end of the Cold War. That means that junior officers and officials from all kinds of agencies have a lot of practical experience on the ground-working with nongovernmental organizations, figuring out the alphabet-soup of agencies they must coordinate with, and getting things done, despite-not because of-guidance from Washington or higher headquarters.

Therefore, we find it is not so bad at the policy lev­el and not too bad on the ground where individuals work together. It is at the intermediate level, the operational level, where the U.S. government under­takes major operations and campaigns, and where agencies in Washington have to develop operational plans such as coordinating recovery operations after a major hurricane. This is where interagency coop­eration is the weakest. This is a legacy of the Cold War. There was never a requirement for federal agencies to do that kind of integrated planning to contain the Soviet Union. Agencies generally agreed on the broad role each would play. There were few requirements under which they had to plan to work together in the field to accomplish a goal under uni­fied direction. Washington has never had an endur­ing formal system to do that.

Arguably, when efforts have been made to "oper­ationalize" decision-making in Washington, princi­pally by trying to coordinate ongoing interagency operations in the White House or at the NSC, they have proved unsatisfactory and Presidents have rightly backed off from the idea of trying to turn the Oval Office into an operations center.[2] No adminis­tration has hit on a satisfactory long-term solution.

Not much has changed. In fact, arguably it has got­ten slightly worse. After the Cold War, a system was developed under Presidential Decision Directive 56 (PDD-56), which established an interagency process to respond to complex contingencies overseas, such as providing assistance to foreign countries after earthquakes and hurricanes.[3] Agencies chafed under a formal process that required them to define an end­state, allocate resources, articulate a plan, and then jointly monitor execution. After a few years, PDD-56 was scrapped. That leaves us where we are.

Today, coordination of major interagency opera­tions in the field is often troubled. Reconstruction activities in Iraq are a case in point. The military, the Coalition Provisional Authority, and the United States Agency for International Development (USAID) all undertook major projects. There was no shared vision, no common operational planning, and no integrated contracting or human capital management process. As a result, these organiza­tions learned lessons on the job and adapted, but they did not keep up with the changing security environment in the country, and after spending bil­lions of dollars, there was very little to show for the investment.

If there is a problem that needs to be fixed, it is this-the ability to coordinate major interagency challenges outside of Washington, away from the offices of Cabinet secretaries and staffs, whether it is coordinating disaster relief over a three-state area after a hurricane or conducting the occupation of a foreign country.

Why We Are Flawed

It should come as no surprise that operational interagency activities have been found wanting. They are flawed by design. I can offer at least eight factors that contribute to that.

Tradition. The divide between civil and military spheres is part of a U.S. tradition that has always placed a premium on civilian control of the mili­tary. In the 19th century it was thought appropriate to "firewall" military activities from civilian func­tions.[4] Even today, military and civilian officials are cautious about "straying out of their lane."

Congress. Congress is ill-suited to promote cooperation between federal agencies. It appropri­ates funds for operations of individual depart­ments. The jurisdiction of committees that oversee the government dovetail with the departments they oversee.

Professional Development. One key instrument for facilitating integrated action is a shared body of common knowledge and practices, common expe­riences, and trust and confidence among practitio­ners. The military achieved improved cooperation between the armed forces by creating a joint profes­sional development program that included activi­ties involving more than one military service and that included requirements for joint education, joint assignments, and joint accreditation. No such professional development program exists for the interagency process.

Operational Organization. Every federal agency has its own distinct operational organization. The U.S. military, for example, has a system of regional commands established under the Unified Com­mand Plan (UCP). It does match the State Depart­ment's regional system, which, in any case, functions nothing like the military combatant com­mands. Federal agencies are always reluctant to support interagency headquarters outside of Wash­ington out of fear that they will usurp policymak­ing authorities from the department secretariats.

Capacity. Outside the Department of Defense, federal departments have very limited capabilities to conduct "operational" activities. Most federal agencies, for example, do not have effective means to mobilize and deploy personnel.

Inspectors General. Interagency operations require effective oversight. This is problematic for a federal inspector general corps that aligns with individual agencies. In Iraq, for example, a Special Inspector General for Iraq had to be established to oversee activities involving multiple agencies.

Politics.Many politicians are rightly uncomfort­able with the notion of "big" government. They are concerned that creating a more effective interagency process would empower government to the point that it might lead to abuse, encouraging Washington to take on missions that are not appropriate.

Operational Models. There are no good opera­tional models on how to undertake major inter­agency activities outside of Washington. The most common is the "lead agency" model, in which one federal agency is responsible for leading a response or planning effort. Where the lead agency has the preponderance of responsibility and the resources, usually other departments act like bystanders- primarily interested in doing as little as possible. Where the departments all have major equities in the process, usually everyone simply agrees to do what they are already doing.

We Could Do It Better

The key to improving interagency operations is to focus on the most pressing problem-and that is not in a Cabinet secretary's office. The answer is not reorganizing the federal government or redistribut­ing federal responsibilities. We need to focus on how to make the interagency process more respon­sive in the operational environment. I have four recommendations.

U.S.-Plan.The first recommendation is that we have to create a place where this collaborative inter­agency process can take.[5] The UCP is still primarily organized to provide global command for the last war. In addition, while each of the geographic com­mands contains a joint interagency coordination group to organize regional activities, in practice, there is little cooperation or planning with outside organizations or departments. Furthermore, com­batant commanders tend to compete with the ambassador (and the ambassador's country team, which incorporates all civilian, military, and intelli­gence personnel assigned to the embassy) in each country in the commander's area of responsibility. Combatant commanders cannot partner with the State Department at the regional level either, because the State Department's regional desks cov­er different geographical areas than the UCP's areas of responsibility..

It is time to replace the UCP with an organiza­tional structure that better supports national secu­rity needs. That organization should emphasize facilitating interagency operations around the world, while still facilitating effective joint combat action. A new structure, the U.S. Engagement Plan (U.S.-Plan), should be crafted at the direction of, and in response to, the National Security Council, rather than the Pentagon.

A possible structure for U.S.-Plan might go as fol­lows. There is still a need for permanent military commands under the direction of the Pentagon; however, the number of combatant commands should be reduced to three. In Europe and Northeast Asia, the United States has important and enduring military alliances and there is a continuing need to integrate the U.S. military commands with them. To this end, EUCOM and PACOM should be replaced by a U.S.-NATO command and a U.S. Northeast Asia headquarters. NORTHCOM should remain as the military command responsible for the defense of the United States. In addition, three "Joint Interagen­cy Groups" (InterGroups) should be established. Joint-Interagency Task Forces have already been used very effectively on a small scale to conduct counter-narcotics operations in Latin America, the Caribbean, and off the Pacific coast of the United States. They incorporate resources from multiple agencies under a single command structure for spe­cific missions. There is no reason that this model could not be expanded in the form of InterGroups to cover larger geographical areas and more diverse mission sets.

The InterGroups within U.S.-Plan should be established to link areas of concern related to nation­al security missions, such as transnational terrorism, transnational crime (e.g., piracy and drug and human trafficking), weapons proliferation, and regional instability. The InterGroups should be established for Latin America, Africa, and the Middle East, and South and Central Asia. Each InterGroup would have a mission set specific to its area. The Lat­in America InterGroup should focus on drug, human, and arms trafficking, counter-terrorism, civ­il-military relations, and trade liberalization. The Africa-Middle East InterGroup should focus on counter-terrorism, weapons proliferation, economic development, fighting AIDS and other infectious diseases, peacekeeping training and support, trans-national crime, and civil-military relations. Central and South Asia InterGroups should concentrate on counter-terrorism, weapons proliferation, training police forces, anti-piracy measures, civil-military relations, trans-national crime, and fighting AIDS and other infectious diseases.

Each InterGroup should include a military staff tasked with planning military engagements, war-fighting, and post-conflict operations. In the event that military operations are required, the military staff could be detached from the InterGroup (along with any supporting staff from other agencies required) to become the nucleus of a standing Joint Task Force (JTF). Using this model, operations in Iraq and Afghanistan would have been command­ed by a JTF.

National/Homeland Security University. Hav­ing established a place for operational action, there will be need for a professional development system to provide personnel qualified to work there. That will require a program of education, assignment, and accreditation that cuts across all federal agen­cies with national security responsibilities. That has to start with a professional school that teaches interagency skills. No suitable place currently exists in Washington, academia, or elsewhere. The government will have establish it.

Operational Concept.One product this univer­sity will have to produce is a suitable concept to frame interagency operations. As a first priority, the concept will have to articulate how unity of com­mand will be established. I would argue that the nature of the task should define who should be in charge. When dealing overseas, there are three crit­ical tasks. They have been described in various ways as, "justice, security, and well-being," or "gov­ernance, security, and essential services." Planning the occupations after World War II, the military planners called it the "disease and unrest" formu­la-preventing humanitarian crises, establishing a legitimate, functioning government, and ensuring the existence of competent domestic security forces to support that government.[6]

Who should be in charge should depend on which of the three missions has priority at the time. In a post-conflict environment, for example, the military should be charge of interagency operations until a stable security environment is in place. Where crisis response is the priority (and security is not a major issue), a civilian agency should take the lead. Ideally, that agency would be an overseas arm of the Federal Emergency Management Agen­cy. Where governance is the issue, building up the capacity of government to be honest and efficient and to promote economic growth and strong civil society (again, when security is adequate), a civil­ian agency should be charge. I would prefer this be something more like USAID, but independent from the State Department, using instruments more like the Millennium Challenge Account and focusing on measures such as those listed in the Heritage Foundation and Wall Street Journal Index of Eco­nomic Freedom.

Funding. Developing the capacity for all federal agencies and nongovernmental agencies-and pri­vate sector contractors, for that matter-to provide the people and services needed has to be a priority as well. There is a simple solution for cutting the Gordian knot of the thoroughly knotty problem of who pays. Congress could appropriate money to the federal agency that will provide leadership for the operation-and that agency would negotiate with other federal agencies, nongovernmental organizations, and private sector contractors to determine what it needs to support what needs to be done. For planning, training, education, and exercises, the lead agency would pay other agen­cies to participate out of an annual appropriation provided by Congress. For operations, it would pay for the supporting agencies to provide person­nel and services (and the salaries of personnel to backfill personnel that are deployed for opera­tions) out of supplemental appropriations provid­ed by the Congress.

The Way Forward

I have suggested in my presentation that the real shortfall in the interagency process is the lack of adequate capacity to conduct operations outside of Washington, where the challenge is not to formu­late national policy, but to plan and execute opera­tions in a way that lets the people on the ground work well together and get the job done. There are impediments that prevent us from doing that now. They can be overcome.

In fact, they can be overcome with what are argu­ably rather modest innovations: creating a regional framework for interagency planning and action; a means to create a corps of interagency professionals; a doctrine that establishes a rationale for creating unity of effort and ensuring that a single entity has the authority and resources to accomplish the mis­sion; and a means to fund the process so that there is reasonable assurance that the essential personnel and services will be available when they are needed. I think this a reasonable and achievable agenda for the Congress and the Bush Administration.

James Jay Carafano, Ph.D., is Senior Research Fellow for National Security and Homeland Security in the Douglas and Sarah Allison Center for Foreign Pol­icy Studies, a division of the Kathryn and Shelby Cullom Davis Institute for International Studies, at The Heritage Foundation. This paper was presented to the Conference on Interagency Operations: Cultural Con­flicts Past and Present, Future Perspectives, hosted by the Center of History, Political Studies, Institute of Par­is, June 15-16, 2006.






[1]James Jay Carafano, "A Better Way to Fight Terrorism," Heritage Foundation Commentary, May 17, 2005, at http://www.heritage.org/Press/Commentary/ed05705b.cfm.

[2]Carnes Lord, "Crisis (Mis-) Management," Joint Force Quarterly (Summer 1999), pp. 72-79.

[3]See William P. Hamblet and Jerry G. Kline, "Interagency Cooperation: PDD 56 and Complex Contingency Operations," Joint Force Quarterly (Spring 2000), pp. 92-97.

[4]James Stever, "The Glass Firewall Between Military and Civil Administration," Administration and Society, Vol. 31, No. 1 (March 1999), pp. 28-49.

[5]James Jay Carafano, Ph.D., "Missions, Responsibilities, and Geography: Rethinking How the Pentagon Commands the World," Heritage Foundation Backgrounder No. 1792, August 26, 2004, at http://www.heritage.org/Research/NationalSecurity/bg1792.cfm.

[6]James Jay Carafano and Dana R. Dillon, "Winning the Peace: Principles for Post-Conflict Operations," Heritage Foundation Backgrounder No. 1859, June 13, 2005, p. 5, at http://www.heritage.org/Research/NationalSecurity/bg1859.cfm .

About the Author

James Jay Carafano, Ph.D. Vice President for the Kathryn and Shelby Cullom Davis Institute for National Security and Foreign Policy, and the E. W. Richardson Fellow