war on the rocks

We Need a Map for Goldwater-Nichols Reform So We Don’t Get Lost

March 17, 2016

Where is Sen. John McCain going with Goldwater-Nichols 2.0? Over the last several months, the Senate Armed Services Committee received testimony from about 35 national security experts in an effort to identify problems and recommend reforms. The original Goldwater-Nichols legislation of 1986 made sweeping changes to the military command structure, including designating the COCOMs as the warfighting commanders, taking the service chiefs out of the operational chain of command, making the Chairman of the Joint Chiefs of Staff the chief military advisor to the President (but not in the chain of command), and greatly enhancing the attractiveness of joint duty. Now that the original legislation is 30 years old, McCain reportedly wants to do something big this year as a follow on. The entire national security community has responded. The Department of Defense has its own group working on the problem and will provide its proposals in about six weeks. Many think tanks (including, in the spirit of full disclosure, the Center for Strategic and International Studies (CSIS) where I work) are doing their own analysis and providing opinions. While it is true that many weaknesses in defense management have been identified, there is no clear consensus about what to do. An alternative approach, then, might be to take some quick wins now and put in place a process for identifying and building consensus for bigger changes later.

In order to help the process of identifying problems and potential solutions, CSIS analyzed the testimony given to the Senate Armed Services Committee and arranged the recommendations into nine categories. The CSIS team then looked for common themes in the recommendations where a consensus might be emerging. One of the themes was management weaknesses in the Defense Department’s organization and decision-making processes. There was little consensus, however, about how to fix these weaknesses. This was in contrast to the situation in the 1980s when the original Goldwater-Nichols legislation was developed. At that time there was a clear consensus that the services needed to work together more effectively and thus make  operations more joint.

Nevertheless, some issues keep coming up for discussion. It’s worthwhile, then, to take a look at these issues to see whether they are ready for decision.

The Combatant Commands (COCOMs) are bloated and should be refocused on warfighting. The original Goldwater-Nichols legislation greatly strengthened the COCOMs in order to enhance joint warfighting. As a result, the COCOMs have become much larger and more influential. The regional COCOMs have become political-military proconsuls and, instead of fighting wars themselves, create subordinate joint task forces to oversee conflicts. Some experts  suggest that one easy way to reduce COCOM staffs is to stop asking them to participate in Defense Department-wide processes such as budget development and acquisition. However, COCOMs are seen as the voice of the war fighter and have been brought into these processes for that reason.

Ironically, if there is any consensus it is to create new COCOMs by upgrading Cyber Command and Space Command to full COCOM status. Sometimes this is balanced by proposals to merge Northern Command with Southern Command, and European Command with African Command. However, many others oppose such consolidations because they would seem to signal to allies and partners that these regions are not important.

The pendulum has swung too far towards joint organizations and away from the services. This theme drove last year’s acquisition reform where the services regained some authority from OSD and the Joint Staff. The problem is real. Gen. Tommy Franks, when preparing for the invasion of Iraq in 2002, made it clear that he did not want to hear from the “Title 10 motherfuckers,” as he called the service chiefs. The service chiefs, with their expertise in organizing, training, and equipping the forces, could be brought back into operational planning. However, this would undo one of the major reforms of the original Goldwater-Nichols legislation. The services were explicitly taken out of operational planning because of perceived parochialism and logrolling.

Ironically, one recommendation that has had some traction is to put the Chairman of the Joint Chiefs of Staff into the chain of command, thus increasing his authority. The original Goldwater-Nichols legislation had explicitly kept him out because of concerns about loss of civilian control and risks in having a single military officer command all military forces.

The Office of the Secretary of Defense (OSD) and the Joint Staff have too much overlap. Over the years, OSD and the Joint Staff have both built organizations that cover the entire spectrum of activities, from budgeting to acquisition to strategy development to operational planning. Today OSD numbers about 3,300 and the Joint Staff about 2,700.  It would be possible to go back to an earlier era and get OSD out of military planning and operations, and get the Joint Staff out of budgeting and acquisition. The problem is that the Secretary of Defense and the Chairman of the Joint Chiefs have statutory responsibilities that require them to operate across the entire range of national security activities.

Pentagon overhead is bloated, inhibits rapid decision-making, and needs to be reduced. There are widespread concerns that management headquarters are too large and need to be reduced, but little consensus about where to take the cuts. The one delayering recommendation is to eliminate the service secretaries based on the theory that their functions, once vital, have now been superseded by OSD, the military staffs, and the defense agencies. Other experts worry that this would weaken civilian oversight of the military. Further, some military officers have argued that the civilian secretaries can handle partisan political issues that the senior military leadership is uncomfortable participating in. Last year, Congress in effect threw up its hands and instituted a $10 billion  reduction by FY 2019 across-the-board in management headquarters rather than target specific organizations.

Under Sen. McCain’s chairmanship, the Senate Armed Services seems to be wisely avoided two issues: using management reform to close budget gaps and legislating reforms to the interagency process. Management reform can make strategic decision-making more agile and effective, but large budget savings are unlikely to result. Indeed, the original Goldwater-Nichols legislation may have increased the size of headquarters as new organizations were layered onto the old. To save money in the infrastructure, reform efforts need to go where the money is: medical care, installation operations, depots, and training establishments. That’s an entirely different effort.

Many senior decision-makers, former Secretary Robert Gates prominent among them, have complained about the highly centralized decision-making of the Obama White House. Indeed, the National Security Council staff has grown to a record size of 400 people. However, the interagency process is at the fault line between Congress and the executive branch. Presidents get the interagency process that fits their decision-making style. It’s not something that Congress can legislate.

So if there is little consensus about big changes to make, what should be done? The answer is to establish some guiding principles, take some quick wins, and establish a process for phased reform over time.

The first principle is that the Department of Defense is not broken. This department is enormously capable, can accomplish the most complex tasks in government, and is highly trusted by the citizenry. That said, improvements are always possible. That leads to two overarching principles for reform: First, do no harm. Be sure that the cure is not worse than the disease. Second, clearly identify the problem that is to be solved in order to ensure that the solution in fact solves the problem. There are a lot of exciting proposals floating around in search of a problem to solve.

The next step would be to take a few quick wins this year. What follows is not a comprehensive list but a start.

  • Authorize another round of base closure (BRAC), increase workforce flexibility by allowing outsourcing (through OMB Circular A–76), and begin a process for easing some of the restrictive civilian personnel rules. These are hard because they contravene powerful parochial interests, but there is widespread agreement that Department of Defense needs these authorities to operate more efficiently. Congress can always override individual decisions that it doesn’t like.
  • Upgrade Cyber Command to a combatant command, but require that the total number of its personnel not change. These new domains would thus receive increased attention, hopefully without increasing total overhead.
  • Allow service chiefs to participate in some phases of operational planning. This would not put them into the chain of command but would provide a mechanism to give their advice and keep them informed.
  • Increase to four years the terms for the Chairman and Vice Chairman so they are comparable to the service chiefs. Although both are routinely given a second two-year term, this would eliminate any perception that they have less room to take risks.
  • Increase education in strategy formulation for both military and civilian personnel. This would improve the strategy formulation process without requiring controversial organizational changes.

Finally, Congress should put in place the mechanics for the next step. The first task is to identify the specific areas of interest. As noted above, COCOM function, COCOM consolidation, OSD/Joint Staff overlap, organizational delayering, and strategy formulation are candidates. Then Congress needs to establish a mechanism for assessing these questions. Although the Department of Defense needs to be consulted, it cannot objectively assess these questions on its own. It has too many established interests. Instead, Congress needs to charter some independent groups, whether a commission, a think tank, or a federally funded research and development center like the Rand Corporation, the Center for Naval Analyses, or the Institute for Defense Analysis.

Goldwater-Nichols is 30 years old.  Sen. McCain is right that it’s time for a review.   The challenge is to replicate its success and to avoid putting in place hasty changes that will need to be reversed in a few years.


Mark Cancian is a senior adviser at the Center for Strategic and International Studies.  Previously he worked many years as a senior civilian in the Pentagon and, before that, oversaw national security programs at Harvard’s School of Government.  He served 37 years in the Marine Corps, active and reserve, including two tours in Iraq.