war on the rocks

Right Ends, Wrong Means: What Congress is Missing on Defense Reform

August 11, 2016

Congress’ current efforts to change the way the Department of Defense — and the U.S. national security apparatus more generally — operates are well-meaning, but replete with pitfalls and hurdles. These changes risk serving as only band-aids on larger problems. The House and Senate versions of the fiscal year 2017 (FY 2017) National Defense Authorization Act (NDAA) propose sweeping changes to the organization of the Department of Defense — the most dramatic since the Goldwater-Nichols Act, passed 30 years ago. Proposals include reductions to organization size and changes to the way elements of the Department of Defense operate, particularly at the combatant command level. Although each of these proposals is well-motivated and driven by real weaknesses of the existing organization, as a whole, they are unlikely to accomplish their goals. We have extensive experience conducting organizational analysis throughout the Department of Defense, and it is our view that the specific proposals in the 2017 NDAA will not solve the underlying problems. While they have merit in addressing the challenges the Pentagon faces, there are significant implementation hurdles that must be acknowledged and addressed up front, before moving too quickly toward an overly constrained solution.


The first set of changes we discuss are those related to the department’s size. Both the House and Senate versions outline reductions to general and flag officers and/or senior executive service staff (civilian equivalents of generals). Similarly, the Senate’s version of the bill defines cuts to the National Security Council’s staff, and the House bill places an additional confirmation requirement on the national security advisor if the National Security Council exceeds a certain size. (Though not within the Department of Defense, the National Security Council is still part of the nation’s defense apparatus more broadly.) As referenced in the bill reports (both Senate and House) and argued in the testimony of numerous experts, the rapid and disproportionate expansion of senior ranks and headquarters staffs renders them ripe for reductions. Moreover, the requirement to make reductions is a good forcing function for the Department of Defense to examine its staff structure: The desire or need to become more efficient is a leading motivator we have seen in requests for our organizational analysis. Unfortunately, by explicitly defining the magnitude of the reductions to be made, Congress hamstrings the potential of this forcing function. Instead, it will lead to cost-avoidance tactics, and it is unlikely to fix the underlying process inefficiencies that produced the bloat in the first place.

Staff cuts hurt, and as a result, we’ve seen in previous reductions that organizations seek to mitigate the pain while still theoretically adhering to the letter of reduction targets. They may do this by trying to limit who “counts” in the reduction, move staff to excepted/exempted places, or threaten to make cuts in places that are politically untenable or operationally dangerous (and thus obtain relief from cuts). In a 1978 study, the Government Accountability Office found that targeted 20 to 25 percent reductions to headquarters staffs were largely accomplished by transferring billets to lower echelons. In other words, very few positions were actually removed from the payroll. Similarly, in accommodating the FY2016 NDAA-directed military headquarters activity staff reductions, we’ve seen several Pentagon organizations attempt to negotiate who counts as “headquarters.”

Even when true staff cuts are enforced, they are more likely to lead to overworked personnel or neglected responsibilities than to more efficient processes. Staffs are more likely to do the same work, the same way, with fewer people. Organizations tend to gain more responsibilities over time, but are rarely successful at shedding them. CNA found this in an examination of the history of the Office of the Chief of Naval Operations. Over the past 40 years, only two responsibilities were shed, both with the creation of new organizations: security assistance responsibilities were transferred to the Secretary of the Navy’s Navy International Programs Office in the 1980s, and doctrine became the purview of the new Navy Doctrine Command in the 1990s. The Kano model for customer-driven organizations, which we have also applied in studies examining U.S. government and military service providers, similarly suggests that over time, nice-to-have functions become required functions and are subsequently replaced by new nice-to-have functions. Even when organizations must dramatically downsize (such as the former Joint Improvised Explosive Device Defeat Organization) or are disestablished (such as Joint Forces Command), we did not see their responsibilities correspondingly diminish or disappear. Thus, all of those positions that have been added over the years are likely supporting real work. We rarely find in our analyses that anyone on a staff is not gainfully employed.

Of course, it is possible to do the same work smarter: to find efficiencies that would allow DoD organizations to meet the same responsibilities without sacrificing quality or morale. However, those efficiencies are the result of organizational and process change or improvement. Identifying inefficient processes and relationships is a resource- and time-consuming proposition — one challenging to accomplish in an existing high operational tempo and limited resource environment. Moreover, although such an examination should yield efficiency dividends and corresponding staff reduction opportunities, there is no guarantee that the reduction opportunities identified will meet or exceed the targets Congress aims to impose. With no guarantee that all that work will get you to the target, there is little incentive to pursue process change. Rather, we’ve seen in the 2016 NDAA reduction requirements that organizations have resorted to two “easy” tactics: “salami-slicing” staffs by requiring equal shares of cuts across the organization and eliminating billets that are currently gapped (the argument being that they are already making due without these positions). Unfortunately, both are tactical changes with limited ability to change the system that has become so inefficient in the first place. To effectively shrink the department, staff reductions should be an outcome of structural changes that improve efficiency and effectiveness, but not the goal in and of themselves.

Internal functioning

The FY 2017 NDAA proposals also seek to improve efficiency and effectiveness by bridging stovepipes within the Pentagon and across the services. The Senate version of the NDAA sets a requirement to establish six new cross-functional teams within the Office of the Secretary of Defense, reporting directly to the secretary and deputy secretary. Similarly, it requires a pilot program to replace the subordinate commands of a combatant command with joint task forces, rather than service components. Certainly, anyone who has spent time in defense institutions has examples of (often unconstructive) parochialism, and cross-functional teams may be one good way to overcome this. Examples in industry and government tout the advantages of this organizational construct. Many missions are intrinsically cross-functional in scope, and the construct formalizes a means to share information across stovepipes.

Our research has documented clear instances of the value of cross-functional teams within the Department of Defense. The premier example in the Navy involves maritime operations centers at fleet commands. These organizations pull together staff members across the traditional Napoleonic centers of expertise — intelligence, operations, logistics, etc. — and employ them in various working groups, oriented around functional processes or timelines. Other examples involve the stand-up of joint task forces established with specialized or limited objectives. Such organizational constructs are well-suited to dynamic and complex environments, as well as to those that demand sophisticated innovation. Because the military is likely to operate in these types of environments, cross-functional teams may of particular value to it generally. However, these environments are arguably less characteristic of higher-echelon combatant commands and Pentagon-based organizations, where Congress aims to install such teams.

Even if cross-functional teams are a reasonable approach at these higher echelons to overcoming stovepipes, there are costs and challenges not well considered by Congress’ proposals. For instance, “special” groups that are external to the basic staff structure of an organization can neuter the influence and effectiveness of the original organization. As an example, CNA observed in several related (as yet unreleased) studies that the creation of special, direct-report Chief of Naval Operations strategy groups limited the influence of his office’s operations and strategy department — the traditional locus for strategic concept development and planning, where strategy is integrated across the Navy. Effectively, the special groups limited the access of the regular staff to senior decision-makers and poached top talent from their ranks. Congressionally directed cross-functional teams could potentially have a similar effect.

Both the cross-functional teams and combatant command joint task forces are also likely to suffer from the dangers of matrix or hybrid organizations, where individuals report to superiors along two axes, or at least have allegiances to two organizations (the team and the service). Both matrix and hybrid organizations can merge complementary expertise in constructive ways, but their management is challenging. Military staff members worry about who writes their fitness reports and who cares about their career growth, and they come to the table with strong career and experiential schemas — all of which challenges their ability to work “across the aisle,” even in the face of a collaborative culture and personalities. This, again, is a trend we have seen in our previous analyses.

When it comes to cross-functional teams, one size does not fit all. Each circumstance has its own nuances and pathologies. Thus, any guidance or doctrine about how to set up the organization should leave much room for interpretation. Although cross-functional teams may be a way to improve the effectiveness of the department, they are not always appropriate and are particularly challenging to manage. Congress’ a priori determination that cross-functional teams are the solution may undermine true effectiveness improvements and should not be explicitly prescribed.

External functioning

Finally, the 2017 NDAA proposals seek to bridge existing combatant command seams to improve organization-to-organization functioning within DoD. These seams were created by Goldwater-Nichols and the establishment of regionally focused combatant commanders. Combatant command areas of responsibility have shifted over the years, but seams have always existed where the edges of these areas meet. The Senate version of the bill proposes a combatant command council, and both versions propose new authorities for the Chairman of the Joint Chiefs of Staff to manage forces more holistically across combatant commands. Both proposals aim to transcend geographically bound management of the military, which the House report, for instance, argues is challenged by threats that “increasingly span multiple COCOM areas and domains of responsibility and require their seamless integration.”

Once again, the motivation behind these Congressional proposals is sound. All organizations have seams — it’s necessary to parse responsibilities in some way to provide effective oversight and allow for specialization. These seams will always introduce some risk where the handoff of information or activities must occur between different pieces of the organization. Information can be misconstrued, balls can be dropped, or context can be lost. A common method suggested in management literature to improve information flow and collaboration across organizational seams is the establishment of standing committees— groups comprising members from each organizational element that meet regularly to collaborate on organization-wide objectives. The combatant command council proposed by the Senate NDAA represents such a standing committee. Thus, there is great potential in this proposal to help bridge existing combatant command seams, but the proof would be in the execution. We have seen time and again in our organizational analyses that truly effective integration and coordination requires not just appropriate mechanisms, but also individual personalities and relationships that are open and collaborative.

Perhaps more importantly, while NDAA proposals aim to bridge existing seams, they do not ask the larger question about whether combatant command seams are the right seams. As we mentioned, all organizations have seams, and all seams yield risks. But different organizational structures will yield different risks, and these different risks may be either more acceptable or more dangerous than current ones. The bill proposals do not question whether the existing combatant command structure is the right one, despite this being a common topic of discussion in Washington circles in recent years. Although we do not claim to know what the least risky organizational option for combatant commands is, the 2017 NDAA proposals offer a band-aid to the risks induced by Goldwater-Nichols-era legislation without instigating a deeper exploration of the seams and risks themselves.


The geopolitical and security environment in which the Department of Defense now operates is dramatically different from the one Goldwater-Nichols legislation was intended to address. Moreover, the decades since the last major Pentagon reorganization have allowed inefficient processes, entrenched positions, and bloat to creep in. These are all things that Congress aims to address with proposals in the 2017 NDAA. As we have shown, however, the specific recommendations Congress makes to address these issues face implementation hurdles, may have unintended consequences, or do not address the true underlying reasons for existing challenges. Consequently, rather than dictating specific solutions via legislation, we recommend Congress use the bill to motivate a deeper, Pentagon-driven introspection to yield truly effective reforms.


Margaux Hoar is the Research Team Leader for the Organizations, Roles, and Missions Team at CNA, in Arlington, VA. Her team examines the responsibilities, dynamics, seams, and operational effectiveness of the U.S. military at various interface levels: within organizations, as elements of command and control structures, at components of U.S. national defense, and as elements of the larger national security apparatus.

Dave Zvijac is a Senior Research Scientist on the Organizations, Roles, and Missions Team at CNA. He has spent a significant portion of his operations research career in the field, mostly at senior Pacific commands. Currently, his major research interests focus on inter-agency coordination, the operational level of war, and military organizational structure.

Image: DoD photo by Army Staff Sgt. Sean K. Harp