The original concept was sensible: Conduct a major study at the beginning of every administration to establish a new strategy and, from that, set the force structure, programs and budgets to support that strategy. Such a study would bring coherence to policy and budgets that had often fallen victim to delayed and ad hoc studies of specific topics.
It worked the first time. The 1993 Bottom-Up Review, though not perfect, set a post-Cold War strategy and reshaped force levels, equipment purchases and budgets to suit. Its success begat the Quadrennial Defense Reviews, which have been less and less successful. Constrained by congressional direction and the inherent difficulty of the task, QDRs have devolved into vague descriptions of grand strategy and unprioritized lists of worthy goals, all disconnected from any discussion of resources. Over time, the reviews have taken longer, cost more and produced less. Critics of the process have not succeeded in changing it.
There is an alternative to expending immense staff effort and senior official time on a big, formal QDR: Focus a streamlined review on the key strategic, programmatic and budgetary changes that must be decided immediately, and conduct broad reviews later if major changes in the strategic environment require them.
March To Irrelevance
Bottom-Up Review. Soon after the Cold War, need and capability came together to produce a new kind of review. In 1993, the Clinton administration took office at a time when everyone recognized the need for a strategic reorientation. The Soviet Union had disintegrated; its military machine continued to decay. The Bush administration had begun this reorientation with its “Base Force,” a 20 percent cut in forces, but the new administration called for further cuts.
The new secretary of defense, Les Aspin, was uniquely qualified to carry out such a reorientation. Long the chairman of the House Armed Services Committee, Aspin had a deep intellectual interest in national security issues, had already conducted several studies on post-Cold War strategy, and had built a capable committee staff, which he brought over to the Pentagon.
The result, published in October 1993, was the Bottom-Up Review. Aspin drove it using existing work from the Bush administration and new work developed by the permanent bureaucracy, his personal staff and the half-dozen political appointees who were aboard at that time. (The fact that a major strategic review can be successfully conducted with half a dozen political appointees ought, in itself, to raise questions.) The Bottom-Up Review laid out a post-Cold War strategy, described options in force structure and modernization for meeting that strategy, and then explained the administration’s decisions:
The short strategy section described a shift from global conflict to regional conflicts.
Based on this strategy, extensive force structure analysis, which was described in the report, called for 12 carrier battle groups, 10 Army divisions, 20 Air Force tactical fighter wings and a 174,000-person Marine Corps.
Following from the new strategy and force structure, some acquisition programs were increased, others curtailed, still others canceled.
The report concluded with a resource analysis showing how the new programs would fit within the lower level of future budgets.
The Bottom-Up Review was so successful in linking strategy, programs and budgets that the Congress codified it into law in 1996 (10 USC 118), requiring that every new administration conduct a similar review. This new Quadrennial Defense Review would constitute “a comprehensive examination of the national defense strategy, force structure, force modernization plans, infrastructure, budget plan, and other elements of the defense program and policies of the United States with a view toward determining and expressing the defense strategy of the United States and establishing a defense program for the next 20 years.” Under this statute, there have been four QDRs.
QDR 1997. The first took place at the beginning of the Clinton administration’s second term. The existence of a strategy, force structure and acquisition program from the first term constrained the review, and the strategic environment had not changed much. Nevertheless, the review made some changes. The review’s short strategy section codified the existing strategy as “shape, respond, prepare,” using the kinds of action verbs that later QDRs found so attractive. The key concern was making fiscal space for increased modernization, which had languished in the 1990s (the “procurement holiday”). Therefore, the review recommended cutting forces and manpower, mainly in the support areas, and hitting reserves hardest. (This latter move later instigated a revolt of the National Guard and reserve.) Base closings and outsourcing would reduce infrastructure costs. The report described three overarching options (“Near-Term Focus,” “More Distant Threat,” “Balanced Approach”), provided a cost and effectiveness assessment of each and then explained the rationale for choosing the “Balanced Approach.” A resource section showed how the cost-cutting recommendations would provide funds for increased modernization and “transformational technologies.”
The review would seem to have been successful: It linked strategy, programs and resources. But congressional critics complained that the review had been “budget-driven, not strategy-driven.” As a result, the QDR legislation was amended so that future QDRs would not be constrained by budgets (10 USC 118 b4).
QDR 2001. The first QDR of President George W. Bush’s administration was also the first of the “strategy” QDRs. It reflected the new congressional direction that QDRs not be budget-constrained, but also a bureaucratic change that shifted staff leadership for the review away from a resource organization, Program Analysis and Evaluation (now Cost Analysis and Program Evaluation). After two false starts, one with the Office of Net Assessment and the other with outside expert groups, neither of which produced actionable recommendations, the QDR was assigned to the Defense Department’s policy cluster. Despite this rocky beginning, and press reports of vigorous behind-the-scenes debates about force cuts, the review came out relatively quickly, in September 2001.
The new review contained a new set of strategic action verbs — “assure, dissuade, deter, defeat” — and broadened its requirements beyond the two regional conflicts of previous QDRs. To support these broader requirements, it prescribed a move to “capabilities-based” force planning. But instead of demonstrating connections between strategy, programs and resources, the review only asserted them. Although force-structure decisions reportedly reflected “assessments across several combinations of scenarios,” none of this was described. The report had only one chart and no dollar signs. Ultimately, the review made no more than small changes to the force structure.
QDR 2006. Bush’s next QDR continued the “strategy” approach, offering few numbers, no tables and no resource section. The post-9/11 strategy, not surprisingly, focused on terrorism (“the long war”), defense of the homeland and “transformation.” It also described in words, but not numbers, a new force planning concept that built on homeland defense, conventional conflicts and long-term irregular warfare, in both steady state and surge conditions. Program changes were few, as might be expected in a second administration. Despite the modest nature of the changes, the review took the longest to conduct, with the final report coming out a year after the inauguration. This was consistent with new congressional language specifying that the QDR be published with the budget. The 2006 version was also the shortest-lived of the QDRs. It concluded: “The size of today’s forces is appropriate,” and instead put effort into “transformational” technologies. Ten months later, the president announced major increases in Army and Marine Corps end strength, thus reversing the QDR’s major precept.
QDR 2010. Published in February, a year after the Obama administration took office, this review was anticlimactic. Defense Secretary Robert Gates had essentially published the new strategy in a Foreign Affairs article the year before and announced major programmatic decisions in April: expanding some capabilities, such as surveillance systems and global partnerships, and canceling major acquisition systems, such as the Army’s Future Combat System and the Air Force’s F-22 fighter. Another “strategy” review, QDR 2010 lasted just one month longer than its short-lived predecessor. It was displaced in January 2011 by the Comprehensive Review, which changed the strategy and cut programs to accommodate budget reductions from the Budget Control Act.
If QDRs were produced by a small group operating independently, then none of this would matter. A QDR would be another government report that might, or might not, affect the real world. However, every QDR takes immense effort by both staffs and policy officials.
The statutory list of requirements has 17 items, from threats to readiness to force structure. It also includes esoteric topics such as defense missions of the Coast Guard, revisions to the Unified Command Plan and effects of climate change on national security. To accomplish these many tasks, QDR 1997 required three levels of review and seven panels, and this was only the tip of a bureaucratic iceberg that included several dozen issue teams. All QDRs have had similar structures.
Feeding this bureaucratic beast takes a lot of personnel. Further, with such important issues in play, every major organization — service staff, service command, agency, combatant command — believes that it must participate. As a result, staffs at every level get beefed up in order to establish “QDR cells” that participate in the many issue teams and respond quickly to proposals. Top-performing individuals, who might otherwise be in the field leading troops and conducting operations, get sent to the Washington bureaucracy. After all, wars in Iraq and Afghanistan may come and go, but the inter-service struggle for budget share goes on forever. One participant likened the dynamics to gang warfare, each gang having its distinctive colors, culture and paranoias.
Policy officials must participate in endless meetings, diverting attention from other matters. In one of the innumerable QDR pre-meetings, a general officer exclaimed to this author, “I’ve been to three military funerals this week, and you’re wasting my time with BS!”
Increasing the burden, the QDR has spawned offshoot reviews. Two reviews are in the QDR statute — the Chairman’s Assessment and the National Defense Panel — both designed as independent checks on the QDR. The Joint Staff, already deeply involved in the QDR, develops the Chairman’s Assessment. Recent assessments have been modest, though independent-minded chairmen can use it to raise areas of special concern, as Army Gen. John Shalikashvili did with modernization in QDR 1997 and Army Gen. Hugh Shelton did with force structure cuts in QDR 2001. The panel is entirely separate, with its own processes, meetings and staff.
Two further reviews have run concurrently with the QDR — the Nuclear Posture Review, which sets policy and structure for nuclear forces, and the Ballistic Missile Defense Review, which does the same for missile defense. Both entail their own set of teams and meetings.
Every QDR spawns a cottage industry of consultants, advisers, critics and kibitzers in the national security community. Much of this criticism aims at the particulars of the policies espoused. While important, they do not involve us here. What does concern us are systemic criticisms that cross all the QDRs. Of these, two stand out: the disconnect between strategy and resources, and the lack of explicit force-sizing constructs.
The disconnect between strategy and resources has been noted. In part, this disconnect is driven by the task’s inherent difficulty — it requires being specific about what a strategy means, and does not mean, and that involves difficult tradeoffs. In part, it is driven by partisan politics, as the out-of-power party seeks to show that the current administration is underfunding its strategy. Thus the QDR statute calls for an unconstrained budget plan. Further, the QDR statute calls for a strategy at “low-to-moderate level of risk.” In the Pentagon, “low-to-moderate” risk requires very robust — and expensive — capabilities. “Moderate” risk is the standard. The QDR statute thus specifies an unachievable goal. Rather than deal with these tensions, QDRs instead ignore resources.
Lack of an explicit force-sizing construct means that there is no quantitative explanation for why the services have the forces they do. For example, QDR 2006 recommended 70 total Army combat brigades. Why 70? Why is 65 too few and 75 too many? The Bottom-Up Review explicitly tied Army forces to two major conflicts — each would require four to five divisions, so the service would need a total of 10. Since then, however, QDRs have mentioned scenarios and force demands that collectively drive service force structures, but leave the connections and actual numbers unstated. Force-sizing constructs do exist deep inside the Pentagon, but the creators of QDRs have apparently not had enough confidence to publish them.
It’s all Paul Nitze’s fault. The memory of Nitze is now fading (he died in 2004), but his signature achievement is still commemorated in the international relations history books. That achievement was NSC-68, a document published in 1950, recommending firm containment of Soviet power. Catching a key moment in postwar history — the Korean War had just begun and the Soviets had detonated a nuclear weapon — NSC-68 set the national security strategy of the United States for two generations. Since then, every policy wonk who ever got the QDR pen has wanted to replicate Nitze’s achievement. As a result, the strategy sections of QDR reports have been elaborate and robust, seeking to create a strategic framework lasting a generation, while other sections atrophied. Further, as Max Boot notes, QDRs have become susceptible to the “intellectual fad of the day,” from “transformation” to “counterinsurgency.”
Getting the strategy right is not enough. Strategists often believe that if institutions get the strategy right, then programs, policies and budgets will follow. Over time, as issues arise, the strategy will shape individual decisions so that the institution will realign itself with the new strategy. Thus, a QDR that focuses on strategy will, eventually, produce the policy and program tradeoffs that the review itself was unable to make.
In fact, this rarely happens. If strategy is not tied to programs, what does change is the justification for the existing programs and policies, as these justifications come into line with the words of the new strategy. The programs and policies themselves remain the same.
For example, no matter what strategy comes out of a QDR, the Navy will argue that carrier battle groups fit the strategy perfectly. And the Navy will not necessarily be wrong. Carrier battle groups are versatile instruments of national power and can accomplish many national security tasks. All strategies are open to interpretation, and the services will build their requirements on their interpretation of strategy. Nor is the Navy alone in this. The Army will argue for “boots on the ground,” the Air Force for precision strike, the Marine Corps for the versatility of its air-ground task forces. Strategies with many unprioritized tasks and goals, like those in recent QDRs, are particularly susceptible to “cherry picking” — that is, having advocates focus on those elements that support their favored programs.
For example, QDR 2001 focused on “transformation,” which favored high-technology Air Force and Navy systems. Army proponents still argued for “boots on the ground” based on the stated goal to occupy an enemy’s territory and change the regime, when needed. Conversely, QDR 2006 focused on “fighting the long war,” a mission that lent itself to “boots on the ground,” but that had a limited role for airpower. Nevertheless, Air Force proponents argued for robust airpower based on the goal of “shaping the choices of countries at strategic crossroads.” The clear lesson is that, if QDRs mean to change programs and budgets, then QDRs need to say so explicitly. Like the Bottom-Up Review, they need to state which programs are, or are not, consistent with the strategy.
The juice isn’t worth the squeeze, as the old saying goes: It’s not that QDRs are useless; it’s just that they are not worth the immense effort that goes into them. The natural bureaucratic response to this problem would be to reform the QDR: Recognize the shortcomings, publish new guidance and vow to do better next time. Unfortunately, recent history has shown that this is unlikely to be successful. The brutal truth is that making the global connections between strategy, programs and budgets is hard — too hard — without some outside forcing function like budget cuts or an obviously changed strategic environment. Instead of a massive “institutional psychotherapy session” every four years, a narrowed focus and streamlined process might produce better results. Future strategy reviews should:
Focus on the few issues where the administration either must make a decision or has pledged to make a decision. Rather than institute a study team for every issue that some claimant identifies, have top leaders specify the handful of issues that must be addressed.
Consider other issues in the regular program and budget cycle. They deserve attention, but not QDR-level attention.
Adopt a modest, 5-to-10-year strategy horizon, rather than the impossible 20-year horizon currently mandated. There is a rich literature documenting how abysmal our ability to predict the future is. As Philip Tetlock has documented, experts are no better at this than “dilettantes and dart-throwing chimps,” so we might as well accept our limitations and move the planning horizon in. We’ll be doing the exercise again in four years anyway.
Narrow participation to speed decision-making and reduce the burden. Narrower participation also helps focus the strategy by excluding lower-priority embellishments. As a result, the tough programmatic tradeoffs are easier to make.
Tie strategy, programs and resources together. Where this is too hard, then the issues being considered are either not ready for change or don’t drive action in the real world. Either way, the process is wasting its time and needs to move on.
Specify a force-planning methodology. Forces are too expensive, and now greatly stretched, to be justified by vague statements about “scenario testing.”
Get it done by summer. Delay brings irrelevance.
Won’t this violate the law? Conscientious administrations could propose legislative modifications to align the QDR with these more modest aims. Alternatively, administrations could just submit the QDR report that they think is appropriate. Statutory requirements to consider specialized topics could be handled in single sentences, rather than long sections. Further, as the Government Accountability Office has pointed out, no QDR has, in fact, complied completely and explicitly with all the 17 statutory requirements.
Changing what is now a 20-year tradition will be hard, but the alternative is harder: another laborious grinding of the bureaucratic mill for limited gain.