Rationalizing McNamara’s Legacy

August 5, 2016

For special access to experts and other members of the national security community, check out the new War on the Rocks membership.

There is no question that Robert S. McNamara’s legacy is alive in the Department of Defense nearly a half century after his departure and a full century after his birth. The character of this legacy, however, is still up for debate. Is it mainly a legacy of cowardice reflecting his lack of public candor regarding the Vietnam War? Is it a legacy of sanity and bravery flowing from his efforts to control the nuclear arms race with the Soviet Union? Or does McNamara’s most enduring influence lie elsewhere?

In a recent essay marking McNamara’s centennial, mathematician and national security analyst Abhijnan Rej argued that the former defense secretary’s real legacy lies in the changes he made to Department of Defense decision-making. Rej maintains that McNamara should be remembered most for “the introduction of a body of knowledge that views national security problems, including those that are related to conventional and nuclear war, as economic problems.” “In the end,” Rej concludes, “McNamara’s legacy lies not in the choices his Pentagon made, but in the ways such choices were formulated and examined.”

Rej is right that Robert McNamara’s most important legacy as secretary of defense is the system for planning and decision-making he installed at the Pentagon. However, he overlooks the most important implications of that legacy for defense planning today. In particular, Rej’s praise for McNamara’s reconceptualization of national security problems as economic problems is misplaced. National security problems are not best understood as economic problems. They are political problems.

Making Defense Rational

Rej’s essay focuses on how McNamara and his “whiz kids” — young RAND Corporation analysts working in the Office of the Secretary of Defense — dealt with defense planning in the nuclear age. The rational choice, game theoretical, and microeconomic models the RAND mathematicians and economists brought to defense planning were particularly appealing to McNamara. He had applied similar techniques to strategic bombing during World War II and again to automobile production at Ford Motor Company after the war. The economic approach also seemed well suited to planning for nuclear war, a novel problem with no record of historical precedents from which lessons could be drawn. In fact, as Alain Enthoven, a RAND economist and one of the whiz kids, once remarked to an Air Force general challenging his analysis: “General, I’ve fought just as many nuclear wars as you have.”

McNamara’s challenge, after being appointed secretary of defense by newly elected President John F. Kennedy, was to impose order and rationality on a chaotic and conflict-ridden Department of Defense bureaucracy. Using the enhanced powers the Office of the Secretary of Defense received under defense reform legislation passed in 1958, McNamara quickly moved to impose his will on the Pentagon. He installed a planning system similar to the one he used at Ford. His Planning, Programming, and Budgeting System (PPBS) continues to exist in modified form at the Pentagon today — known now as Planning, Programming, Budgeting, and Execution (PPBE). PPBS would provide the Department of Defense with rationality by organizing the budget around functional mission outputs, instead of using military service requests as the starting point. Systems analysis would then compare the ability of different programs to perform the same mission. PPBS would produce five-year spending plans — a Future Years Defense Program (FYDP) — using the programs that proved most efficient and cost-effective based on the comparisons.

The most infamous failure of this analytical approach came in Vietnam. The Office of Systems Analysis was officially put in charge of evaluating progress in the war effort in 1966, but McNamara’s methods had been influencing the way the Johnson administration weighed costs and benefits throughout the conflict. The problem, as noted by the preeminent Cold War historian John Lewis Gaddis in his classic Strategies of Containment, was that systems analysis required quantitative metrics that could be easily manipulated. “The most notorious example,” Gaddis wrote, “was the use of enemy ‘body counts’ as the chief indicator of ‘progress’ in the ground war.” He continued:

The argument has been made that in such a conflict, where conventional indices — territory taken, distances covered, cities occupied — meant little, emphasis on these kinds of macabre statistics was unavoidable. That may be, but what seems odd is the importance accorded them, given their widely acknowledged inaccuracy.

Even if accurate body counts were available, Gaddis observed, these quantitative measures still would have likely failed to deliver an accurate picture of American progress in Vietnam. Statistics of enemies killed said nothing about Vietnamese willingness to absorb those losses in pursuit of their political goal: expelling American troops and uniting the country under communist rule.

It was not only in wartime assessment that McNamara’s quantitative, economic approach to defense planning ran afoul of political reality. Other problems in the approach are evident in the story Rej tells about the evolution of the former secretary’s thinking on planning for nuclear war.

The Kennedy administration came into office repudiating the Eisenhower administration’s reliance on the threat of massive nuclear retaliation to deter Soviet aggression in Europe. Kennedy and McNamara sought a “flexible response” that provided limited response options. The nuclear component of flexible response relied on counterforce strategy. “The no-cities doctrine fit perfectly with the economists’ view of war,” Rej writes, “If the point of nuclear assets is not just to prevent war… but also to minimize loss of life including those of civilians if it ever broke out, why not neutralize the adversary’s capacity to strike first with nuclear weapons?” McNamara thought the Soviets would follow the American shift to a “no-cities” doctrine, because both sides would appreciate the significantly lower civilian casualties it promises — a type of “mirror imaging” Rej concedes is inherent in the rational actor assumptions under which McNamara operated.

However, a few problems with counterforce targeting became immediately evident. Counterforce requires a first strike. Leaving aside obvious moral questions about launching a preemptive nuclear attack, counterforce creates pressure for an adversary to strike first to avoid allowing their nuclear forces to be destroyed. Each side also has incentive to continually expand their arsenals so that a first strike becomes infeasible. Rej credits “economic considerations” with convincing McNamara to discard counterforce just a few years after announcing it. But if McNamara wanted to distance himself from it, others at the Pentagon, the U.S. Air Force in particular, had organizational interests in perpetuating counterforce capabilities. And though public statements on U.S. nuclear strategy over the past five decades focus on the American arsenal’s retaliatory capability, counterforce remains a key justification for the preservation of the U.S. nuclear triad — the delivery systems of which are slated for upgrades that will cost in the neighborhood of a trillion dollars.

Meeting Political Reality

The promise of McNamara’s economic approach to defense planning was supposed to be the rational pursuit of greater efficiency, and thus more “bang for the buck.” Parochialism, such as the Air Force’s continued pursuit of counterforce, could be overcome through rational analysis. But McNamara’s approach to defense planning was wielded as a political weapon as much as it was an analytical tool. As one biographer wrote, the secretary of defense relished bombarding members of Congress with statistics to impress upon them the precise nature of his analysis. To paraphrase MIT political scientist and Niskanen Center fellow Harvey Sapolsky, Eisenhower fought political battles with his generals’ stars; McNamara fought them with heaps of statistics.

The problem with McNamara’s pursuit of greater economic efficiency is the lack of an objective output against which efficiency can be measured. The U.S. military is what James Q. Wilson referred to as a “procedural” organization. Because procedural organizations, while conducting many visible activities, do not provide a visible output, it is not possible to judge competing approaches in terms of their efficiency in producing it. With measurable outputs unavailable while the military was at peace (and, as Vietnam showed, still very problematic when at war), McNamara instead invented some, framing outputs in terms of major forces, such as “strategic forces,” “general purpose forces,” and “mobility forces.”

But building the budget around artificially constructed outputs as a means of measuring efficiency reinforced parochial tendencies. Because McNamara’s system required five-year planning cycles unconstrained by resources — over which Congress, rather than the military, has control — the services have no incentives to seek tradeoffs. As the late public administration scholar Aaron Wildavsky argued, because resources inside governments are allocated by the political process instead of markets, the services will fight for expensive plans they have already devised, even if funds are not available. “Since nothing is lost by trying harder, no participant has reason to stop lobbying for more,” Wildavsky wrote.

Even if economic analysis produces the most efficient plan possible, that doesn’t mean it won’t face political resistance, or won’t require the expenditure of political capital to get it approved and implemented. For example, early in his tenure, McNamara argued that the Air Force budget — which accounted for about half of the total defense budget under Eisenhower’s “New Look” strategy — should be reduced. However, even armed with statistics to back up his argument, he went further than Congress was willing to accept when he cancelled the service’s B-70 bomber program. Congressional sponsors of the Air Force balked at the decision, and only the intervention of President Kennedy himself was able to convince powerful legislators on the House Armed Services Committee to back down.

Even commanders-in-chief do not have an unlimited supply of political resources. A president can intervene only so many times on behalf of his secretary of defense, as John F. Kennedy did, against the interests of a military service and a powerful Armed Services Committee chairman. During the B-70 controversy, McNamara was able to count on the support of Kennedy, a decorated war hero. Few of McNamara’s successors could say the same.

Les Aspin, for example, could expect little political support from President Bill Clinton when conducting his post-Cold War “Bottom-up Review” of the military and challenging the “Base Force” plan General Colin Powell devised shortly before the Berlin Wall fell. Clinton had a notoriously poor relationship with the military, having been referred to early in his presidency by one Air Force general as a “pot-smoking, skirt-chasing, gay-loving draft-dodger.” And while Clinton’s Republican successor promised the Pentagon that “help is on the way” after eight years of poor relations with a Democrat in the White House, Secretary of Defense Donald Rumsfeld’s attempt to transform the U.S. military faltered as his brusque management style alienated many within the defense bureaucracy. Politics trumps rational analysis and plans are meaningless absent the political capital needed to overcome obstinate military bureaucracies or recalcitrant legislators with political interests of their own.

Rationalizing a Legacy

None of the above should be taken to mean that economic tools are irrelevant to national security analysis. They are in fact very valuable. As Rej notes, the whiz kids’ mathematical modeling in the 1960s helped show that the conventional military balance in Europe was not nearly as lopsided in favor of the Red Army as was previously believed. Today, quantitative methods can provide insights into intractable defense problems, such as military readiness. The usefulness of these tools, however, depends on recognizing the issues at stake for what they are. It means recognizing the political logic of the problems that need to be addressed. Economics is a useful tool. But the essence of national defense is still political.

McNamara’s legacy is still very much alive and well at the Pentagon though. His PPBS system is still in place in its slightly modified PPBE form. The FYDP it produces still serves as political touchstone upon which various organizational actors can lay claim to resources instead of adjusting to new political realities. The economic mode of thinking McNamara brought to defense planning was supposed to impose rationality and efficiency on a complex and often chaotic bureaucratic system. By ignoring the political nature of defense decision-making though, the economic tools McNamara ironically left the system less efficient than he found it. One hundred years after his birth, and five decades after he served as secretary of defense, that is McNamara’s legacy.


Matthew Fay is a defense policy analyst at the Niskanen Center and a PhD student in the political science program at George Mason University’s School for Policy, Government, and International Affairs.

Image: DOD

We have retired our comments section, but if you want to talk to other members of the natsec community about War on the Rocks articles, the War Hall is the place for you. Check out our membership at warontherocks.com/subscribe!

3 thoughts on “Rationalizing McNamara’s Legacy

  1. Matt, I commend you on an excellent article. All I would add is that the additional book to read by McNamara’s colleague, economist Dr. Alan Enthoven and K Wayne Smith entitled “How Much is Enough”. It is available from RAND online free of charge. Dr. Charles Hitch proposed PPBS, but McNamara was the one, with Enthoven, that imposed it on the Pentagon. He also created and served as the acolyte for (now) three generations of analysts, many of whom helped lead the fight to pass Goldwater Nichols. The legislation cemented many of McNamara’s centralized acquisition and test and evaluation principles in place within DoD that now appear to be a root of delayed programs and massive cost overruns. One might now ask how much analysis and operational testing is enough?

    It is amazing that a 50 year old output budgeting system designed to produce cars at Ford Motor company remains the guiding principle of U.S. defense acquisition.

  2. What amazes me most, given the phenominal accounting done in military procument; is how little of use it produces.

    Over the fence here in DoD contracting land, after some study we decided to use widget A with the military’s concurrance. Widget A is designed to operate using subwidget 1. It is inoperable without subwidget 1. No substitute for subwidget 1 is possible. Subwidget 1 uses a battery designed only for the DoD. Subwidget 1 is only designed for the DoD. The battery for subwidget 1 is also an exclusive DoD product. So to integrate subwidget 1 with Widget A as the engineer, I have to request the military provide me one each subwidget. Unfortunately I separately have to request, not to be given a scheduled resupply of said battery, but a single battery as needed. And then the mandated test equipment also uses subwidget 1’s cousin, subwidget 2, also using a battery. Yes, a different battery.

    I have probably burned less than 100 hours designing an interface for Widget A. Getting subwidget 1 has consumed over 500 hours and batteries even more as it has to be repeated regularly. Then there is the hours spent explaining that”Yes, I have subwidget 1, but also need subwidget 2 and who do I discuss batteries with, please”?

    Yet were the DoD actually to adopt the methods of Ford motor, a subcontractor told to integrate Widget A would have triggered the supply chain to also provide subwidget 1 and a fresh battery on it’s replacement cycle.

    So my conclusion as an outsider is the DoD picked up the horrid bureaucracy of Ford, without any of the efficiencies it can produce.