We all want acquisition reform. We want our forces to acquire new weapons faster, cheaper, and with the most advanced capabilities possible. The problem is that these goals trade off against one another. If we attempt to accelerate weapon development schedules, we usually increase costs and have to sacrifice system capabilities. If we seek lower weapon costs, we likely are asking for stretched schedules and/or lower weapon performance. Set higher performance requirements and usually you are encouraging higher costs and more development time. Moreover, because weapon development advocates, the services and also civilian officials and contractors have proven in the past to be overly optimistic in their estimates of weapon system development schedules, costs and capabilities, we have been adding more and more reporting requirements, decision reviews, cost analyses, and independent testing, all of which complicate the weapon acquisition process and add to its costs and hinders its agility. And because the government is the ultimate customer, we lace the acquisition system with social goals — audits for the detection of fraud and favoritism, requirements for the use of small businesses, procedures to assure access for minority and women owned firms, assignment of work in areas of high unemployment — that cannot be ignored and that add to the process’ complexity.
But we are smart. Surely, we ought to be able to figure out an optimal path through the acquisition maze. Unfortunately, security is a cyclical enterprise for America, froth with emotion and politics, which means it is impossible to make the acquisition process totally into an engineering design problem. We routinely go into periods of security panics, mobilization for war or near war — Korea/the beginning of the Cold War, Vietnam, the Reagan Buildup, and more recently, the Global War on Terror — when we cannot spend enough on defense and worry only whether or not our troops have the very best equipment. Wars end or fade away. What follows are periods of security relaxation and puzzlement where there is confusion about the nature of the threat, regrets over past spending, and demands to put defense dollars to different uses. On the panic side of the cycle concurrency is common practice, with weapon systems being rushed into production before development is completed, a certain formula for trouble but one temptation that is almost never resisted. And on the down side of the cycle weapon developments are stretched and new ones are hard to start. On the up side rules are waived. On the down side rules are remembered and enforced.
Security involves trends as well as cycles. The most important trends affecting acquisition are those favoring greater centralization and less competition. Almost from the enactment of the National Security Act of 1947 there were efforts to increase the supervisory powers of the Secretary of Defense and weaken those of the services. These efforts culminated in the Eisenhower years with executive reorganizations that led to the full subordination of the service departments to the Secretary of Defense and his staff. Later the Packard Commission reforms brought increased centralization in acquisitions, making its management an OSD task, while the Goldwater-Nichols Act sought to force Jointness, centralization by another name, on the services themselves. And with the collapse of the Soviet Union came the end of the nuclear arms race, a mudding of the threat, fewer new systems starts, mergers among the defense contractors, and less competition among both the contractors and the services.
The United States cannot avoid security cycles. There will be times when we are at war or fear one is fast approaching. That is when we throw money at problems and abandon procedural constraints. We want ballistic missiles and we want them now! We want MRAPs and we want them now! Rules become obstacles and are bypassed. But on the downside of a cycle, when the security panic/war is over, we complain about how indecisive the acquisition system is. Why is it taking so long for the F-35 to become operational? Don’t we need a dedicated close air support aircraft? Should we be developing a new tank? For what war is the Army preparing? Does the Navy have to design another frigate? Wouldn’t more small carriers rather a few bigger ones be better? Strategy is not much of a guide for acquisition then because without an agreed upon and pressing threat it is difficult to have widely accepted, well-articulated strategy that provides such answers. Instead, we fight over procedures and budgets, champion a favorite program as being transformational, and foolishly heap blame on the acquisition system for the resistance and uncertainties we experience.
How then do we reform the system? There is little to be done about the rapid run ups to the cycle peaks, times of stress when the rules will be pushed aside for the immediate needs of combat or its anticipation. We know not to do concurrent development and production, yet we let it happen anyway. Political leaders cannot take the heat, the risk of being accused of not doing enough for our forces. No set of rules can change this. But we can at least marginally improve the acquisition experience during the down time before threats clarify and remobilization begins. The way is to seek more decentralization and competition in the system, to resist and reverse the security management trends noted earlier.
The answer lies, I think, in giving more power, not less, to the services. If we are uncertain about the threat — is it China, Russia or terrorism? Is it growing or declining? We need more eyes on the problem. The services see themselves as better situated for certain threats and will emphasize and prepare for them. Their overlapping capabilities push them into each other’s business. The services’ assessments and preparations can be compared and form a reserve of countermeasures when threats clarify. We can have additional hedges by encouraging them to build prototypes and conduct experiments. The services are each other’s toughest critics, and through their bickering help policy makers keep informed about options. The services are also the natural basis for acquisition competition. Weapon acquisition is technically a monopsony with the government as the only buyer. Jointness emphasizes the crucial limit of monopsony, the opportunity to make very big and lasting mistakes. Contractors have to line up with the preferences of a joint buyer or risk their existence. With more service rivalry, there is room for contractor differentiation as there are multiple buyers with different preferences seeking to develop systems. Competition encourages innovation, monopoly stagnation. For acquisition reform that will make things at least a bit better, we need to have more inter-service rivalry, not more Jointness.
Truth be told, the acquisition system, even as presently structured, is hardly a disaster. No other nation fields forces with better equipment than ours. We are at the frontier in nearly every area of military relevant technologies. We have the best aircraft, space systems, warships, and combat vehicles, and are capable of having our forces operate and be sustained anywhere in the world, no matter the terrain or conditions. Our forces do not long for Russian made submarines, French aircraft or Chinese tanks. Their gear is envied, not disparaged. Neither Apple nor Google could do better in developing weapons subject to the same decision making and budgetary constraints as are our current contractors. And no level of our civilian government has a better acquisition record. Witness Boston’s Big Dig, the DC subway system, the Denver VA hospital and a million other projects, none of which work on truly advanced technologies as those required for our military.
Why then all the complaining about the acquisition system, the search for reforms? It is in part the cycle again. Participants can recall a time when there was money to do everything and the rules were not that burdensome. Programs moved faster and with less friction. There was room in the budget to do many things. No one’s favorite weapon was shunned. Why can’t we get it right now, they ask. But really they are asking why aren’t we on a war footing again. It is also, though, that the costs of new weapons keep growing. The bills are staggering. New aircraft now exceed a hundred million dollars a copy. Warships cost in the billions of dollars. And armored vehicles are priced individually as high as the average seaside mansion. Something must be wrong. But the weapons must deal with past challenges as well as new ones. An aircraft carrier, for example, needs to cope with threats posed by mines, submarines, aircraft, missiles, and surface ships all of which have evolving capabilities. Costs are high because we value greatly the lives of our servicemen and women. We keep getting better at war fighting, and so do our opponents.
Harvey Sapolsky is Professor of Public Policy and Organization and recently retired from teaching political science and directing the MIT Security Studies Program.
Find Archived Articles: