Something truly astonishing happened this month, although many in Washington were too busy chattering about “the broken defense acquisition system” to notice. When Defense Department officials totaled up the cost growth on the 38 major defense acquisition programs (MDAPs) launched in the wake of the landmark Weapon Systems Acquisition Reform Act of 2009, they found that the grand total was…negative. That is, costs actually declined.
These 38 programs represent half of all MDAPs, so this result is probably the best evidence to date that acquisition cost growth is not, in fact, the grave threat to humanity and national security we had imagined. And while it is premature to declare victory over cost growth, perhaps it is at least appropriate to consider whether such growth is really the acquisition system’s central problem. To make a long story short, the answer is: no.
But hold the parties on K Street and the clapping of slide rulers in certain halls of the Pentagon; the end of history has not come. The real challenge to defense acquisition, the enemy in whose service cost growth was only ever a minion, is the rapid erosion of the technological advantage upon which U.S. national security strategy depends. The United States is being challenged by Russian aggression in the Ukraine and elsewhere, underpinned by dramatic improvements in Russia’s military technology and operational proficiency; and by Chinese aggression in the South China Sea, enabled by military reforms that have driven, among other things, advances in anti-access area denial capabilities. It is even being challenged by ISIS and other non-state actors, who are using car bombs and other relatively low-tech systems to confront national armies.
While the defense acquisition system has contributed to the problem — maddening program failures and drawn-out acquisition timelines are still too common — the source of this erosion lies elsewhere. It is driven by competitors’ focused investments in advanced capabilities, and by the increasing availability of commercial technologies that bear on military problems. Consider the iPhone, which essentially hosts in one pocket-sized device all of the imaging, networking, and geolocation capabilities that the U.S. military developed over multiple decades, then used to overwhelm opponents. The diffusion of sophisticated military capabilities is outside of anyone’s control, and is unlikely to be limited or reversed. U.S. adversaries will use them to undermine, challenge, or even exceed U.S. systems in certain niche areas that support their objectives.
In many ways, the defense acquisition system was designed to confront this sort of challenge. Born of the competition with the Soviet Union, the system produced cutting-edge capabilities and upgraded them to stay one step ahead: for example, the C-130 and F-16, whose block upgrades and multiple variants have allowed them to accept new missions. But most of these modifications took years to develop and field, and that pace is now far too slow. As vividly demonstrated by the amazing variety of improvised explosive devices fielded in Iraq and Afghanistan, even unsophisticated adversaries can use cheap weapons and commercial technologies to upgrade their striking power over the course of months and even weeks.
The Defense Department’s much-discussed Third Offset Strategy, its new outreach to Silicon Valley, and its internal Better Buying Power 3.0 initiative are all responses to this fundamental problem. Each is important and worthwhile, but even more needs to be done. There are important steps the department and Congress, working together, can take to add agility to the defense acquisition system so that it fields an enduring technological advantage. Among them: increase the prominence of DoD’s head of research and engineering, as suggested by Dr. John Hamre, former deputy defense secretary of Defense and president of the Center for Strategic and International Studies.
Needed: A New Lane
The past decade of war taught us how to field weapons and gear faster through dedicated rapid-acquisition lanes. Now we must apply those lessons to create a new lane for adaptable systems. The best example is the Predator drone, which has been continuously modified, upgraded, and morphed into new variants to respond to new threats and new technology. Today a program of record, Predator didn’t start that way and has never followed the typical path for a major weapon. If we’re lucky, it never will.
To start with, this new adaptable-systems lane needs more financial flexibility. Today’s new efforts generally take at least two years to start receiving funding; even when existing budgets contain relevant funding, a year can pass before the money can be redirected. This is far too slow for areas of fast-moving technology. The rapid-acquisition lane has successfully used flexible funding accounts, rapid reprogramming, and rapid acquisition authority; the new adaptable-systems lane needs these as well. Just as important is flexibility in the “color of money” — that is, the rules that separate the funds for development and testing from those used for production. A truly adaptable system will be simultaneously in production and development for most of its operating life.
Also needed: a modified requirements process. The standard approach focuses on system-level attributes, extended analysis, and inflexible performance parameters that become pass/fail criteria. Adaptable systems need dynamic requirements that can be revisited and modified constantly, as well as a recognition that system-level attributes will often be less important than the subsystem attributes they enable.
Even baselines and other foundational acquisition concepts need a rethink. In the standard approach, baselines are established when a program enters development, are rarely modified, and are used to judge the program’s success. For an adaptable system, however, deviation from the baseline is to be expected. It is a feature, not a bug. A better approach would ensure reasonable planning and accountability, but would also allow the program to evolve through use of a dynamic baseline. Having a dynamic approach to requirements and baselines will also require testing regimes that keep up with many program changes.
Ultimately, having technology that is ready to go and organizations prepared to do things differently will be critical. Prototyping and experimentation will provide the technology feedstock for the adaptable systems lane. As well, Defense Department leaders must offer their support as systems enter the adaptable systems lane to ensure they receive the resources and attention they need.
This concept of an adaptable systems lane complements current Congressional efforts on defense reform. Recent hearings in the House and Senate have sought to identify areas where reforms are possible. A public web survey on defense reform that my think tank, the Center for Strategic and International Studies released on March 14, finds that a large majority of the public believes acquisition reform remains deeply needed. The chairman of the House Armed Services Committee, Rep. Mac Thornberry, has said he will include in the 2017 defense authorization bill language to foster more prototyping and the use of modular open systems architecture to make it easier to modify and upgrade systems. His counterpart in the Senate, Sen. John McCain, has likewise indicated that he is looking to make reforms this year. The time is ripe to give the DoD the tools it needs to field an enduring technological advantage.