What we really have to do in acquisition is shift the culture that is solely focused on the balance sheet — on cost — and shift to a competitive mindset… Competing means we have to speed up acquisition. It’s far too slow in this century. The model from the Cold War brought into this century is too slow. If you look at the pace of technology change, we should be fearful of having a slow acquisition system. Who knows what the next technology that’s going to change national security is going to be. I’m sure you could get a lot of people who will come give you a talk on that. But you wouldn’t believe any of them. There’s so many possibilities between AI, autonomy, synthetic biology, and ubiquitous sensing. Who knows?
That was William Roper on the CSIS podcast, “Implementing Innovation Series: A Perspective from Will Roper, Air Force Acquisition Executive.”
A couple points before some additional quotes. First, the cost of financing a truly competitive environment isn’t insignificant. Flat out high costs on sustainment has put pressure on investment budgets, and are the consequence of past R&D decisions in a non-competitive environment (i.e., a vicious cycle). Cost growth on existing investments also takes from new program starts. The DOD “night court” zero-based budget reviews is intended to find money to put into new programs, but is it also financing competition?
The second point is the fact that the existing acquisition system was built on the premise third-party experts could effectively evaluate future technological arrangements. If you read the old congressional hearings and texts on rationalizing defense management, and you’ll see that technology was supposed to be objectively evaluated by the expert. However, if outputs cannot be forecasted with much accuracy, then creating financial plans based on expected outputs might not be efficient.
Here’s some more from Roper:
Innovation happens when you get the culture right. You have to organize a culture where people can connect to the mission. We’ll we’ve got mission in spades. We’ve got to clarify what the mission of acquisition is.
If we make the mission wringing dollars and cents out of programs — is that inspiring to hear that said? It doesn’t inspire me. It’s good to do, it’s diligent, it’s a necessary function of acquisition. That does not get me up in the morning. I don’t think it gets anyone up who is building cutting edge systems for the future warfighter. Competing against an adversary is. That is a mission. What you do today counts.
So the big shift we have today in the Air Force is that I’m seeing take root out in the field, so that innovation can happen at scale, is shifting from focusing on money to time. Everyone can plug into time. Not everyone’s job in acquisition connects to cost directly. But everyone can decide to push the envelope, do things differently, do things faster. So when you do that, and you delegate the authority so that people have a reasonable amount of control over their destiny, then creative people take the reins.
So it takes a lot of energy pumped into the system to keep this going. A bureaucracy has a natural kind of drag; it resists any kind of velocity.
This is an incredibly important point, and is often difficult to understand for outsiders to the defense business. The government wants to trace where every dollar was planned to go, and then where it actually went, in a major contract. This necessarily comes at the expense of adaptation and learning.
It brings up the distinction between cost analysis and opportunity cost. The DOD can focus on analyzing how much cost went to this kind of labor or that kind of material, but that really says little about the value being generated by that cost decision. All that valuation came up front in the paper analyses evaluated by experts.
What gets people excited to work in defense isn’t executing standing orders, or turning dollar outlays into purchases according to a preconceived plan. Rather, it is exciting to discover higher valued uses for the money, to use one’s creative energies to contribute to the acquisition process. After all, it is obvious that the same dollars used to purchase labor and material could be applied in different ways. Whether or not those resources were worth the cost or not depends on the specific actions that they take, and on the subjective opinion of the decider as to what value will result in the future. The knowledge to determine the optimal pattern of production is often held at the lower levels, and is generated in the act of production itself — not ahead of experience by those with the most scientific and engineering knowledge.
And a final part from Roper:
What have we learned that is more general about doing acquisition fast? I think that they’re not going to be new lessons, they’re going to be rediscovering old lessons. The Packard acquisition policies — you read them from 1971 — a lot of what’s in those policies sounds like it’s what we ought to be doing today. Focusing more on hardware, less on paperwork. Fly-before-you-buy. Trade off performance to keep schedule.
Those were religious dogma for me at SCO [Strategic Capabilities Office], and they matter to me just as religiously here in the Air Force. Those are old lessons that we are re-learning. I think that we’ll see that we’re just regaining what acquisition was originally meant to be, which is buying things smartly and buying them fast, as opposed to where I think we went, which is trying to make every acquisition a chance to really trace dollars and cents to such a degree that it stymied programs.
I can’t agree more that we have to look to history to rediscover old lessons. However, I don’t think we should be re-hashing the Packard era reforms. We have to move beyond that.
What Laird and Packard seemed to miss was the idea that the program budget created a lot of rigidity against a culture of speed and adaptation. After all, when congress authorizes a budgeted program, it then requires a long-tail of reporting mechanisms to ensure that the funding was indeed spent on the approved activities. That extends down into contractor business systems, and requires a great deal of work to satisfy accounting by a slew of widgets in the Work Breakdown Structure, and so forth, which then becomes the basis for estimating and justifying future programs.
Does that mean we just have to go one step further back, to the acquisition practices of the 1940-1950s? I think there’s a lot of wisdom in the use of after-the-fact controls and strong in-house capabilities, but it wasn’t perfect. I think inadequate operational testing, transitory personnel assignments, a reliance on paper studies, concurrency, a bias against competition, and a lack of procurement authorities for using commands were some real problems of that era. Many of these problems and more seemed to be the outgrowth of attempts to hyper-rationalize defense according to the so-called principles scientific management.
Be sure to listen to the whole podcast, Roper has tons of additional insights that are well worth reflecting upon.
Leave a Reply