Podcast: Big questions in defense acquisition with Col. Bryon McClain

I was pleased to speak with Colonel Bryon McClain on the Acquisition Talk podcast to discuss the biggest questions facing the defense acquisition system. He is the senior acquisition course instructor at the Eisenhower School, and before that he was the materiel leader for the rapid acquisition branch in the MILSATCOM directorate at the Space and Missile Systems Center. We touch on:

  • The role of interservice rivalry in military innovation
  • How today’s processes address risk but not uncertainty
  • Moving past cost accounting when monitoring contracts
  • How unintended consequences plagued the C-5 development
  • Whether reform should focus on workforce culture or rules/regs

Agility Prime

One interesting idea Bryon touched on is the concept of anti-statism, in this case related to the idea that government won’t pick winners and losers but instead write the rules of the road. In defense, however, the government writes the contracts with industry. So it has to pick winners and losers in competitions, right?

In some cases that is true — DoD is living with the winners of previous competitions and they are quite good at delivering that capability. For example, there’s only a few yards qualified to build sustaining capabilities like destroyers. But for disruptive capabilities, DoD could help create markets.

Bryon points to the Air Force’s agility prime initiative, which seeks to support commercial eVTOL companies with the expectation of military applications to follow. While government may let contracts which still has a winner/loser aspect, there are many other ways about it. For example, providing companies access to test ranges and facilities, or providing industry subsidies and tax benefits. Here’s Bryon:

If you focused and tailored that money, you could either (a) help suppliers get over that hump or (b) you could create a market, thus incentivizing suppliers to jump in on it, put in their investment because they see that market that’s out there. Another example, one of the examples she uses is solar panels and electric cars.

As companies move down the cost curve along with private capital, DoD becomes well placed to access disruptive innovation earlier and create dynamism in its industrial base. This strategy can work well for other interest areas including quantum.

Principal-Agent Problem

Bryon uses the lens of the theory of agency, seeing the defense acquisition outcomes as resulting from a chain of overlapping relationships between individuals. Of course, there is no “government” or “industry,” it’s all a network of people with responsibilities to one another. The program manager is the principal to the prime contractor, but is also the agent on the program executive officer, who him or herself is an agent to the acquisition executive, and so forth. One implication of that is the need for a monitoring mechanism. Here’s Bryon:

If I want to  level the playing field, I can spend time learning, researching, becoming smarter, but that’s time I could be spending doing something else. So there’s an opportunity cost. I can also create some form of a monitoring mechanism.

 

In the government contracting world, everybody knows that when you do cost plus contracts, it comes with a huge overhead of cost accounting systems… They’re all about leveling the information playing field. So the question to me becomes, are there other information tools that we could use to level that playing field?

That is a great way of honing in on the problem. Listen to the whole thing for Bryon’s recommendations. I wanted to point out, however, the point is to have a successful relationship without leveling the information playing field. As Admiral Rickover said, “As long as a man will accept dictation in a technical matter he is not a professional person.” Tightly specified development contracts that go through an IBR are tantamount to having professionals execute standing orders — all the problems have already been solved.

Ultimately we want to unleash information asymmetry to promote innovation rather than restrict it to stop the bad. Unfortunately, I don’t see any other way around it but focusing where Rickover did: training. The one thing I might add is shortening feedback loops as much as possible. Good performance should lead to growth and poor performance to a filtering process. Here’s Bryon:

So there’s a lot of movement in program managers. If you pick up a contract that’s a relational contract that may have been running smoothly, but you don’t have a level of technical knowledge on the specific topics. And that allows the contractor some wiggle room. Then who do you blame?

Unintended Consequences of Systems Analysis

Bryon recounts a great story about the unintended consequences of incentivizing performance. The Air Force, he says, creates all its cost models using weight. A lighter aircraft should be cheaper to produce and sustain according to the models, so the Air Force wrote strong penalties for going over weight in their contract with Lockheed on the C-5 transport. And surprise, the C-5’s weight started to grow in development. Here’s Bryon:

The Lockheed solution was to look at the wing spar and realize that there was a lot of weight there because they had over-designed it. So they reduce some of that over-designed by creating some holes in the wing spar. And that in theory didn’t break it, it still matched it, but it may have made it a little bit easier to break.

 

Sure enough, after we’ve accepted delivery, many years down the stress fracture start showing up in the wing spar and we end up replacing those at cost to the government. And so that to me is a specific example of unintended consequences. We wanted to keep our costs down in the long-term by putting a proxy metric based off of weight and the end result of that was increased costs in the longterm.

That’s a good example of Goodhart’s Law: “When a measure becomes a target, it ceases to be a good measure.” The cost per pound of the airframe or the empty weight might have a lot of analytical value of actual data, but it is probably a bad metric for writing speculative contracts that push the boundaries of engineering

Thanks Col. Bryon McClain!

I’d like to thank Colonel Bryon McClain for joining me on the Acquisition Talk podcast. I have uploaded a working copy of some nice slides from Bryon on a unified framework for acquisition. As for that agile development chart Bryon mentioned, it’s here. There’s a bibliography of literature Bryon discussed in the episode below, and full transcripts of our discussion below that.

Bibliography

Transcripts

2 Comments

Leave a Reply