I was pleased to have Russell Rumbaugh join me on the Acquisition Talk podcast to discuss some of the most fundamental issues facing defense management. He is the systems director at the Aerospace Corporation’s center for space policy and strategy. Though we went off script for nearly the entire episode, it was the conversation I wanted to have with Russ. We hit questions including:
- Is weapons choice a political or technocratic process?
- Is there a boomerang effect in defense S&T?
- How can shared data be used to improve analysis of the force structure?
- Is the DoD organization at odds with portfolio management?
Download the full-text transcripts here.
Was defense really more innovative in the 1940s and 50s?
Certainly in the popular imagination, defense programs seemed like a tremendous source of innovation in the middle of the 20th century. Today, there seems to be more of a malaise about defense technology and admiration for the commercial sector. Should we conclude that defense simply worked better in those days, or is something else going on?
Russ makes two key points on this question. First, greater financial risks were taken in the 1950s:
When Bennie Schriever out at the Western Development Division, he was getting 5% of the entire defense budget. Absolutely he did amazing things. Aerospace is very proud of being part of that history, but we, as a nation, we’re also throwing resources at it. Similarly, same timeframe, the Navy’s Polaris program that made it up to 4% of the defense budget. In contrast our last big crash course, that JEIDDO the IED office. They never got 1%. Even missile defense, with the peak of Star Wars, still only got about maybe 2%.
The Polaris ballistic missile chewed through $2 billion a year in the early 1960s, which comes out to roughly $14 billion a year if you adjust for inflation. But that would represent just 2% of today’s DoD budget. If real prices in defense acquisition grew at 1.5% above inflation each year (which is about the same as Navy shipbuilding real price growth), then you get to $32 billion a year — getting you back to 4% of today’s budget. By comparison, total F-35 acquisition funding for FY2021 stands at just $12.8 billion. Imagine if the F-35 program was two or three times larger than it is today!
With that level spending, DoD also had better access to talent in the early Cold War era when it was the primary engine of R&D. In 1960, the federal government spent 50% more on R&D than the entire US commercial sector, whereas today it is at an 8-to-1 disadvantage. The Vietnam War accelerated this shift in access to talent as universities began to distance themselves from the military.
But we must ask: why was there a greater willingness to take financial risks? Why did Congress allow DoD to prioritize new systems like nuclear submarines, ballistic missiles, and jet aircraft at the expense of the “legacy” force structure? Sure, that led to some imbalances that McNamara tried to undo with his “flexible response” strategy, but the force structure of the 1960s was generations ahead of what existed just ten or twenty years before.
Russ caught me with a good point on this front:
How do we do getting rid of the old stuff? That’s a harder question right now. I’ve walked back into your redundancy question. If we’re talking about getting rid of old stuff, aren’t we talking about getting rid of the redundancy?
Unfortunately, military systems cannot be tested in real environments like consumer products to provide data as to whether something adds to resiliency or whether it is simply inferior. But that probably strengthens the idea that RDT&E should provide interesting and novel options for rigorous experimentation to provide a basis for such judgments. Any individual question or retirement cannot be made in advance of the replacement’s development.
Onto Russ’ second point on why we should be skeptical whether the defense acquisition in the 1940s-50s period worked better:
The real way to close this question is there’s a Harvard study from 1962, Peck and Scherer, and they argue all the programs they looked at — for development — were averaging 3.2 times the estimated costs and 1.4 times the estimated time. If in 1962, the programs are already going off the rails, that’s probably a good sign this isn’t a new thing.
While that’s true for the 12 missile programs Peck and Scherer looked at, cost growth just tells us how much money was spent relative to some prediction. Controlling cost growth means little or nothing about value achieved per dollar spent. There is even a theory of cost growth that talks about organizational incentive to intentionally underestimate costs to get a program started.
The institutional context of the 1950s was very different from today, which has become fixated on cost growth. But the idea was that by killing cost growth leaders could achieve long-term stability of the total program portfolio, and that such stability correlated with increased value. The question is whether such stability comes at the expense of innovation, pushing program choice towards incremental changes like JSTARS Recap rather than ABMS.
Portfolio Budgeting
Here’s a good edited-down exchange where Russell gives me the Socratic method:
Russ: The army really looks at things only from a land perspective. And the Navy only looks at things from an sea perspective. And yet almost always the problem the president faces are not domain specific. So that tension is fundamental. I love portfolios. If we could get rid of all of our existing organizational structure and only use portfolios, I’m for it. It seems a little rash.
Eric: Can I ask why do you think there would be a major reorganization with portfolios? Because my imagination is just consolidate budget lines around the existing program executive offices, laboratories, and other lifecycle management centers. And so there’s no reorganization.
Russell: What happens when your portfolio champion disagrees with the chief of staff of the army about where to move money?
Eric: The chief of staff is ultimately the boss, right? So my view is not all discipline has to come through the budget. Discipline that guy through the administrative hierarchy.
Russell: So your portfolios are only within military services. Portfolio is don’t overlap the services.
Eric: The JPO [Joint Program Office for the F-35] would be a multi-service one, that’s an organization that would be a portfolio and it currently fits. But for the most part, they would be separate. The current organizational structures deliver to joint war fighting in the combatant commands.
Russell: What happens when the chief of staff of the army and the commander of central command disagree with each other about what the joint warfighting constructs?
Eric: I’ll take a quote from John C. Rice, “On troublesome issues where sharp differences occur among subordinates, no substitute exists for the consideration of opposing views ably argued. The secretary must be able to examine the proponents carefully convinced them that he is familiar with the consequences of the decision and is intolerant of superficiality and is willing to use his political power fully if necessary in resolving the issues.”
Russell: But it turns out there’s more disagreements in the department of defense than the secretary of defense has time for, and that’s a knowledge problem.
Eric: I’m always trying to consider the invisible hand context. How do we use dispersed knowledge to self-coordinate in such a way that we actually arrive at better decisions.
And it goes on from there, so be sure to listen to the whole thing! This exchange gets to one of Russell’s points about the fragmentation of the defense establishment:
For more than 40 years, when the nation has decided a military function is not being fulfilled, instead of holding the military services accountable, it has created a new organization using new funding and organizational models to focus on that mission.
Thanks Russell Rumbaugh!
I’d like to thank Russell Rumbaugh for joining me on the Acquisition Talk podcast. You can find some of his recent work at his Aerospace Corporation page, and be sure to read his The Fragmentation of DoD paper. Watch Russell on the Space Policy Show on Force Design, the Space Budget, and USSF organization. He’s written some interesting papers at the Congressional Research Service including Defining Readiness. See more of his works from Google Scholar including Real Acquisition Reform and an interesting take on the COCOMs in The Best Man For the Job?
Leave a Reply