Here’s my last excerpt — I promise — from Jason Crawford’s excellent interview on the Venture Stories podcast, “Progress Studies in 2020.”
Jason: I want to understand better the history of all of these funding models, and how things have changed: the decline of the corporate research lab, the rise of government funding agency, those kinds of shifts. I want to understand in a much deeper way how those have affected progress. I want to understand the stagnation hypothesis better and it’s causes — and just be able to better tie together into a story that gives people a better picture and starts to point to prescriptive answers.
Erik: what’s something you’ve changed your mind on for how you’ve changed your view on the broader field?
Jason: Certainly, everything I said about funding models is not something that I predicted going into this. It’s something that really came out from the stories I was reading. Some of the things about the different models, in particular how effective some of the corporate research labs were in the 1920s and 30s, how much fundamental science came out of them, how many Noble prizes got award to people in corporate labs, that was surprising to me. I won’t say we can or should go back to that model, but I think there is something to learn about that in an age where corporations are accused of — perhaps rightly — of short-termism and managing to next quarter’s earnings results. The fact that some corporations a century ago could be extremely far sighted and have these very long term plans, I think we should look more at what is it that causes people to think long term, what kinds of institutions and structures allow people to think and invest for the long term, that might be one of the most important things we can look into.
I was very happy to hear Jason say that he stumbled onto funding models as one of the most important aspects to technology progress. I didn’t expect it either when I dove into the history of defense acquisition. I didn’t really have a hypothesis coming into that besides a vague notion of liberal institutions, competition, and complex adaptive systems as important aspects.
Reading the literature from the 1940s through the 1960s it became clear that the Planning-Programming-Budgeting system was the most important aspect of the acquisition process. Robert McNamara and many of the economists at RAND certainly understood that the DoD pivoted on the budget process. But that was also understood much earlier by Ferdinand Eberstadt, who pointed to the program budget in 1949 as the most important aspect to the Secretary of Defense’s powers for controlling the rivalrous services.
(Eberstadt, by the way, was chairman of the Munitions Board in WWII and developed the Controlled Materials Plan, and after the war was a major player in trying to hold back the Department of Defense from completely usurping the prerogitives of the military services — he saw the budget as a way for the SecDef to create top-level policy without micro-managing administration/execution. Unfortunately, that program budget process was flat out wrong headed, because program policy inevitably descends into detailed planning prior to experimentation.)
The corporate research labs, of course, are interesting because so much new science and technology came out of them, but the companies themselves rarely were able to scale them into profitable products. Often these labs were quite independent, which fostered their creativity while at the same time making it hard to productize what came out of that. The government might have an advantage of being big enough and independent of sales to foster a conscious effort to have a diverse set of overlapping organizations that can do S&T and acquisition programs.
Final note: it’s interesting that long-range thinking is important to progress and generating new s-curves. The DoD certainly does long-range planning. It fixes acquisition program plans decades into the future. I think that’s a different kind of long-range planning. Rather than having a general road map or ancitipation about where one should go, and then pivot-pivot-pivot along with trial-and-error discovery and new theorizing, the DoD locks down a precise requirement and schedule of activities for years to come. That locks-in a direction which can not be knowable in advance, and leads to tremendous errors and wasted effort.
Leave a Reply