A viable path from start-up to a program of record has yet to be charted, and this remains a significant obstacle to entry… Creating a process whereby initial investments have gates or milestones for advancement that facilitate integration into the wider enterprise will create incentives for commercial partners (and indeed small start-ups) to work with the national security enterprise as a whole. Changes designed to accelerate Space Force acquisitions (and indeed Department of Defense acquisitions writ large) and make them more experimental, adaptive, and innovative must be brought from the periphery to the core, otherwise, the overall national security space enterprise will continue to stumble along its current path, overburdened by bureaucracy and red tape.
… If the broader, static environment is not changed, and if the large dollar items are not adjusted, small bets such as those made by AFWERX and SpEC will have little chance of transitioning beyond the early stages of development and achieving big payoffs.
That was from MAINTAINING MOMENTUM IN NATIONAL SECURITY SPACE: RECOMMENDATIONS FOR THE BIDEN-HARRIS ADMINISTRATION, authored by Mike Rodgers (R-AL) and Glenn Nye (former D-VA and current president of the Center for the Study of the Presidency & Congress).
The report recommends defining an “innovation pathway” from experiment to program of record using a sequence of gates. This is in effect the same recommendation Armen Alchian and Kenneth Arrow made at RAND in the 1950s to the systems analysts: rather than do one big study and then build the single best solution, start with a number of options and use sequential evaluation as a filter. These recommendations were actually adopted by the systems analysts, and McNamara instituted three “Key Decision Points” which are today’s Milestones A, B, and C.
See a figure depicting DoDI 3200.6 R&D cycle, dated June 7, 1962. (Reproduced from Martin Meyerson’s 1967 article, “Price of Admission into the Defense Business.”)
But ultimately the three Milestones that DoD has today are not incremental enough. There needs to be more continuous evaluation of progress and funding, as is understood by agile commercial companies. Here’s a good bit from Doing Agile Right:
This sounds obvious, but it’s surprising how many traditional budgeting systems spend months or years deciding whether to invest in a project, then never spend a day determining how the investment actually panned out. Agile budgeting is different. Agile constantly questions whether incremental investments are justified.
In DoD, there’s a bit of a misalignment that makes things worse. You have progress through milestones that depends on technical events, you have access to money that depends on multi-year budgeting process, and you have validation of requirements that depends on threat-based scenarios. In other words, creating more incremental decision-points for program milestones does not solve the misalignment problem with budgeting cycles and requirements that are the true source of the “valley of death.”
And so, Space Force requires the portfolio budgeting accounts that it desired last May 2020. Space Force would be a great place to start piloting some of these methods that have also been recommended by the NSCAI for command and control capabilities. The CSPC study came at portfolio management more from the requirements angle than budgeting:
The Space Force approach to requirements should shift from low-level system specifications to high-level capability requirements — in essence, define the ends desired, not the way it should be achieved, thereby allowing the companies to innovate
That’s another way of creating a portfolio, but ultimately budgets have to line up with that in order for it to be effective… in which case you’re still in need of portfolio budgeting. So the conversation needs to happen.
Make substantive changes
The CSPC study made a number of other interesting recommendations for the Space Force, including using an IDIQ contract model for space launch. But I like this part here:
Adjusting the margins and shaving a month or two off a multi-year acquisition program is insufficient to the challenge the U.S. space enterprise confronts. Getting the same equipment and capabilities, only slightly faster and at an equal price, is not enough progress. Fundamentally changing what is bought, and how it is bought, is the only pathway to long-term resiliency on orbit.
The recently announced awarding of the Next Generation Overhead Persistent Infrared5 satellites is a fitting example—five satellites to be deployed by 2028, the earliest of which may fly by 2025, at a cost of nearly $9 billion. While the award may have been “swifter” than previous competitions, it by no means leverages next-generation capabilities, mega-constellations, or other cutting-edge technologies. If anything, it is a continuation of the deployment of “juicy targets.”
Amen brotha…the stage gate approach is also not sufficient to maximize use of space services or take advantage of hosted payloads to me because they assume the end state is “program of record” versus a capability that might come from various sources but collectively represents the most affordable and resilient approach. For instance, the best approach to military communications might be buying bandwidth on various commercial vehicles and hosting some specific comm payloads on different commercial buses. As long as the sats have an S and X band channel and the system has an adaptable TT&C ground segment, they should be able to plug into the larger architecture. This provides affordability, security and resiliency that represents the future of space acquisition.
Good point as always. I was always uncomfortable with the Kenneth Arrow depiction of technology development. It always felt too linear — I start with a well defined problem and many possible solutions, and as I learn more about how each project is progressing I filter out the ones that don’t work until I arrive at a winner. But there is no well defined problem, but only multiple overlapping and changing problems. And projects will skip steps or find solutions to other problems or require interoperability with something new. There’s just something very wrong with the concept of the program of record, so the solution isn’t just making a better process to get there.
Reminds me of Wildavsky: “Program budgeting is like the simultaneous equation of society in the sky. If program budgeting worked, every program would be connected to every other with full knowledge of the consequences, and all social problems would be solved simultaneously.”