The PPBE was really a revolutionary and radical idea. One of the things it was trying to do was reduce duplication and overlap. You might have multiple systems like an aircraft carrier and a long range bomber — which was one of the issues in the late 1940s — and there might not be enough resources. The budget would not be able to control what the services were actually doing.
So the primary purpose of the PPBE was the programming system, where budget line items look specific weapon systems — an F-16, Stryker combat vehicles, for example. And while this seems natural enough, it’s actually a sharp break from traditional budgeting practices as well as practices that now exist in the commercial sector and international governments.
The reason is that this programming system requires a great deal of prediction and consensus about future technologies and concepts of operations. These predictions are not only prone to error, they create very decision long lead times and lock-in effects.
The average age of a major development or procurement program is over 14 years today — that was right when Amazon AWS was coming out with the cloud — so you can get a good idea of how old the requirements and technologies formulated for these programs are.
That was me speaking on Government Matters, Defense budgeting system hinders rapid acquisition of commercial technology, says procurement researcher. Watch the whole thing! I discuss how the Senate Armed Services Committee included language about a major PPBE reform commission in their version of the NDAA, while the House had a scoped down report from the SecDef looking specifically at the issue of tech transition.
Here’s what I have to say on that: “Of course, the ‘valley of death’ issue is really a symptom of a broader governance struggle in the PPBE.”
You won’t be surprised to learn that I disagree, Eric. 🙂
First, whatever its original rationale might have been, the purpose of the PPBE is to convince Congress to fund specific capability investments. There’s no point in saying that the Services should fund flexible portfolios that would let them shift money around as needed — Congress simply will not authorize funds for that. It has been tried. Until Congress gets over their need to micromanage spending choices, the Services are stuck with needing to say exactly what they are going to spend the money on, every year.
Second, this isn’t the cause of the Valley of Death. The Valley of Death is a natural result of a set of issues that are only tangentially related to PPBE:
1. The Services do not invest enough in technology development, as a fraction of spending
2. The Services (and DARPA) do not mature new technologies far enough to be ready for use by acquisition programs
3. The Services use the wrong metric for assessing technology development success. The correct metric is not “what fraction of new technologies eventually make it into an acquisition” — it is “what fraction of acquisitions are able to make use of already-matured technologies”.
For acquisition programs to be able to use already-matured technologies, there need to be many more matured technologies “on the shelf” than will ever be used. The value of a hardware store is not that you’re going to need everything in it; it’s that it has the thing you need when you need it. That requires investment in stocking the store with far more goods than will ever be purchased.
Thanks for the comment Dave! Of course I disagree as well! Haha. First, the Congress issue you brought up isn’t a knock on what I was saying, but that members have self-interested reasons to keep PPBE. I won’t get into that. However, Congress did provide “lump sum” appropriations from the founding fathers all the way up until Kennedy, so the process I’m suggesting worked for a long time. Today, leaders like Jack Reed, Adam Smith, and others are openly talking about the need for change.
Second, the mature technology argument has basically been around since Packard and has been tackled so many times that it seems there’s nothing to squeeze out anymore. I won’t pull the numbers, but we know that BA 6.4 prototyping has seen dramatic increases in funding the last several years but that hasn’t solved your issues #1 – #3. Your #3 is an especially good point and is definitely sacrificed on the alter of the “program of record” concept. Programs have their baseline and cannot deviate from it without creating this unfunded risk. They were tasked to get a fixed configuration system, not to manage a portfolio of evolving and interoperable things.
My basic point is that “mature” technology comes from empowering technical people and subjecting them to rigorous scrutiny. It does not come from consensual decisions about what technology to go mature, and then find someone to go manage it and bring it to life. The best systems all seemed to have a “parent.” So a lot of this needs to be integrated from the minds of people rather than integrated from a bunch of disjointed processes spread across the bureaucracy.
Thanks for engaging with me here, Eric. I think we’re perhaps disagreeing about some of the data, and not just the philosophy.
For example, I’m not sure why you think lump sum funding worked significantly better, back in the day. After all, this is the military-industrial complex whose pathologies Eisenhower warned us about. When the veil was finally lifted (e.g. _Pentagon Wars_), the resulting round of increased oversight was fully justified by what had been going on.
Also, I’m not sure where you see “dramatic increases” in 6.4 funding. Increases that are essentially invisible in the pie chart of RDT&E vs Procurement vs O&S are not going to help much.
We agree that mature technology is a push process, not a pull process. When 6.1-6.4 funding is 30% of the defense budget, we’ll know someone is serious about that. When 80% of acquisition programs are only allowed to use mature technology, we’ll know someone is serious about that, too.
A few of responses. First, Pentagon Wars was a story about post-PPBE world from the 1980s. Second, the military-industrial-complex was not about acquisition effectiveness, it was about seizing political power.
Also, I was talking about BA 6.4 increasing, not the proportion of RDT&E to procurement or O&M. You said not enough maturation happened prior to program initiation, which is usually BA 6.5 but sometimes 6.4.
But of course I agree on push vs. pull and needing a balance, but that currently we are probably too far on the pull side. 30% of budget in BA 6.1-6.4 is interesting. Another way to think about it is free up procurement/O&M allocation to reward winners of commercial funded R&D.