November 9, 2015
Of DEF and RNDF
Starkly different defense conferences discussed why military procurement is still so broken.
By James Hasik
First, let me recommend last month’s insanely great video of Jetmen Yves Rossy and Vince Reffet escorting an Emirates A380 into Dubai. It just must be seen—if you didn’t see him steal the show at Oshkosh in 2013. Jetman is what one gets when a crazy Swiss fighter pilot decides he’s Tony Stark, builds a personal jetpack from German hobbyist engines, recruits an equally crazy (aren’t they all?) French wingsuit pilot, and convinces a Gulfi airline of the awesomeness of the whole thing. Instruments aren’t necessary, Rossy explains, because the pilot can see and feel everything. There’s but one mechanical control: a rolling throttle switch. The pilot’s body serves as fuselage and sole control surface. Add a parachute and legs in lieu of landing gear, and you’ve got a human flying machine. The simplicity (someone call Dan Ward) of the whole thing is brilliant. Oh, and there’s a TEDtalk too.
So Rossy has also shown us how to build a small, tailless batwing fast. Building a big, tailless batwing is evidentally another matter. I’m still annoyed by the protest of the Long Range Strike Bomber (LRS-B) decision, rather as I was annoyed by the protest of the Joint Light Tactical Vehicle (JLTV) decision. As I wrote the other day, giving every would-be contractor a fee-free do-over mostly just adds a hundred days to every big procurement process. Why am I so irritated? Innovation means new and at least intendedly beneficial, but benefits are perishable, as technologies evolve and political conditions change. So innovative later may not be as valuable as innovative now.
Jetman could be one way to build an air force from scratch—if you only needed to fly at 180 knots for 10 minutes. For longer missions, there’s budding enthusiasm in ISIS for little drones. They may not kill us all, RAND assures us, but they’re bound to be annoying. Rethinking the possible from the ground up, or considering problems from the most impoverished point of view, is what Vijay Govindarajan and Chris Trimble of Dartmouth call reverse innovation. Lt. General Robert Brown of I Corps recommended this mode of inquiry to us at the 2014 DEF. After all, those off-the-shelf MRAPs from South Africa were less inexpensive than Strykers and Bradleys, but IEDs were a lot less expensive yet.
So why can’t the Pentagon do this stuff? The classic case remains the MRAP. Byron Callan of Capital Alpha wrote to me from the RNDF about how two speakers had invoked that vehicle as an example of how materiel can be fielded rapidly—in just 18 months. He had another recollection; I’m writing my doctoral dissertation on the MRAP, so I did too. A cyber defenses expert at DEF asked me whether I was looking at the slow phase or the fast phase of MRAP. Indeed, the details are a little different—the Army ordered its first (a Buffalo from Force Protection) in February 2000, but its 1000th MRAP only in February 2007. Seven years is not fast with off-the-shelf equipment, so if you thought that was fast, you probably need to rethink what’s fast.
When bureaucracies are bypassed, though, things do go faster. Journalist Erik Schechter asked me over the weekend why that couldn’t always be the case—why couldn’t the whole of the Pentagon’s acquisition secretariat act like its Joint IED Defeat Organization? One can often, I cautioned, spend too much and test too little when moving too fast (in this case, someone call Dan Kahneman). The JLTV is indeed a good counterexample of that. It’s a much lighter vehicle than its forerunner the M-ATV, but with comparable protection. That’s what’s possible in a three-year engineering process, when one isn’t racing to meet a wartime need. But mostly, the entirety of the American military acquisition system is moving all too slowly.
Need another example? At DEF, Petty Officer Richard Walsh of the US Navy told us how he watched his “beneficial suggestion” (again, is there another kind?) for a crowdsourced, natural-language-processing maintenance tool take four months to get up to and past his captain. That was two years ago. A sympathetic admiral tried to track down the paperwork for him, but failed. Meanwhile, Walsh and his mates have written the code and tested the prototype. Why did the MRAP take so long? In part, for a similar reason—someone at Quantico lost the urgent request for a thousand of them in 2005. In 2008, the Inspector General couldn’t prove it, but word was that rear-area support really preferred to wait for the JLTV. Remember, that vehicle was conceived all the way back in 2005, but is still not out of the protest process. Three years is good, but ten years is not.
What keeps bureaucrats from acting more like jetmen? It’s not just the risk-aversion over the over-supervised. It’s also the petty tyranny of the administratively terrorized, and the casual disinterest of the chronically under-motivated. I had to sharply disagree with Maura Sullivan, the Navy’s chief of innovation, when she asserted that “it’s all about the incentives.” It’s also about staffing, education, command structures, organizational culture, and even fundamental assumptions about how militaries must do business. From the RNDF, Pete Newell of BMNT Partners (tagline: “Startups + Defense = Life-Saving Products”) may have put it best on Twitter (@peteranewell):
#Kendall on dif btw .com & .gov is .gov has a req to be fair. Last time I was shot at “being fair” isn't what came to mind. #RNDF #DEF2015
As I wrote back in April, fairness is “an economically intractable concept.” Worse, as Colonel Newell (a former commander of the Army’s Rapid Equipping Force) points out, it's militarily madness. ISIS does not care about fair. BMNT is on point with this; the firm sent another partner, Jackie Space, to DEF to tell us about their work connecting Silicon Valley to the Pentagon. I’m still processing her assertion that defense acquisition doesn’t require an overhaul, but just tweaking. Ultimately, I don’t care if the next new thing needs to be a jetsuit, an armored vehicle, a software suite, or a drone defense system. Whatever rulebook Defense writes for guiding its staff in making decisions about materiel, the policies mustn’t continue to be taken as convenient excuses for inaction.
The mass drone-selfie at the conference didn’t come off, because the drone was broken and couldn’t come out. So I guess that I’m still safe from the swarms of flying killer robots. When they do show up, though, I won’t call the Pentagon for help. Because Skynet will have another hundred days in which to hunt me down while the GAO is regrading the papers.
James Hasík is a senior fellow at the Brent Scowcroft Center on International Security. The Atlantic Council was a sponsor of DEF 2015.