High Drama Over F-35 Cost Overruns: Can the Pentagon Fix This Procurement Mess?
These revelations come as no surprise considering the history of this program. The Government Accountability Office concluded that F-35 estimated acquisition costs have increased $46 billion and development extended two-and-a-half years compared to the program baseline approved in 2007.
The price per aircraft projected at $69 million in 2001 is now up to $112 million, according to GAO. The Pentagon plans to acquire 2,443 jets for the Air Force, Navy, and Marine Corps. Foreign nations also are expected to buy the aircraft.
Predictably, Pentagon officials were grilled today on Capitol Hill. The Defense Department acquisition team assured lawmakers that there is no need to panic, that problems are under control and that it will be smooth sailing from now on. But will it?
Cost overrun bombshells such as the F-35 are exactly what acquisition reforms over the past two decades were supposed to prevent. Last year, another major piece of legislation, theWeapons Systems Acquisition Reform Act, dumped even more strictures and oversight into the process. Obviously this latest round of reforms would have come too late to affect the course of the F-35.
But even the Pentagon’s own acquisition experts are skeptical about the ability of WSARA or other reform initiatives to avert massive overruns in future programs.
WSARA harps on the need to estimate costs accurately early in the development phase, and to come up with “independent” cost calculations before a program even gets off the drawing board. That would allow the Pentagon to reject low-ball bids from contractors who under-price systems to beat their competitors. In hindsight, if the Pentagon had accurately estimated F-35 costs when the program was conceived in the mid-1990s, would it have done things any differently?
It appears that until Defense Secretary Robert Gates took over at the Pentagon, nobody had seriously calculated the cost of F-35, nor had the program been funded to fully cover it.
“For the first time the program has been funded to the independent cost estimates for production,” said Matthew J. Schaffer, deputy director for conventional forces, cost assessments and program evaluation at the office of the secretary of defense. “This emphasizes the secretary’s interest that we implement these reforms,” he said last week at a Precision Strike Association conference in Arlington, Va.
The principles that underpin the WSARA legislation -- stable requirements, funding independent cost estimates, securing favorable contract terms, injecting mature technology -- are all commendable, said Schaffer. But they are not new. “We tried the same things in the past,” he said. The only difference now is that “We have decision makers who are willing to make decisions based on these principles.”
But Schaffer cautioned that these early cost estimates are easier said than done.
Members of the Pentagon’s new CAPE organization (Cost Assessment and Program Evaluation) gathered for three days in Williamsburg, Va., in February, to discuss the impact of WSARA. The burning question there was “How are we going to estimate costs that early in the procurement cycle?” Schaffer said.
Predicting accurately how much a system will cost early in its design phase -- years before it’s actually produced -- is a daunting prospect, he suggested. “There won’t be the usual details that are used for a detailed cost estimate,” he said. “It is going to be much more ambiguous. How do we go about doing it?”
These are difficult questions, Schaffer said. Early cost appraisals will have to consider whether the technology is mature, if there are alternatives, or if an “evolutionary” approach is warranted, rather than trying to bite off too much technology all at once. Schaffer conceded that with large and complex programs such as the F-35, the Pentagon will have to take some risks. “In some cases, we’ll have to take a large bite,” he said. “In F-35 there is no evolutionary approach. I can’t have half the stealth. Some things are predetermined.”
Other acquisition experts agree that forecasting costs so early in a program (before Milestone A) may be a taller order than what Congress envisioned.
Pentagon analysts have at their disposal mountains of technical, performance and cost data to make somewhat credible estimates, but information and research alone are not enough, wrote Martha A. Roper, senior operations research analyst for the office of the deputy assistant secretary of the Army for cost and economics.
What is needed is an “analysis culture with the policy, procedure and willingness to develop and/or accept cost estimates that are less precise than those developed at Milestone B or Milestone C,” Roper argued in an article published in the January 2010 issue of theDefense Acquisition Review Journal. Without a “large-scale, department-wide culture change within and around the analysis community,” early cost estimating will be tough to accomplish.
Privately, industry experts blame the Pentagon’s procurement woes (it’s not just the F-35) on a shortage of skilled buyers and competent managers overseeing major programs. Regulatory schemes such as WSARA also could use more “teeth” to enforce discipline, some industry insiders contend. “If they enforcedNunn-McCurdy and canceled programs, that would send a message,” one insider said.
Another acquisition expert who offered his views on the condition of anonymity said the problem is the excessive complexity that usually leads to Pentagon buyers chasing rainbows. The technical specs often are so intricate that they create major blind spots for the government, particularly in trying to predict costs, this expert noted. “With high technology systems, the difficulty of accurately estimating production costs is extreme for systems that have not yet undergone a preliminary design review.” Unless the Defense Department starts enforcing cost estimates -- and budgeting accordingly -- in the nascent R&D phase of a program, “defense acquisitions will continue to enter their Milestone B process effectively blind-folded with respect to long-term costs,” he added. “The more complicated and software-intensive we make our weapons systems, the harder it is to predict both their performance and their cost.”