Ares

A Defense Technology Blog
See All Posts
  • JSF - Talking Real Money
    Posted by Bill Sweetman 12:00 PM on May 03, 2010

    Pentagon acquisition boss Ashton Carter does not seem too impressed by the way that his predecessors managed the Joint Strike Fighter program. In an Air Force report of a speech in Dayton on April 20, Carter says that "a combination of unforeseen engineering changes and other factors went unacknowledged and virtually unmanaged for two years"

    That is a harsh assessment of what went wrong with JSF, and if anyone can come up with a similarly damning judgment on any other program from a top Pentagon official I'd be interested to hear it. One estimate, Carter is quoted as saying, is that the delay "resulted in $3 billion in additional program costs."


    If that is what Carter is thinking, Lockheed Martin faces an uphill struggle in its campaign to convince the Office of the Secretary of Defense and Congress that the Pentagon's official cost estimates (covered here and here) are not only overstated but are based on selective and inadequate information.

    Aviation Week visited Fort Worth in force last Tuesday. Amy Butler has a report here, but what went on was what Sherlock Holmes called "quite a three-pipe problem" and I needed the weekend and some paper printouts to sort some of it out.


    Essentially, Lockheed Martin is worried that the high cost projections revealed since early March will cause Congress to cut the FY11-15 buys to reduce concurrency. Those are the main years in which annual production ramps up to 200+ jets.

    Lockheed Martin argues that the program is already preparing to build on that scale. Evidence of that is visible in the vast assembly hangar at Fort Worth (which in 33 months in WW2 spat out 3,034 B-24s, each one about the same in empty and loaded weight as an F-35). One large segment of the hangar is occupied by wing assembly fixtures. At the same time, the JSF team is preparing contracts to cover second sources for wings in Italy and center fuselages in Turkey.

    That's why Lockheed Martin is concerned. According to the company, cutting the numbers now would cause unit costs to rise sharply, to the point where customers would further cut and delay orders to stay within fixed budgets: the classic procurement death spiral.

    But this isn't necessary, Lockheed Martin leaders say, because the government's cost estimates are badly wrong on the conservative side. If planned production rates are maintained, the company asserts, the program will deliver F-35As at less than 75 per cent of the cost predicted by the CAPE.

    It's the production cost of the F-35A that is the key area of disagreement between Lockheed Martin and the government's budgeteers in the Capabilities Assessment and Program Evaluation (CAPE) office. Since three-fourths of the currently planned aircraft in the program are F-35As, and most of the research and development money has been spent or committed, the unit cost of the F-35A variant drives the cost of the program.

    And that's about the only thing anyone agrees on.

    The debate over F-35 costs includes a head-exploding range of cost definitions. Costs are variously quoted as base-year (2002 dollars) or then-year.

    Lockheed Martin likes to talk about recurring flyaway cost (URF), the cost of producing a single airplane, where the budget people and Congress prefer the average procurement unit cost (APUC) which is what the government spends to put an airplane on the flight line, including initial spares, training equipment and so on.

    Just to add to the fun, Lockheed Martin also tosses in data on "air system contractor" unit flyaway costs - which in my not always humble opinion should not be called flyaway costs at all, since the JSF is not going to fly very far without an engine.

    Basically, though, the argument is this: 

    The CAPE office and the latest Selected Acquisition report cite an average URF of $63 million for the F-35A, in base-year (FY2002) dollars. Converted to then-year dollars, and including all the APUC elements and the other two variants, this corresponds to the $113 million APUC that was reported to Congress in March. Since this was 57 per cent higher than the original planned cost, it threw the program into a Nunn-McCurdy breach.

    Lockheed Martin argues that the $63 million (FY02 URF for the F-35A) is much too high. The previous SAR, in 2007, pegged this number at $49.5 million, but Lockheed Martin says it can bring the jets in at $46 million.

    Why is the Pentagon estimate too high? Lockheed Martin people point to two reasons. First, last year's Weapon Systems Acquisition Reform Act now obliges the Pentagon to go with an independent non-program estimate if it is higher than the program office's numbers. Second, the CAPE office did not accept any Lockheed Martin numbers and based its estimates on "legacy parametric data" - numbers from the F-22 and F/A-18E/F programs. 

    Indeed, Lockheed Martin points to the fact that its own contracts (not including the engine) for the first, second and third low rate initial production (LRIP) batches were signed at values corresponding to the 2007 SAR, and that the LRIP-4 contract is being negotiated below the line. This, the company says, is evidence that the CAPE numbers are too high.

    The DoD tersely disagrees, noting that the CAPE's full assessment is still under way and saying that it is likely to result in new cost estimates above the latest SAR line:  the $113 million APUC, according to the CAPE office, is at the low end of its current range.

    The two sides cannot both be right.

    Moreover, a detailed look at the SAR report shows that the CAPE assessment was not just a vague application of historic data to the current program, but that the difference between the 2007 SAR and the 2010 price is based on specific factors.

    Overall, CAPE estimated that the total F-35 procurement cost would be about $23.6 billion more than the December 2007 SAR estimated. Two elements accounted for the lion's share of that increase. $9.2 billion was attributed to increases in the cost of building the wing section - which in JSF terms is the wing plus the complex and densely packed body section to which it is attached - plus other elements such as a commonality update.

    $8.6 billion goes to propulsion, which is not reflected in Lockheed Martin contracts. Some of this is influenced by the cost of the lift system for the F-35B:  the Navy is on the hook for nearly one-third of the $8.6 billion even though only 330 STOVL aircraft are planned. That tells you something about how much the Marines' jets are likely to cost.

    So the CAPE estimates - even those made in December 2009 - are about much more than legacy costs. They appear to be about real issues with building the F-35, about manhours and labor rates.

    The next round in this battle will be joined as the CAPE office completes its new estimate - which, as we were told in March, could show APUCs almost 90 per cent higher than Lockheed Martin and P&W contracted for in 2001. And that's for a program that has been fully funded all the way, and where any changes to requirements (such as deferring capabilities to later blocks) have been in the contractor's favor.

    And the LRIP contracts with Lockheed Martin are only part of the answer. They don't include the engine - responsible for a large part of the CAPE increase - and none is fixed-price and none has been completed. (The LRIP-1 jets are due to be signed over in September, the LRIP-2 contract is 65 per cent complete and the LRIP-3 has 70 per cent left to go.) 

    And all cost-plus and fixed-price incentive contracts, one way or another, split the "execution risk" between the contractor and the government, and Lockheed Martin people are not very specific on how this has been done on LRIP 1 through 3 or how it will be done on LRIP-4, except to say that the contracts will not be "firm fixed price" until a multi-year is signed beyond 2015.

    Global warming or not, it's going to be a long, hot summer.

    Tags: ar99, jsf, lockheed martin

Share:
  • Recommend
  • Report Abuse

Comments on Blog Post