This chapter provides a brief overview of the types of costs involvedin producing the JSF for those unfamiliar with how DoD and its con-tractors determine and allocate costs to defense programs. In par-ticular, the objective is to describe indirect costs and then to describehow these costs might vary across locations and how that variancemight affect JSF FACO costs. Indirect costs are an important consid-eration in the final cost of the JSF because these costs normallyconstitute more than half the costs incurred at any manufacturinglocation for DoD systems.
Because of the lack of a true commercial marketplace for military air-craft, where other comparable products can be used for price com-parisons, the production price paid by DoD for the JSF will be anegotiation between DoD contract administration personnel andthose of Lockheed Martin. These negotiations will be largely basedon forecast costs derived by government cost and price analysts aswell as those of Lockheed Martin. Early in production, estimatorswill use the actual costs of other aircraft previously built and adjustedfor JSF differences and/or the actual costs of producing the JSF SDDaircraftagain, adjusted for differences in content between devel-opment and production aircraft. DoD and its contractors for mostmilitary-unique products follow this process. Over the years, acomplex categorization of the various types of costs involving mili-tary production has evolved so that the government and its contrac-
62 FACO Alternatives for the Joint Strike Fighter
tors can use this categorization to expedite the negotiations on finalprices for military systems.1
The first major division of these costs is into two categories calleddirect and indirect costs. Direct costs are those that can be asso-ciated directly with the development and/or production of a specificDoD system, including such items as direct labor (engineering, tool-ing, manufacturing, etc.) and direct materials (which may range fromunprocessed basic materials to complete assemblies). The viewpointtaken in this report is from the prime contractors perspective, buteach intermediate manufacturer goes through the same categoriza-tion when it converts material into final products, which may besubassemblies for use in the next higher level of manufacturing.Direct costs are accounted for and tracked in the contractorsaccounting system to a specific system being developed or manufac-tured for DoD.
Indirect costs are those not easily identifiable with a specific systemor those that, because they represent relatively small sums, may notjustify separate tracking to a specific system. These indirect costs aregenerally further subdivided into two major categories: overheadcosts and general and administrative (G&A) expenses. Overheadcosts are those that are indirectly related to an area of developmentor production and include such items as electrical power; deprecia-tion; heating, ventilation, and air conditioning (HVAC); computerservices; and employee fringe benefits. Corporations often furthersubdivide overhead into areas identifiable to a functionengineer-ing overhead, manufacturing overhead, etc. G&A expenses coveractivities related more to the overall functioning of the corporation,with little identifiable relationship to specific development or manu-facturing operations. Included in G&A are the salaries of corporateofficers, the cost of operating a corporate headquarters, independentresearch and development, and bid and proposal costs. This discus-sion implies that the division between the various cost categorieswould be the same with all DoD contractors; but in reality this cate-gorization varies somewhat among DoD contractors, between sites
______________ 1Two excellent sources of information on the categorization of DoD costs are Indirect-Cost Management Guide: Navigating the Sea of Overhead by the Defense SystemsManagement College (DSMC) and the Defense Acquisition Deskbook at http://web1.deskbook.osd.mil.
Indirect Costs 63
of the same contractor, and even at the same location over timesodirect comparisons of these costs is difficult at best.
If one company were developing and manufacturing only one DoDsystem at one location, all costs could be considered direct costsbecause any costs incurred by that company would be directlyrelated to that system. Thus, whether a cost is direct or indirectwould be a theoretical discussion because all costs could be billed tothat one program. Such a case is rare because most contractorsdevelop and produce several DoD systems at more than one locationor may produce both military and commercial systems or may pro-duce items for Foreign Military Sales. One might ask whether itwould be easier for DoD to just pay the costs annually at specificlocations where only DoD-related activity took place, rather thangoing through a complex cost determination and allocation process.Many arguments could be offered for keeping the current allocationprocess, but the strongest is that it provides the best visibility into theindividual costs of various DoD programs so that senior decision-makers and Congress have the most accurate cost informationrelated to DoD programs.
Because indirect costs are not readily identifiable with a specificproduct, some system of allocating these costs among DoD pro-grams, as well as to commercial programs, is required. This is animportant issue because indirect costs are normally larger than thedirect costs incurred at a manufacturing plant. The first step indeveloping an allocation scheme is to organize the indirect costs intocost pools, which relate to the direct function supported. For ex-ample, all indirect costs forecast to be incurred by a company relatedto its manufacturing function would be categorized into a manufac-turing cost pool. Next, some basis of allocation among programs isdetermined. The most commonly used allocation method is onerelated to a direct expense, such as direct labor hours. The totaloverhead costs divided by the basis of allocation produces an over-head rate. Thus, if a company forecasts it will incur $100 millionannually in the manufacturing overhead account and also forecaststhat it will use 1 million manufacturing hours, the manufacturingoverhead rate would be $100 per hour. If a specific program wereprojected to use 250,000 manufacturing hours, it would be billed for$25 million of overhead costs.
64 FACO Alternatives for the Joint Strike Fighter
Indirect costs have a fixed and variable component. The fixed part ofthese costs does not change over the short run. For example, depre-ciation on a plant would not change whether the production activitywere relatively high or low because depreciation is normally basedon the original cost and the passage of time. The variable part ofoverhead costs changes by activity level. An example of this might beemployee fringe benefits or electrical power consumption. Withhigh production output, increased fringe benefit costs would resultfrom larger numbers of employees working on the program, orgreater usage of machines and test equipment would result in higherelectrical consumption. By their nature, indirect costs tend to have alarger fixed component compared with the variable portion. Thesecosts vary by location, not only in the magnitude of the total, but alsothe fixed and variable components.
Annually, DoD auditors and contract administrators go through aprocess to develop FPRAs, which involve detailed analyses of directand indirect costs incurred in the past and forecast to be incurred inthe future. FPRAs are expressed in terms of a dollars-per-allocationbasisagain, normally direct labor hours. These rates are oftentermed wrap rates because an hour of labor used to produce a DoDsystem would include the wages of the worker, the overhead costs ofsupporting the worker, and an element of the G&A expense. Thetotal number of units in the basis of allocation (direct labor hours orsome other factor) at a facility is often termed the business base toreflect the level of activity at a location. Generally, several ratesreflect the different direct cost activities; therefore, a location mighthave an engineering rate, a tooling rate, a manufacturing rate, etc.
Because the indirect costs have a large fixed component, one way tominimize the costs of a program is to produce the greatest number ofgoods possible at a specific location (maximize its business base),thus spreading the overhead costs over as many systems or units aspossible. In the defense industry today, production rates are lowcompared with capacity. DoD is continually pressuring defense con-tractors to lower their overhead costs to keep the FPRAs under con-trol, in some cases by closing facilities. Few, if any, locations in theUnited States producing DoD systems are anywhere near maximumproduction capacity. (One might take this argument to its logicalextreme and ask whether DoD should require its defense contractorsto produce all of their work at a single site. However, diseconomies
Indirect Costs 65
of scale may arise. These occur when the sheer size of the facilityand amount of work performed there become too large to beproperly managed.)
EFFECT OF INDIRECT COSTS ON JSF FACO LOCATION
A prime objective of this study is to determine what the costs to DoDwould be if the business base resulting from FACO activities werespread to more than one location. At the first level of analysis, itseems that the rates at Fort Worth would increase because the samefixed costs would be spread over a smaller business base if some JSFbusiness base were moved elsewhere. This would increase the costof the JSFs assembled at Fort Worth, as well as increase the costs ofother DoD programs at the facility. Theoretically, if some portion ofFACO were moved out of the Fort Worth plant, the FPRAs would risebecause the indirect costs would have to be covered by less directactivity. However, if the FACO business base were moved to anotherLockheed Martin location where DoD programs were being con-ducted, the FPRAs at that location(s) should be reduced due to theaddition of the JSF FACO business base. Thus, the DoD programsmanufactured at either (or both) Palmdale or Marietta could bedecreased by such a movement of business base. This reductionwould be at least partially offset by added overhead costs at Palmdaleor Marietta for such costs as new facilities, equipment, and environ-mental compliance costs. New site overhead rates would be calcu-lated as follows:
existing overhead costs + additional FACO overhead costsexisting business base + FACO business base.
One of the key questions in the analysis of creation of more than oneFACO site is whether moving some or all of the FACO business basefrom the Lockheed Martin Fort Worth plant to another DoD produc-tion location would change the total cost of JSF FACO and the totalcosts of other DoD programs. In other words, would the negativeeffect on the rates and DoD programs at Fort Worth by reducingFACO activity be offset by the positive effect on the rates and pro-grams by increasing activity at another Lockheed Martin location? As
66 FACO Alternatives for the Joint Strike Fighter
the analysis in Chapter Ten will show, moving the FACO activity fromFort Worth has a tangible effect on total overhead costs.
RAND has investigated the overhead rates at the four potential loca-tions, gathering as much specific data as possible. Specifics are notreleasable because overhead rates are proprietary to the companies.The link between FACO activities and overhead rates at the differentsites has been incorporated into the model. Thus, the RAND analysisestimates the total impact on all DoD programs (including JSF) atFort Worth and alternative sites by incorporating the puts and takesof this business base at the sites by adjusting forecast rates at eachlocation under several production scenarios. This analysis providesa more accurate assessment of JSF FACO options in terms of the totalDoD costs involved (see Chapter Ten for details).