of mission and cost-effectiveness, and the late availability exposes the country to unaddressed threats and increases the probability of early system obsolescence. As would be expected, cancellation of programs reported in the SARS is rare. The Roland and Copperhead weapons systems are two recent examples and even in these cases a minimum number of orphan units will be produced. The same resistance to cancelling production programs extends to new starts. Again, DOD clearly prefers stretchouts and decreased funding to cancellation. of the 18 new start programs approved by the Deputy Secretary of Defense as part of the FY 1983 Program Objectives Memorandum and budgeting process, 11 have not been funded to the level proposed by DOD (through authorizations), and only one was increased. Not one of these programs has been cancelled. Obviously, they will proceed at a slower pace than that proposed in the original service submissions. The extent of the stretchout and suboptimum production rate problem goes even deeper. In developing their proposals, the services have their own affordability concerns. OSD officials believe that the services do not necessarily request the most cost-efficient programs, but rather that they are driven to propose what will sell. Thus, even at the outset of a program, there are inherent inefficiencies which are later compounded by overwhelming pressures to launch new start efforts and to eventually force them, along with existing programs, under a relatively inflexible total DOD budget line. This process is believed to be an important contributor to program instability (see Issue OSD 23). Conclusions Much has been written about the so-called procurement "bow wave" and the recurring production schedule stretchouts, unit production reductions, unit cost increases, and the compounding effects of price escalation. Recent increases in defense spending have permitted improved schedules and have moved rates of production closer to economic levels. However, it is asserted in the literature and by those interviewed in this survey that overall defense systems are still being produced at suboptimum rates, and unit volumes are continually set on the basis of what can be squeezed into the budget, rather than what is economical or what is the optimum required for the intended military mission. Part of this costly problem is caused by the continuing competition for available funds from new programs being added to the acquisition pipeline. If the 39 systems currently in full-scale development or production had had to pass tougher affordability criteria and continuing cost scrutiny at an early stage, fewer of them would have been subjected to the recurring stretchouts and unit volume reductions that have occured. Although demonstration of affordability in a rudimentary sense is required as part of the new start and milestone review process, the requirement only applies to the specific systems under consideration. There is no requirement to demonstrate that adding new systems to the acquisition pipeline will not cause production stretchouts or volume reductions in other systems already in the procurement chain. In the view of the OSD Task Force, the new start decision process is inherently deficient since it provides virtually no meaningful control. Given the pressures to keep programs alive once they pass the new start hurdle, and the noted inability to cancel programs downstream, it is absolutely essential to stiffen the entry requirements for major new system starts. DOD must make a more demanding and realistic review of long-range affordability in the new start decision process. Given its interest in profitability, the private sector cannot afford to continue marginal programs. Because DOD operates in a different environment, it has escaped many of the severe penalties that would be paid by an industrial concern if it routinely stretched programs and operated at inefficient production rates. However, if DOD is to fulfill its mission at a cost that is affordable to the U.S. economy, it must introduce more of the kind of discipline that characterizes private sector operations. DOD policies are a reflection of the system as it has long operated. The policies did not create the system. The OSD Task Force observed that there are few pressures, if any, on DOD top management to face up to this affordability issue. Rather than retard the flow of new programs, the current DOD management has indicated it intends to do a better job of cutting out entire programs if they do not meet objectives. However, there is no evidence that this can be achieved, even though the objective is sound. Top DOD management has the power to institute changes in the system, but there are pressures that mitigate against placing tough controls on the system because: Such controls would be unpopular with the Political pressures are great, given Congressional constituent interests in various Political appointee tenure is short, and fighting to improve the system would require a major commitment of scarce resources and time. Similarly, the services have little incentive to sponsor controls because: They compete with the other services and fear resources. They already are hard pressed and feel they need all the resources they have. The Chairman of the Joint Chiefs of Staff and the service Chiefs of Staff, who are represented on the DRB, also are not likely to limit new start programs. The Chairman has no power to set priorities among the services, and the service Chiefs appear to support the other services, lest their service be subject to restraints as well. The DRB, as it deals with new starts, simply concurs with service recommendations. The facts support this. In the last two years, DRB has recommended for approval one more new start than the services requested. No program was rejected. OMB has indicated that the total projected costs of acquiring all weapons systems in the pipeline (from new start through production) are approximately 230 percent of the funds that are likely to be available for such purposes. Although it is important to overprogram somewhat in this regard to take into account the likelihood that a few weapons systems will indeed be cancelled at some point along the line, the OSD Task Force believes that this percentage is much too high. We believe that the percentage should be 140 percent at the time of the new start decision and that this percentage should progressively decrease through the later milestone decisions. Strict adherence to such limitations will ensure that a more limited number of programs make it through the process to the production phase. Recommendations The lack of effective controls over new starts results in each program's being given less than the required resources which, in turn, increases system costs and delays introduction of capabilities. As a result, the system may end up being obsolete and may be built in insufficient numbers to meet the mission. OSD 21-1: DOD should develop and aggressively implement a modification to the program review and approval cycle, which would: O Subject every new start to in-depth affordability review over the projected life cycle of the program. Review new start affordability in relation to all other new starts being proposed and programs already in the system extending out at least ten years and show the effects on existing programs. Base affordability evaluations on conservative estimates of DOD total budget resources and price escalation for the ten-year planning period. Allow for less program attrition and establish specific ceiling percentages for overprogramming at each acquisition phase and hold rigidly to these ceilings. The OSD Task Force suggests that the percentage ceiling be 140 percent at the new start stage, and that this percentage be progressively decreased at later milestone decision points. While this appears conservative, the record shows that cost overruns justify such conservatism. Modify the SAR reporting system (as proposed in This recommendation would inject into DOD spending decision discipline analogous to the capital spending budgets used by industry to control tooling and facilities expenditures. Such capital budgets are developed on the basis of conservative projections of sales and profits and corporate financial objectives. Capital budgets force prioritization of new projects and, therefore, provide a realistic means to assess affordability. Under the proposed system, the percentage limits on overprogramming would serve to prevent inefficient funding of programs and associated stretchouts. The changes would tend to force DOD management to decide program priorities and to extend those requirements to the services. Savings and Impact Analysis Major program new starts are described in DOD Directive 5000.1 as those programs for which the DOD component estimates costs to exceed $200 million (FY 1980 dollars) in RDT&E funds or $1 billion (FY 1980 dollars) in production funds, or both. The average number of new starts approved in 1981 and 1982 (for FY 1983 and 1984, respectively) was 14. As discussed above, OMB has indicated that the total projected cost of all programs is approximately 230 percent of the funds that are likely to be available. We propose that DOD limit such overprogramming to 140 percent as of the time of the new start decision, with progressively lower percentages for later milestone decision points. A reduction from the current average over-programming of 230 percent to a level of 140 percent would result in a 40 percent reduction in RTD&E expenditures. Assuming there are 14 new start programs a year and that each has an average total RDT&E cost of $200 million ($2.8 billion total) over a five-year period, an average savings over a typical five-year period of 40 percent, or approximately $1.1 billion, might be obtained. On a moving average basis, this would provide annual savings of $1.1 billion for all programs in the system by the fifth year. Applying these criteria to the FY 1983 new starts funded in the FY 1983 Congressional Authorizations Bill, the RDT&E new start budget in the first. year would be reduced from $557.3 million to $334.4, a $222.9 million or 40 percent reduction. These savings could be used in any of three ways: (a) to reduce defense spending, (b) to fund the remaining new starts to optimized levels, or (c) to fund programs in the procurement phases to more efficient production rates and volumes. Assuming 10 percent inflation over the five-year, phased-in implementation period, savings would develop as shown on the following page. |