Previous studies have shown that the Department of Defense (DoD) and the military departments have historically underestimated the cost of new weapon systems. Quantifying cost growth is important, but the larger issue is why cost growth occurs. To address that issue, this analysis uses data from Selected Acquisition Reports to examine 35 mature, but not necessarily complete, major defense acquisition programs similar to the type and complexity of those typically managed by the Air Force. The programs are first examined as a complete set, then Air Force and non-Air Force programs are analyzed separately to determine whether the causes of cost growth in the two groups differ. Four major sources of cost growth were identified: (1) errors in estimation and scheduling, (2) decisions made by the government, (3) financial matters, and (4) miscellaneous sources. Total (development plus procurement) cost growth, when measured as simple averages among the program set, is dominated by decisions, which account for more than two-thirds of the growth. Most decisions-related cost growth involves quantity changes (22 percent), requirements growth (13 percent), and schedule changes (9 percent). Cost estimation (10 percent) is the only large contributor in the errors category. Less than 4 percent of the overall cost growth is due to financial and miscellaneous causes. Because decisions involving changes in requirements, quantities, and production schedules dominate cost growth, program managers, service leadership, and Congress should look for ways to reduce changes in these areas.

In recent decades, there have been numerous attempts to rein in the cost growth of U.S. Department of Defense (DoD) acquisition programs. Cost growth is the ratio of the cost estimate reported in a program's final Selected Acquisition Report (SAR) and the cost-estimate baseline reported in a prior SAR issued at a particular milestone. Drawing on prior RAND research, new analyses of completed and ongoing weapon system programs, and data drawn from SARs, this study addresses the following questions: What is the cost growth of DoD weapon systems? What has been the trend of cost growth over the past three decades? To address the magnitude of cost growth, it examines cost growth in completed programs; to evaluate the cost growth trend over time, it provides additional analysis of a selection of ongoing programs. This sample of ongoing programs permits a look at growth trends in the more recent past. Changes in the mix of system types over time and dollar-weighted analysis were also considered because earlier studies have suggested that cost growth varies by program type and the cost of the program. The findings suggest that development cost growth over the past three decades has remained high and without any significant improvement.

This report is one of a series from a RAND Project AIR FORCE project, "The Cost of Future Military Aircraft: Historical Cost Estimating Relationships and Cost Reduction Initiatives." The purpose of the project is to improve the tools used to estimate the costs of future weapon systems. It focuses on how recent technical, management, and government policy changes affect cost. This report focuses on the accuracy of cost estimates. For our analysis, we used a very specific sample of Selected Acquisition Report (SAR) data, namely only programs that are complete or are nearly so. The analysis indicates a systematic bias toward underestimating the costs and substantial uncertainty in estimating the final cost of a weapon system. In contrast to the previous literature, the cost growth was higher than previously observed. We also found few correlations with cost growth, but observed that programs with longer duration had greater cost growth and electronics programs tended to have lower cost growth. Although there were some differences in the mean cost growth factors among the military departments, the differences were not statistically significant. While newer programs appear to have lower cost growth, this trend appears to be due to factors other than acquisition policies.