blog-main.jpg

Inside the Barrel

Is Joint Strike Fighter Cost Growth Different than Cost Growth for Other DoD Acquisition Programs?

Is Joint Strike Fighter Cost Growth Different than Cost Growth for Other DoD Acquisition Programs?

Many reports have been published on the cost growth experienced across the Joint Strike Fighter (JSF) program and indeed organizations like the Government Accountability Office (GAO) and RAND regularly publish formal studies highlighting cost growth that occurs in DoD acquisition programs. Often, these studies and reports discuss the causes and magnitude of cost growth; however, I was interested in a different question. Is there a statistically significant difference among the total JSF cost growth and major DoD acquisition programs? RAND published a study that determined the total cost growth for major DoD acquisition programs, at Milestone-B (early in their life cycle) averaged 46% (adjusting for quantity changes) or 65% (unadjusted for quantity changes) and within their data sample, they also highlighted that aircrafts have an average cost growth of 35%. My analysis, which was published in the Winter 2013 Edition of ICEAA World, revealed the JSF program experienced a growth of 69% over the original baseline estimate. In general, as analyzed by RAND, this is certainly more than the historical averages for aircrafts and major acquisition programs. However, is the difference statistically significant? To answer this question, I first graphed the mean cost growth factors for each acquisition commodity (i.e. aircraft, helicopters, missiles, etc.) using a Normal Quantile Plot to assess whether the distribution of the data was normally distributed, and to calculate the average cost growth and standard deviation for all commodities. The next step was to conduct a Student's T-test using the hypothesized cost growth of 69% for the JSF (previously calculated) and indeed the p-value of the Student's T-test was 0.0056. Therefore with statistical rigor I concluded that total cost growth for the JSF program is statistically different from the broader population of DoD major acquisition programs. To add further analytical rigor, I also conducted a Wilcoxon signed-rank test (a non-parametric statistical hypothesis test) to compare the results of this test to the Student's T-test. The result was a p-value of 0.0098, which is far less than a significance level of 0.05 and therefore my original conclusion was corroborated with this test. Given the nature of DoD funding and rising costs across all of government, this analysis demonstrates that not only is cost growth within the JSF program more than other acquisition programs but it is statistically significant and therefore should be scrutinized even further to determine how future cost growth can be curtailed. For more information on Cask's cost estimating and analysis capability please click here.

Written byBrad Ellis