Review of DOD's Report on Budgeting for Exchange Rates for Foreign Currency Fluctuations

The Department of Defense (DOD) expends a significant amount of funds overseas, particularly from its Operation and Maintenance (O&M) and Military Personnel (MILPERS) appropriations. As the rate of overseas currencies fluctuates on a daily basis, such fluctuations have an impact on the various expenditures that DOD makes. For budgeting purposes, DOD establishes foreign currency exchange rates to determine its O&M and MILPERS funding needs. During the fiscal year, DOD incurs expenditures at the actual exchange rate, which varies from the budgeted rate. For example, if the dollar depreciates in value, more dollars are needed to pay for goods and services than originally budgeted. Concerned about whether DOD's method for selecting foreign currency rates has produced realistic estimates in its budget submissions, Congress required DOD to consider alternative methods. Specifically, the Ronald W. Reagan National Defense Authorization Act for Fiscal Year 2005 required the Secretary of Defense to submit a report on the foreign currency exchange rate projections used in annual DOD budget presentations. The act required that DOD identify alternative approaches, including the feasibility of using private economic forecasting and approaches used by other federal departments and agencies, for selecting foreign currency exchange rates that would produce more realistic estimates of the amounts required for DOD to accommodate foreign currency exchange rate fluctuations. DOD also was required to discuss the advantages and disadvantages of each approach and to identify the department's preferred approach among the alternatives and provide a rationale for preferring that approach. Finally, the act further required that we review DOD's report, including the basis for the Secretary's conclusions for the preferred approach. DOD submitted its report to Congress on April 15, 2005. In response to the act, we examined (1) the extent to which DOD evaluated alternative approaches for selecting budgeted foreign currency rates--such as private economic forecasting companies or approaches used by other federal departments' and DOD's basis for selecting its preferred rate selection approach and (2) the extent to which DOD's preferred approach for forecasting foreign currency exchange rates would produce a more realistic estimate than its historical approach.

DOD evaluated several alternative approaches for selecting budgeted foreign currency rates and selected an approach that it believed would more accurately and objectively predict exchange rates. The alternative approaches included (1) estimates from a private forecasting company; (2) methods used by other federal departments, such as selecting exchange rates on the basis of past execution data and using multiple exchange rates for individual offices; and (3) various statistical methodologies. DOD selected a statistical method referred to as the centered weighted average, which combines both a long-run average of exchange rates and the most recently observed exchange rates to predict future exchange rates. DOD chose this approach because it was based on historical and current data and could be universally replicated; therefore, it was not dependent on subjective judgment. DOD used this method in developing its fiscal year 2006 budget submission. DOD believed that using a combination of both recent and historical data would be an advantage in more accurately predicting future exchange rates because the methodology allows DOD to weight individual currencies to minimize the impact of their volatility over time. According to DOD, they did not choose the private forecasting alternative because of the lack of visibility over key assumptions used in generating the forecast nor did they choose one of the methods used by the other federal departments because the other departments either did not forecast foreign currency rates or they did not have a uniform procedure for setting departmentwide rates. DOD selected a reasonable approach for forecasting foreign currency rates that can produce a more realistic estimate than its historical approach. Unlike DOD's historical approach, the centered weighted average approach provides a straightforward statistical calculation of historical data that can be easily replicated with no hidden assumptions and is not dependent on subjective judgment. In reaching our conclusion, we compared both the new rates generated using the preferred approach and the rates DOD used to prepare the fiscal year 2004 and 2005 budget submissions based on its historical approach to current actual exchange rates. We found that the new methodology generated rates that more closely reflected actual exchange rates that occurred during the budget year.