Evaluation of commercial economic forecasts for use in local government budgeting

Using a decision-making under uncertainty framework, this paper proposes an approach to evaluating commercial macroeconomic forecasts as used by local governments in forecasting revenues. The approach is applied to a case study of Kansas City. Forecasts of GNP and CPI provided by DRI, Chase and Wharton Econometrics are evaluated along with simple time series extrapolations as inputs to the revenue forecasting process. The results indicate that the variance of forecasts errors is minimized by either extrapolating exogenous variables using time series methods, or relying directly on time series extrapolation methods to forecast revenues.