| Abstract: | Five versions of a regional economic forecasting and simulation model are implemented to evaluate the forecasting accuracy and significance for impact analysis of alternative regional labor market closures. The five versions correspond to the following specifications: downward-sloped labor demand and upward-sloped labor supply, vertical labor demand and upward-sloped labor supply, an input-output version, and two general equilibrium configurations of labor demand and supply. It is found that the estimated impacts of an exogenous employment stimulus differ greatly across the model versions. Also, post-sample forecasts for 1981-1988 are run for the fifty states plus Washington D.C. with each model version to test their relative forecast accuracy. The forecast comparison shows that the general equilibrium version that specifies inelastic supply is inferior to the other versions for short-term forecasts of wage rates and long-term employment forecasts. For both short- and long-run population forecasts, the versions with completely immobile labor are more accurate than those with completely mobile labor. However, versions that specify an upward-sloped labor supply (partial labor supply adjustment) are the most accurate. |
