Near optimal regional investment using a simultaneous equation economic model

Abstract

A region seeking development objectives should attempt to schedule public investment optimally between human resources and physical infrastructure. A regional growth model and a suitable objective function are presented. It is desired to calculate both the optimal public investment for nominal parameter values and the sensitivity of the optimal solution with respect to changes in parameter values. However, the system is not in state form, but is represented as a set of nonlinear simultaneous equations. For this case the necessary conditions for optimality are difficult to solve. The approach suggested here is to choose a nominal parameter value of economic significance for which the model may be put into state form. Then the existing optimally sensitive control algorithm may be used with some modifications to obtain a near optimal investment strategy for the simultaneous equation system when the parameter deviates from the nominal value. In general, the results indicate an emphasis on education and training expenditures. Sensitivity results indicate that the relative emphasis on long vs short run considerations (i.e. public investment vs consumption) depends on the growth rate parameters in the model.