A Spatial Panel Simultaneous-Equations Model of Business Growth, Migration Behavior, Local Public Services and Household Income in Appalachia

In this paper we develop a spatial panel simultaneous-equations model of business growth, migration behavior, local public services and median household income in a partial lag-adjustment growth-equilibrium framework and utilizing a one-way error component model for the disturbances. This model is an extension of the “jobs follow people or people follow jobs” literature and it improved previous models in the growth-equilibrium tradition by: (1) explicitly modeling local government and regional income in the growth process; (2) explicitly modeling gross in-migration and gross out-migration separately in order to spell out the differential effects, which used to be glossed over under net population change in previous studies; (3) explicitly incorporating both spatially lagged dependent variables and spatially lagged error terms to account for spatial spillover effects in the data set; and (4) extending and generalizing the modeling and estimation of simultaneous systems of spatially interrelated cross sectional equations into a panel data setting. To estimate the model, we develop a five-step new estimation strategy by generalizing the Generalized Spatial Three-Stage Least Squares (GS3SLS) approach outlined in Kelejian and Prucha (2004) into a panel data setting. The empirical implementation of the model uses county-level data from the 418 Appalachian counties for 1980-2000. Generally, the results from these model estimations are consistent with the theoretical expectations and empirical findings in the equilibrium growth literature and provide support to the basic hypotheses of this study. First, the estimates show the existence of feedback simultaneities among the endogenous variables of the model. Second, the results also show the existence of conditional convergence with respect to the respective endogenous variable of each equation of the model and the speed of adjustment parameters are generally comparable to those in literature. Third, the results from the parameter estimation of the model indicate the existence of spatial autoregressive lag effects and spatial cross-regressive lag effects with respect to the endogenous variables of the model. One of the key conclusions is that sector specific policies should be integrated and harmonized in order to give the desirable outcome. Besides, regionally focusing resources for development policy may yield greater returns than treating all locations the same.