Agglomeration and spillovers in a simultaneous framework

Abstract

Previous studies have emphasized reduced transaction costs, knowledge spillovers, diverse intermediate goods suppliers, and deep labor pools as possible drivers of firm agglomeration. Among these candidates, knowledge spillovers lie at the heart of discussions in the literature. If knowledge spillovers are so important, it follows that firms tend to locate in proximity to capitalize on the knowledge stock in neighboring firms. Therefore, agglomeration and spillovers reinforce each other. However, most existing empirical studies ignore the potential endogenous relationship between the two. Moreover, few studies have even attempted to measure knowledge spillovers. In this paper, I address both issues by developing and estimating a simultaneous equation model of agglomeration and knowledge spillovers. The results suggest that traditional single equation models do indeed suffer endogeneity and that the accuracy of the estimation can be substantially improved by employing a simultaneous system.