Measuring regional spillovers in long- and short-run models of total factor productivity, trade, and FDI

This article applies the novel concept of global panel cointegration to analyze the role played by trade and foreign direct investment (FDI) activity in driving regional total factor productivity (TFP). Using West German state-level data for the period 1976-2008, the approach allows us to identify the magnitude of direct trade and FDI effects as well as spatial spillovers from these variables. The author finds that the inclusion of spatial lags significantly improves the fit of the empirical model and allows us to strongly reject the null of no cointegration among the variables in the full spatial specification. For the long-run cointegration equation, the empirical results hint at export- and FDI-led growth. Additionally, outward FDI activity shows to have positive spatial spillover effects among German regions, while the spatial patterns of import and inward FDI activity indicate substitution effects of interregional input-output linkages in favor of international ones over the sample period 1976-2008. In the short run, TFP growth is predominantly affected by changes in exports, inward and outward FDI stocks, where the latter variable also provokes positive spillovers. The author uses alternative spatial weighting matrices based on geographical information as well as interregional goods transport flows to check the robustness of the obtained spillover effects in the long and short run. Here, the results turn out to be similar for the different empirical specifications employed throughout the analysis. Finally, summing over the four variables to get a direct and indirect net effect of internationalization activity, the author finds that the direct effect is always positive, while the indirect net effect is positive in the short run but slightly negative in the long-run equation.