Nonlinearities in FDI allocations : why good institutions matter more with rising FDI locational concentrations

Abstract

This paper extends an established finding: that institutions are important for foreign direct investment (FDI). Our results show that institutions are both more and less important than previous empirical results suggest. This is because the concentration of FDI in a location matters. Theoretically, if institutions serve a risk-mitigating role, then rising locational concentration of FDI compromises the risk diversification function that multiple locations for FDI provides. This can be offset by high-quality institutions. The implication is that the impact of institutions on FDI will be enhanced with rising FDI concentration. Empirically, we examine the locations of outward FDI from South Africa from 1996-2019, confirming the presence of a strong association between FDI and an institutions-FDI concentration interaction term. The result is robust to many alternative means of measuring institutions and to a number of alternative means of representing the implied nonlinearity in estimation. The inference is that for any location that is intent on attracting strong concentrations of FDI inflows, the general precept that sound institutions are important in attracting FDI flows is enhanced, both in terms of the strength of institutional improvement, and the breadth of institutions that require attention.