Spatial clustering and nonlinearities in the location of multinational firms

We propose a semiparametric geoadditive negative binomial model of industrial location which allows to simultaneously address some important methodological issues, such as spatial clustering and nonlinearities, which have been only partly addressed in previous studies. We apply this model to analyze location determinants of inward greenfield investments occurred over the 2003-2007 period in 249 European regions. The inclusion of a geoadditive component (a smooth spatial trend surface) allows to control for omitted variables which induce spatial clustering, and suggests that such unobserved factors may be related to regional policies towards foreign investors Allowing for nonlinearities reveals, in line with theoretical predictions, that the positive effect of agglomeration economies fades as the density of economic activities reaches some limit value.