Determinants of geographic concentration of industry in the enlarged EU

As theoretically suggested by Krugman (1991), Krugman and Venables (1995), and Puga and Venables (1997) economic integration would be effective in the formation and evolution of core and periphery structures. New economic geography literature suggests well structured theoretical modelling for the linkages between trade liberalisation and geographical concentration/dispersion of economic activity under a variety of scenarios. Designed models are well suited to make theoretical predictions for the European economic integration case. Since the European cooperation got underway in 1951 among 6 countries, moves toward closer integration resulted in an economic and monetary union of 28 member countries by 2007. Along with several waves of enlargement, increased heterogeneity among the members of the EU raised concerns regarding economic disparities. Thus, the link between European economic integration and spatial distribution of economic activity deserves prominent academic and political interest. However, the empirical literature seems to lag behind the theoretical achievements of the theoretical literature.

There exist very limited attempts to validate/falsify the theoretical framework. To date few studies have empirically analysed the determinants of industrial location in the European Union (for example, Midelfart-Knarvik,et.al, 2001; Combes and Overman, 2004; Head and Mayer, 2006). To the best of the author's knowledge, none of the other studies have extended beyond the 2000s and they have used less comprehensive datasets. This paper aims to make a contribution to the literature in testing the theory with a dataset that is larger and up to date in its cross sectional coverage. This paper addresses the question of how the process of economic integration among the EU countries affected industrial location decisions within the economic region. Existence of a core and the periphery structure of the EU is investigated in its relationship to the dynamic integration process. An econometric estimation model developed by Midelfart-Knarvik, et.al. (2001) is employed to analyse the determinants of industrial location and the evolution of concentration/dispersion patterns. Estimation is undertaken for 20 EU member countries and over the period 1995 - 2009. Including the new members of the EU to the dataset will allow us to analyse the changing patterns of concentration through the enlargement process of the EU.