A Spatial Analysis of Innovation in Europe

The purpose of this paper is two-fold. Firstly, it surveys different works related to the analysis of Innovation and the way it spreads locally. We will see that theoretical predictions and empirical facts are rather contradictory. This will lead us, secondly, to shedding light towards the concentration of innovative activities in the European Union, at a more precise level of analysis: the NUTS2 nomenclature. We contradict the usually accepted empirical fact that innovative activities tend to be less and less concentrated, thus supporting the theory of knowledge spillovers, which is probably due to the important changes the European Union went through in terms of trade structure. This claim relies on a rather simple but well-founded analysis. Patent applications to the European Patent Office will be used as a proxy for the level of Innovation. These patent applications have a geographical component that will be exploited to determine whether spatial autocorrelation of innovative activities is present. After showing the importance of considering spatial autocorrelation in an analysis of Innovation, we will propose an empirical methodology to assess how it spreads. Regarding this, no results are presented, as it goes beyond the scope of the paper.