The regional distribution of knowledge-intensive services in Europe : a spatial approach

There is a rich debate in the innovation literature about to what extent innovation has become an international (or globalised) phenomenon, or, on the contrary, it maintains its local/regional character. As Koschatzky (2001) notes, given the fact that knowledge is commonly tied to personal capabilities; it has a clear geographical component. In the case of knowledge-intensive services (KIS) most of analyses come to the same conclusion: distance is particularly relevant when knowledge (mainly of a tacit type) is diffused. Starting from this premise, a burgeoning literature on the contribution of those knowledge-intensive business services (KIBS) to regional innovation has emerged. Most of these papers adopt a national perspective, that is, analyse regions in a specific country. On the contrary, comparisons of regional features have been carried out in very few papers: Germany and the UK (Simmie and Strambach, 2006) or Germany and France (Muller and Zenker, 2001) are two examples. The objective of this paper is to take a step further and examine the distribution of knowledge-intensive services (KIS) in the European regions. For so doing we employ the data provided by the Regional Innovation Scoreboard (RIS) 2009. This database provides information on the innovation performance across 194 regions of the European Union and Norway. The methodology employed is known as spatial analysis and evaluates whether there are clusters in the location of KIS in the European regions, which involves three processes. First, to evaluate the existence of spatial autocorrelation by means of global statistics; the Moran's I and the Geary's C. Once verified the existence of positive spatial autocorrelation, it is possible to identify "clusters" of regions with high and low participations of KIS by using a local indicator of spatial autocorrelation (LISA). Finally, employing an econometric model, some potential explanatory factors for the concentration of KIS are examined. The results obtained support the hypothesis that KIS are spatially concentrated and confirm that spatial clusters are different in northern/central and southern/eastern regions. Moreover, a close relationship between location of KIS and regional innovation performance is found.