Knowledge and the diversity of innovation systems: a comparative analysis of European regions

The main goal of this paper is to shed some light on European regional diversity in terms of knowledge accumulation and socio-economic performances. Dynamic links between knowledge, innovation and performance are complex to address because they take place in different contexts, involving heterogeneous agents interacting through different institutions. Studies on national systems of innovation (Edquist, 1997) stressed the role of the institutional context in these dynamics and identify various configurations associated with these national systems. This conceptual framework, used at the regional level, leads to the identification of regional systems of innovation (Cooke, 2001) and thus underlines the limits of a regional scoreboard only based on high-tech indicators as it is usually proposed. This paper constitutes a first attempt to propose a more exhaustive effort in characterizing the diversity of "regional knowledge an innovation systems" within Europe. The study is performed through data analysis using the conceptual framework of "social systems of innovation and production" (SSIP) proposed by Amable, Barré and Boyer (1997). A Social System of Innovation and Production can be defined as a coherent combination of different components referring to Science-technology-industry (STI) configurations articulated with financial system, labour relations, education and training and economic performances. This framework can be adapted at the regional level by identifying specific arrangements of each part of the system even if the concept of system is questionable at this level. The analysis is performed combining data from three sources (Eurostat, the Cambridge Econometrics database and OST (Observatoire des Sciences et des Techniques)) over a sample of NUTS-II european regions and using multivariate data analysis (principal component analysis, hierarchical anova). Putting together the SSIP and local economic performances allows defining different regional configurations in order to identify regional trajectories and patterns of articulation between knowledge dynamics and performance. Our hypothesis is that regional growth in not a problem of best practice but of coherent knowledge combination: institutional differences may lead similar (or different) STI structures to different (respectively same) performances.