Patterns of innovation in the Flemish business sector: a multivariate analysis of CIS-3 firm-level data

Abstract

The aim of this paper is to improve our understanding of innovation behavior of firms in the Flemish business sector, including both manufacturing and services. After presenting some new results from the third Community Innovation Survey and a brief international comparison, we use multivariate data analysis to group firms that are markedly similar in terms of innovation activities. Based on data from the CIS-3 for Flanders, it is possible to identify a set of analytical clusters that exhibit clearly distinctive configurations of innovation- related indicators. These clusters can be interpreted as specific “innovation patterns” or “innovation strategies”. Furthermore, we explore the correspondence between the identified innovation patterns and the structural characteristics and performances of the corresponding cluster members, as well as the sources and impacts of their innovations. Finally, we analyze the distribution of the innovation clusters across industries, firm-size classes, and regions. A major finding of this study is that the “sectoral determinism” of innovation (Pavitt) does not hold true in the case of Flemish firms, as multiple innovation strategies seem to co-exist within the same industry. Even though one should be cautious in drawing policy conclusions at this stage of the research, it is clear that policy-makers should fully appreciate the wide diversity of innovation practices among Flemish firms.