Technological Profiles And Economic Performance In The Italian Manufacturing Sector∗

Abstract

This paper applies multivariate analysis to a rich data set on technological innovation, in order to identify typical patterns of technological change. To single out the dominant forms of technological behaviour, twelve variables have been selected approximating firms' technological sources and innovation results. Multivariate analysis was applied to identify six main clusters or dominant technological profiles. Each cluster was then interpreted on the basis of a larger set of variables concerning firms' technological and economic performance. The results confirm recent emphasis placed by scholars of technical change on variety of bchaviour in technical change and industrial organization and, more specifically, Pavitt's seminal attempt at a classification of this variety. However, our clusters are somewhat more numerous than Pavitt's, thus showing a more complex combination of technological input and output. Moreover, representative industries of each cluster are less easily recognized as a less marked sectoral characterization of clusters emerges. The latter result seems to imply that variety in technological change is also shown by the existence of different technological trajectories and strategies within the same sector. We suggest that the diffuse nature of recent technical change has given firms greater freedom of strategic choice among different technological strategies.