Clustering dynamics and the location of high-tech firms

The location of productive activities and the emergence of clustering dynamics has been an important research topic since the early works of Weber (1929) and Marshall (1920 and 1921). This thesis aims at relating the processes of firms' location decision and the development of high-tech clusters within an encompassing theoretical and empirical framework. The thesis shows the empirical relevance of the clustering of high-tech sectors and highlights the importance of the issue through the construction and use of an original database on the location of high-tech establishments and employment (at two different geographical levels) in four major industrialised countries. It also contains a critical review of a number of different streams of theoretical and empirical literature which are directly connected, or which have been explicitly put in connection by the author, with the topic of study. In the thesis we develop a composite modelling framework for analysing firms' location decisions and the growth of high-tech clusters, and we empirically test a number of crucial hypotheses in order to draw some guidelines for economic policy. The models presented in the theoretical chapter derive from two different streams of literature. The first derives from the analysis of population ecology, the second from the theory of innovation diffusion. These modelling frameworks have stressed the existence of a critical mass and a maximum dimension of the cluster and their effects on the early and late phases of development within the "life cycle" of a cluster. They also highlighted the role of rank, stock, order and epidemics effects in the location decision of an individual firm which has to decide whether to locate into a developing cluster. The empirical evidence presented in the thesis has focused on the crucial elements of the location process by verifying the empirical relevance of different locational factors, has stressed the relative importance of agglomeration versus scale economies in determining the industrial specialisation of an area, and has measured the competitive effects which arise between the development of different clusters and the synergistic effects which are generated within the cluster. Finally the thesis presents empirical evidence which shows that local competition and industrial specialisation are the key elements for the success of an industrial cluster. A final chapter extracts some crucial policy conclusions on the role of entry versus growth policies, on the different development path that an industrial cluster may follow depending on the excludability condition, presents an original taxonomy of specific policies, applies some of these findings to a brief survey of the phenomenon of science parks and finally produces a series of guidelines for policy makers. The conclusion summarises the results obtained in the thesis and present a brief agenda for future research.