Abstract

Among the advantages of belonging to successful local systems like clusters, the regional economic literature has stressed the critical role played by localized knowledge. For some time, scholars have been arguing that knowledge spreads unevenly among local actors, rather than pervasively and widely, but, its drivers, underlying social structure, and evolution over time remain poorly understood. Particularly, on the one hand, heterogeneity of firms and the way they are perceived are fundamental features to understand evolutionary patterns of clustered firms acting in a world of uncertainty and imperfect information; on the other hand, different ties among the same set of actors simultaneously diffuse specific knowledge. This PhD thesis aims to go deeper in this debate investigating a framework to study local development in relation to architectures and dynamics of local systems and focussing on a network perspective; particularly, stressing the role of both individual heterogeneity and relational multiplicity; testing the efficacy of the identified framework within the same industry, but, with two different and original databases; implementing two different methodologies of social network analysis for the study of knowledge network structures and dynamics (Exponential Random Graph Models and Stochastic Actors Oriented Models); and identifying a few policy implications. To organically achieve these aims, the thesis aims to answer the following general research questions: What is the state of the art of knowledge networks within local systems? To what extent do multiple ties as different relational sets through which knowledge diffuses impact on the local exchange of knowledge? To what extent does status as the perceived relative qualities of a firm in a given market or organizational field affect knowledge network evolution over time? To answer these questions, the first chapter “Knowledge Networks within Local Systems. Their Structures and Dynamics” provides a literature review on knowledge network structures and dynamics within local systems and it offers an original explanation of local systems evolution with a knowledge network perspective. The second chapter “Complementary Inter-Firm Relations of Multiple Knowledge Networks in Industrial Clusters: Evidence from a Growing Wine Cluster in Italy” shows that different kinds of relationships positively impact on the spread of technical knowledge, but they are different in magnitude and they follow complementary patterns rather than substitutive ones. The third chapter “Status and the Assortative Dynamics of Knowledge Networks in Industrial Clusters: Evidence from a Successful Wine Cluster in Italy” shows the presence of an assortative network change, where high-status firms are more likely to interact with other high-status firms but not with low-status firms (and vice-versa). Finally, the last part concludes with a summary of the main findings and it offers a few possible policy implications. Also, the main limitations of the study as well as a few future possible extensions are discussed.