Knowledge network dynamics in clusters: past performance and absorptive capacity

Purpose - The purpose of this paper is to explore the degree to which absorptive capacity and previous innovative performance affect network dynamics, specifically in the creation or destruction of inter-business relationships. Design/methodology/approach - The empirical study has drawn on the data collected in an industrial cluster located in the Valencian Community in Spain. This population of firms allowed us to test various roles played by network endogenous forces, absorptive capacity and former performance in the creation and dissolution of inter-organisational linkages. The authors followed an evolutionary approach and applied network analysis techniques. Findings - Empirical evidences suggest that absorptive capacity and previous innovative performance are predictors of inter-business relationships. Absorptive capacity affects the emergence of linkages in the technological network, due to the tacit nature of technological knowledge. On the other hand, previous innovative performance is an indicator to identify leaders companies. These companies tend to increase the reception of request for advice from local firms. Moreover, prestigious companies tend to be more selective in choosing collaborators. Practical implications - This study helps researchers and managers better understand network dynamics. The generation of inter-organisational relationships in clusters can be determined by cognitive proximity and prestigious status in the network. These evidences are relevant in a mature cluster where knowledge is asymmetrically distributed. Originality/value - Over the last few decades, studies on industrial clusters have evolved from the canonical standpoint. Nowadays, researchers usually take a more sophisticated and richer view of this reality, mainly as a consequence of the inclusion the proximity concept, intra-cluster heterogeneity and advanced analysis of overlapping between networks and territory. Thus, the authors try to add some empirical evidence along these lines.