Explaining path-dependence in the evolution of networks. The case of an electronics cluster in Argentina

Abstract

One of the current open areas of cluster research concerns the factors driving the dynamics of local inter-organizational networks. Since embeddedness in local networks is considered to be one important dimension of firms? capacity to innovate, understanding their dynamics can also help to interpret a cluster?s development trajectory. Conceptual works about networks in clusters claim that most networks tend to grow according to a very stable pattern, but very little research has analyzed the micro dynamics responsible for the path dependent evolution of local networks in clusters. To address this limitation we develop a novel conceptual framework where we distinguish between micro dynamics that lead to path dependent changes of local networks, and path destructive micro dynamics. We apply Stochastic Actor-Oriented Models (SOAM) for network change to the case of an electronics cluster in Cordoba (Argentina), and find that the cluster is characterized by a path dependent change of its local network, and argue that this is associated to the coexistence of different (path dependent) micro dynamics, such as: reciprocity, preferential attachment, and social and institutional proximity. In contrast, we do not find that firm level agency and inter-firm heterogeneity contribute to generate path destructive processes of network change. This paper provides evidence that contributes to the refinement of the incipient theories about cluster network dynamics and cluster evolution.