Do clusters follow the industry life cycle? : Diversity of cluster evolution in old industrial regions

Purpose – This paper aims to examine whether in old industrial regions, the trajectory of clusters follows that of their corresponding industry or deviates from it and which are the factors that account for cluster evolution. This paper deals with the issue of how established clusters either renew or transform themselves in such regions and how they adapt to changes in their corresponding international industries. Design/methodology/approach – This research paper draws from in-depth case studies on six industrial clusters, takes a longitudinal perspective and uses a multi-level and qualitative analysis. Based on existing literature, the paper suggests and exploratory analytical framework with four alternative scenarios for cluster evolution and three broad factors: cluster knowledge base, social capital at cluster and region-level and public policies. Findings – Clusters do not always follow the life cycle of its dominant industry. The paper clearly shows a diversity of cluster evolution across clusters and even within clusters (at subcluster level). This study suggests that cluster knowledge diversity and heterogeneity allow to broaden the scope of evolutionary trajectories available; the same goes for social capital at cluster and region levels. Research limitations/implications – The main limitation of this paper lies in its qualitative approach that makes its conclusions more suggestive than conclusive. In any case, further research on other Basque clusters may corroborate or question its findings. Originality/value – The paper offers an empirical and longitudinal study on cluster evolution, very much needed to the ongoing theoretical discussion on this issue. So far, there are very few empirical studies on cluster evolution with this perspective. At the same time, it presents a theoretical framework to analyse diversity of cluster evolution in old industrial regions that builds on Menzel and Fornah’s (2010) model.