Industrial clusters in Mexico and Spain

Purpose – The purpose of this paper is to analyse the structure of knowledge networks and the geographical patterns of knowledge networking in mature industrial clusters. To such end, it is assumed that proximity is not really what matters in innovation, but rather the embeddedness of firms into localised networks, enhancing collective learning and knowledge diffusion. Design/methodology/approach – The research is contextualized in the footwear industry and applies the microeconomics of innovation (grounded in the resource based view and social capital approach) and industrial clusters/districts as theoretical frameworks. Methodologically, the paper adopts an exploratory perspective and employs a qualitative approach to conduct a cross‐case analysis of the Leon‐Guanajuato cluster (Mexico) and the Vinalopo cluster (Spain). Findings – Firstly, this paper endorses recent research trends suggesting that knowledge is unevenly and selectively distributed among clustered firms. Secondly, it evidences how internal resources determine a firm's access to valuable repositories of knowledge. Thirdly, key knowledge players are usually involved in extra‐clusters networking, indicating that mere reliance on localized knowledge may result in declining trajectories. Research limitations/implications – Because the case study approach and qualitative methodologies are used, readers are advised not to generalize the findings. The research on the subject matter is offered as a means to substantiate or refute the latest research premises, and provide evidence on the selected clusters. Originality/value – This paper shows how knowledge networks differ depending on geographical specific characteristics and the resources of the main players. Managers‐owners should be conscious that being close to one another is not enough. It should be combined with both solid internal resources and access to repositories of knowledge outside the cluster. Policy makers should prepare customized public programs based on the particular structure of each cluster.