Cluster-based routines and paradigm-bound innovation

The paper explores the limitations of innovation in clusters, proposing that innovation advantages of clusters are contingent upon technological paradigms. Technological paradigms manifest in the heuristics of ‘how to do things’ and ‘how to improve them’ in a domain, embedded in organizational routines. The paper argues that new product development routines can be enacted in clusters, turning into cluster-based routines. Cluster-based routines are efficient in guiding search for rapid solutions within established technological trajectories but become ineffective during paradigm shifts. Consequently, cluster-based routines tend to promote paradigm-bound innovation rather than paradigm-setting innovation. Using an original, product-level database of mobile handsets in China from 2007 to 2016 — a period which witnessed a paradigm transition from feature phones to smartphones — the study presents robust evidence that being in a dominant cluster in Shenzhen has a positive impact on product innovation in the feature phone regime but casts significantly negative effects on paradigm transition and subsequent innovation in the smartphone era. The findings indicate that the temporal and spatial processes of innovation are deeply interwoven.