Knowledge clusters and knowledge hubs: designing epistemic landscapes for development

Abstract

Purpose With globalization and knowledge‐based production, firms may cooperate on a global scale, outsource parts of their administrative or productive units and negate location altogether. The extremely low transaction costs of data, information and knowledge seem to invalidate the theory of agglomeration and the spatial clustering of firms, going back to the classical work by Alfred Weber and Alfred Marshall, who emphasized the microeconomic benefits of industrial collocation. This paper aims to argue against this view and show why the growth of knowledge societies will rather increase than decrease the relevance of location by creating knowledge clusters and knowledge hubs, due to sharing of tacit knowledge and research and development outputs. Designing epistemic landscapes of knowledge clusters and hubs is, therefore, proposed as a viable development policy. Design/methodology/approach Secondary data from cluster research as well as the authors' own survey data, mainly on Vietnam, Malaysia and Singapore, are used to show and measure clustering of knowledge‐producing organizations. Findings The paper shows that sharing of tacit knowledge is a crucial variable to explain why clustering of knowledge‐intensive industries takes place despite the development of information technology. The importance of knowledge management for development is highlighted. Practical implications Designing epistemic landscapes of knowledge clusters and hubs is, therefore, proposed as a viable development policy. Originality/value The paper argues against the mainstream of industrial clustering theory and proposes a KM‐based research strategy. The discussion of the difference between knowledge clusters and knowledge hubs enables the use of a new and consistent evaluation of a K4D (knowledge for development) strategy.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 90 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.