Does Proximity Matter for Knowledge Transfer from Public Institutes and Universities to Firms

Abstract

'National Innovation Systems' theories are built upon the assumption that linkages among organisations matter to innovation. Specifically, proximity is a crucial factor in most of the explanations of regional innovation systems. Yet several thing, such as the rapid growth of the internet and email, suggest that the role of proximity could be breaking down, particularly for large firms with the financial resources to seek out knowledge anywhere in the world.However, the need to access tacit knowledge in rapidly evolving science-based technologies could counter the centrifugal features of modern communication technologies. This study examines the effect of proximity on the sourcing of knowledge by firms from suppliers, customers, joint ventures, competitors (via reverse engineering) and publicly-funded research organisations (PROs). The focus is on PROs, since they are an essential component of National Innovation Systems. Relevant data for up to 615 firms are available from the 1993 PACE survey of Europe's largest industrial firms. Descriptive results show that compared to four other information sources proximity effects are greatest for PROs. The factors that influence the importance of proximity to the use of information from PROs are explored through an ordered logit model. The dependent variable is the relative importance of domestic and foreign PROs. The independent variables include firm size, activity in foreign markets, R&D intensity, a proxy for codified knowledge, and two proxies for the quantity and quality of the scientific base of a country. The ordered logit model results show that proximity effects decline with an increase in the firm's R&D expenditures, the importance attached to basic research results in publications, and activity in the North American market, but increase with the quality and availability of outputs from domestic PROs.

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 — 50 citations exist upstream.

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