With or Without Clusters: Facilitating Innovation through a Differentiated and Combined Network Approach

Abstract

Over the past decades, economic and innovation policy across Europe moved in the direction of creating regional clusters of related firms and institutions. Creating clusters through public policy is risky, complex and costly, however. Moreover, it is not necessary to rely on clusters to stimulate innovation. A differentiated and combined network approach to enhancing innovation and stimulating economic growth may be more efficient and effective, especially though not exclusively in regions lacking clusters. The challenge of such a policy is to mitigate the bottlenecks associated with “global pipeline”, “local buzz” and “stand alone” strategies used by innovative firms and to combine these strategies with a view to their complementarity in terms of knowledge effects. Private and semi-public brokers will be key in the evolving policy, as timely organizational change is crucial for continued innovation, while brokers also need to mitigate governance problems. This requires region-specific knowledge in terms of sectors, life cycles, institutional and socio-cultural factors, and yields spatially differentiated and differentiating adjustment strategies. The role of public policy is to assist in recruiting, provide start-up funding and monitor brokers. With this, policy moves towards a decentralized, process-based, region-specific, spatially diverging and multi-level system of innovation that is geared towards the evolving innovation strategies of firms.

Notes

Please note that we separate the cluster from the network concept, defining clusters as “geographical concentrations of firms involved in the same of similar activities, which may, but need not specialize, subcontract and co-operate with one another”, and networks as “a set of strategic, purposeful, preferential, sometimes repetitive and usually co-operative interactions between actors, which may, but need not operate in close vicinity” (Visser & Boschma, 2004, p. 801). There are different explanations for clustering in the sense of a long-term and endogenous process of geographical concentration (see Appendix 1). Clusters—in the sense of geographical concentrations, are out there, or they are not, which is radically different from the position of some “cluster enthusiasts” who put “the promotional cart before the analytic horse”, believing that “the detailed structure and workings of a cluster will become obvious soon enough once we begin to think about an activity in cluster terms” (Martin & Sunley, 2003).

Another source of opportunity costs is the lack of balance, within and across regions, of an all too strict focus on clusters. Not only does this increase the dependence of regions on the performance of cluster(s), implying enhanced vulnerability of the regional economy towards external shocks in specific markets (oral communication with Ulrich Witt, 1 December 2006; for a recent example see Van Dooren, 2006; for a more extensive discussion of this problem, see Frenken et al., 2007), but it also contributes to growing inequality within and between regions and in terms of innovation, economic growth and prosperity, which are a threat to stability in certain societies, as we have seen in developing nations (Todaro, 2000; Visser, 1996) and currently in the Russian Federation and China.

The Utrecht, Gooi & Eemland region has got a few specializations, but this does not mean that cluster processes (in the sense of Brenner, 2004) are on their way. The four identified clusters in Media, Business Services, IT Consultancy and Chemical subsectors are embryonic in at least two ways: (a) low concentration and specialization ratios, oscillating between 1.25 and 3 (which is low compared with Italian clusters; see Capello & Faggian, 2005); and (b) weak statistical relations with several innovation indicators (Atzema & Visser, 2005b, pp. 40–51).

The choice between, to combine or to exclude local, global or stand-alone strategies is likely to be path- and context-dependent, including differences across sectors, the phase of an industry or technology in the life cycle, institutional settings and socio-cultural factors.

These investments entail: a) informing other actors about one's “mental model” (North, 1994) and knowledge routines; b) explaining how and why one has arrived at this model; c) understanding alternative models and explanations.