Visualising the operating behaviour of SMEs in sector and cluster: Evidence from the West Midlands

Abstract

Cluster policy in the UK, pursued by the Regional Development Agencies (RDAs), has readily adopted a simplistic definition based upon industrial sectors and location quotients. Evidence drawn from a study of the operating behaviour of SMEs belonging to two traditional manufacturing industries within the West Midlands—automotive components and clothing—provides a critique of this approach. Whilst the automotive components industry has been designated part of a key, high priority cluster, the clothing industry has not. Using case studies from both industries, this paper shows firms both within and outside RDA cluster definitions display a remarkably similar range of behavioural characteristics. Yet, based on weakly defined cluster policy, one industry enjoys considerably more policy support than the other. The paper begins to question the logic of RDA cluster policy and to ask whether a more sophisticated and locally sympathetic manner of visualising clusters and business behaviour rather than an emphasis on employment numbers would have a greater impact for policy.

Keywords:

• Cluster policy
• regional development agency
• automotive & clothing sectors

Acknowledgements

The authors gratefully acknowledge Gordon MacLeod for comments on an earlier draft and colleagues at the University of Oxford for permitting the data to be used in this paper. An earlier version of this paper was presented to the European Regional Science Association Annual Conference, Dortmund, 27–31 August 2002.

Notes

1. The paper draws on data from a project undertaken by Oxford and Coventry Universities under the EU's Fifth Framework Programme (Battersby et al., 2001). The project was concerned with regional adjustment strategies to technological and economic change in the context of European integration (RASTEI). See Tully & Berkeley (2002) for a fuller description of the project methodology.

2. All the businesses interviewed could be easily grouped (with varying degrees) with at least one other of the firms in their industry into three ‘types’.

3. The names have been changed to protect the anonymity of the firms.

4. Maria Corte-Real, Postgraduate Researcher, University of Cambridge, mjrc2@hermes. cam.ac.uk