To Cluster or Not to Cluster? Three Views on Cluster-based ED

Cluster-based economic development policy has gained wide acclaim in recent years as interest has grown in utilizing “cluster theory” in economic development. Some have found this idea to be appealing while others have been a bit less receptive of its ideas. Three recently released papers address cluster-based economic development from different angles. The first article explores options for utilizing cluster-based economic development in less advantaged regions. The second examines into evaluation techniques of cluster policy, while the third addresses concerns about “cluster theory” in general. Each is summarized below.

Creating Smart Systems: A Guide to Cluster Strategies in Less Favoured Regions

In Creating Smart Systems: A Guide to Cluster Strategies in Less Favoured Regions, released in April 2002 and written by Stuart Rosenfeld, strategies are presented to apply cluster-based economic development policies in lesser advantaged regions. One of the basic questions that is often asked in regional economic development is, “Why do some regions have the ability to develop and support innovative and competitive cluster and others do not?” Rosenfeld identifies and details five barriers that less favored regions face when attempting to develop cluster economies — weak infrastructure; lack of access to capital, technology, and innovation; regional insularity and isolation; low education levels; and a low skilled workforce. Rosenfeld presents six regional actions that regions can invoke in order to better apply cluster based economic development polices:

• Understand and benchmark regional economies;
• Engage employers and institutions;
• Organize and deliver services;
• Build a specialized work force;
• Allocate and attract resources and investments; and
• Stimulate innovation and entrepreneurship.

Warning that there is no single recipe to follow that will meet the needs of all clusters and regions, Rosenfeld simply presents a menu of actions that can be applied as required. The appropriate action depends on various factors such as geography, stage of development, resource constraints, special societal needs, cluster priorities, market imperfections, and local preferences.

Creating Smart Systems: A Guide to Cluster Strategies in Less Favoured Regions is available at: http://europa.eu.int/comm/regional_policy/innovation/pdf/guide_rosenfeld_final.pdf

The Challenge of Evaluating Cluster Behavior in Economic Development Policy

While Rosenfeld proposes actions for the use of cluster-based economic development in less favored regions, some question the evaluation methods of these policies. In The Challenge of Evaluating Cluster Behavior in Economic Development Policy released May 2002, Philip Raines addresses the need to redefine evaluation techniques for cluster-based economic development policies. Raines identifies the uniqueness of cluster-based policy and explains the differences with traditional spatial development.

While measures such as employment, output and jobs are important, they do not capture how processes are affected through policy, Raines asserts. Linkages between policy-induced changes in firm behavior and regional impacts go unevaluated.

Raines emphasizes limitations exist in applying traditional evaluation techniques to cluster-based policy. In spatial development evaluation, the link between output and result tends to be assumed; for cluster policy, however, this link is critical to the analysis. The main objective of cluster policy is to maximize the benefits and value that cluster agents can gain from networking. Crucial to this, and to the evaluation of policy, is how networking influences business performance. Because of the idea that cluster-based activity is a collective action, it presents new sets of problems for spatial development evaluation.

Raines asserts that more emphasis needs to be placed on understanding the links between individual business behavior, collective action and wider economic impacts in evaluation. Raines identifies five challenges that exist in evaluating cluster-based policy and proposes options that can be explored for addressing these issues. They are:

• Identifying the clusters whose policy impacts are to be evaluated;
• Defining the behavior to be measured in each cluster;
• Linking changes in cluster behavior with changes in individual agent and cluster performance;
• Establishing that changes in cluster performance result in improved economic impacts within the cluster and regional economy; and
• Providing a standardized base on which the performance of highly-differentiated clusters can be compared and decision on the cost effectiveness of their policy support can be made.

The Challenge of Evaluating Cluster Behavior in Economic Development Policy is available at: http://www.regional-studies-assoc.ac.uk/events/Raines.pdf

Deconstructing Clusters: Chaotic Concept or Policy Panacea?

Questions concerning not only evaluation techniques but also the actual constructs of cluster-based economic development exist. Ron Martin and Peter Sunley provide a unique critique of cluster theory in Deconstructing Clusters: Chaotic Concept or Policy Panacea? which is forthcoming in the Journal of Economic Geography.

While cluster theory has been popularized recently — largely because of wide readings of Harvard professor Michael Porter — Martin and Sunley stress the idea of specialized industrial localization is nothing new, dating back to 1890. The authors argue that instead of being a theory or model to be rigorously tested, the cluster concept has been accepted in economic development practice mainly on faith.

The two main constructs of Porter’s work on clusters involve linkages and geographical proximity — Martin and Sunley take issue with each of these. Regarding linkages, they assert that Porter’s concept provides no mechanism or definition for the strength of linkages necessary for the cluster concept. With respect to geographical proximity, the authors again take exception, saying the concept is never defined with precision. The Porter concept, they argue, accepts that clusters can be found at any level of spatial aggregation. With this definition, clusters can occur in rural or urban areas, large and small economies and even encompassing networks of countries. This sort of ambiguous geographical definition allows an unlimited scope in the characterization and application of the concept, Martin and Sunley contend.

They write, “The existence of clusters, appears then, in part at least, to be in the eye of the beholder or should we say, creator.” The authors then argue that this ambiguity of the cluster concept has lent to its convenience, there appear to be no rules or guidelines involved in its use and applicability.

The authors conclude that a concept so flexible as that given for clusters cannot provide a complete model on how spatial concentration is related to regional economic growth. Economic geographers and regional analysts have long understood that a simple association between high-growth industries and geographical concentration does not establish causality between the concentration and economic success. The practical case for clustering remains in its formative years and repeatedly errs in moving from particular relationships to general causality and applicability.

Deconstructing Clusters: Chaotic Concept or Policy Panacea can be downloaded from:

http://www.ges.bham.ac.uk/NewEconomyIntranet/Seminar2/Papers/Deconstructing_Clusters Jun.pdf

To learn more about Clusters...

Cluster-based ED will be analyzed further next week at SSTI's 6th Annual Conference, Building Tech-based Economic Development: From Policy to Practice. Dr. Jerry Paytas, Associate Director of the Center for Economic Development at Carnegie Mellon University, will conduct a breakout session next Wednesday at 10:30.

Dr. Paul Sommers, Senior Research Fellow with the Northwest Policy Center at the University of Washington, will lead a session later that same day that can help frame what approach a community should take in developing its tech-based economic development strategy. The session is entitled Why Do High Tech Firms Locate Where They Do?

More information on the conference is available at: http://www.ssti.org/conference02.htm [expired]