Emergence of clusters: by chance or by design. The rise of the Oslo Cancer Cluster

The paper is motivated by an interest in how clusters come into being and why they emerge in particular locations. The literature includes two main views on this topic. A first view underlines that new clusters often start out in a particular location more or less by chance, often because some pioneering firms for some reasons choose to locate there in the first place. The second main approach maintains that the emergence of clusters relates to previously developed local capabilities, routines and institutions. The paper proposes a middle ground between these two approaches. It consists first of some pre-existing, regional conditions that must be present if specific industries and clusters are to evolve in some places, i.e. some conditions that exist prior to the emergence of a cluster. Secondly, the paper points at the need for triggering factors that cause clusters to emerge in just some particular places, and first of all factors that stimulate entrepreneurship. The paper will discuss how this theoretical approach stands up when subjected to empirical analyses of the emergence of Oslo Cancer Cluster (OCC). The cluster consists of 60-70 firms and organisations focused on research, development, commercialisation and production of cancer drugs in Oslo. The cluster evolved late 1990s and early 2000 and is still in a very early phase. Mere chance is out of the question to explain the emergence of OCC. The paper demonstrates how high quality and capacity in cancer research in Oslo has evolved over long time and demanded much resource. But nor is OCC ?designed' even if public funds have (partly) financed the building of a knowledge and commercialisation infrastructure in Oslo that are important prerequisites for academic start-ups. These local assets also have to be turned into productive use by entrepreneurs. In the Oslo Cancer Cluster case the initiative by the large Norwegian company Hydro to build new activity in pharmaceuticals, which contributed to combining academic knowledge and industrial know-how, seems to be one important triggering event that is difficult to see in advance and plan. Thus, the analytical framework which includes both pre-existing conditions and triggering events seems as a useful middle road between chance and design in order to explain the emergence and location of regional clusters.