Efficiency criteria for intermediaries involved in the innovation process

Public or private intermediaries are proliferating all over the world, especially in France; they try to facilitate innovation by linking scientific and productive areas, and often benefit from public financial support. Any attempt to measure their efficiency faces the problem of choosing adequate criteria, since we still lack a well–based model representing the innovation process. Two models can be used, but each of them is partly irrelevant. The first one belongs to the allocation–optimizing models family: intermediation can be considered as a substitute for an efficient technological information market and as an instrument for redistributing this information and for exploiting externalities and economies of scale. However, this first model is not able to represent the interactive process that links knowledge accummulation with innovation strategy because one of its main assumptions is that technological knowledge is exogeneously produced and has only to be transferred by information channels (the 'canal–lock model'). The second model, based on 'evolutionism', seems to be more relevant since it leads us to assess intermediation efficiency in terms of its overall performance in handling the innovation process rather than in terms of minimizing information costs. However, it proposes no clear method of assessing the collective efficiency of the intermediaries, especially when their influence is focused on small and medium–sized businesses' innovation capabilities. So, after having examined the difficulties of finding general criteria for evaluating the efficiency of intermediation, one can perhaps focus on their particular role in each national production and innovation system.