The geography of knowledge flows

Research on the geography of innovation in the past twenty years or so to 2010 has largely revolved around two basic questions. Does knowledge tend to flow more easily across spatially proximate agents than across agents localized far apart in space? To what extent may localized knowledge flows also be characterized as spillovers? The purpose of this chapter is to provide a review of the empirical literature that has attempted to assess the existence and spatial extent of knowledge flows. From a purely methodological perspective, two distinct approaches can be identified in the literature. A first group of studies has adopted a knowledge production function approach, interpreting a positive correlation between regional innovative inputs and outputs as evidence of localized knowledge spillovers. A second group of studies has instead focused on the much more complex task of providing a direct measurement of knowledge flows. Although this group comprises a more heterogeneous set of empirical approaches compared to the studies based on the knowledge production function, a few common features can nonetheless be identified. First, the search for a direct measurement of knowledge flows implies the shifting of the unit of analysis from the level of regions or clusters, to the level of individual innovations or to the level of individual researchers. Second, most of the studies included in this group attempt to disentangle the role that different transmission channels may play in the spatial diffusion of knowledge.