Abstract

Modem (endogenous) growth theory teils us that knowledge spillovers are crucial for the growth of high-income econornies. Against this background the paper discusses how geographically lirnited knowledge diffusion can help to explain clusters of regions with persistently different levels of growth. The paper analyzes this topic in three steps: First, the concept of knowledge spillovers is outlined by discussing (i) the different types of knowledge, (ii) the spatial dimension of knowledge spillovers, and (iii) the geographical mechanisms and structural conditions of knowledge diffusion. This discussion shows that the literature on knowledge spillovers focuses on the hypotheses that such spillovers lead to dynarnic externalities and - in the geographical dimension - to agglomeration effects, both of which constitute path dependency in the econornic growth of regions. Second, the paper analyzes the empirical evidence for these theoretical findings. According to existing empirical work, the evidence suggests that the present studies mainly support the theoretically derived hypotheses. This applies especially with focus on the spatial lirnited character of knowledge spillovers as weil as the importance of knowledge transfer for regional productivity and innovative behavior. Third, the question is asked whether these theoretical and empirical findings give reason for political intervention, and - given such reasons - which policy design should be chosen, if one takes into account the localized knowledge transfer networks as an important source of regional growth activities, but also the increasing econornic inequalities between regions, which result from the agglomeration effects.