1 Knowledge Spillovers and Economic Growth – Regional Growth Differentials across Europe by Marjolein C.J. Caniëls Summary of my PhD dissertation, presented at the BABES colloquium on March 5th, 2002 The aim of the dissertation is to implement concepts from geography into a technology gap model and thereby broaden the understanding of knowledge spillovers (across space). More precisely, the impact of local knowledge spillovers on regional growth is ana lysed. The book is divided in three parts. The first part deals with the description of the literature on diffusion of knowledge and provides a methodological framework. Chapters II and III are devoted to these issues. The second part focuses on analysing the influence of local knowledge spillovers on regional gaps. Chapters IV to VI are devoted to setting up a model that gives an impression of these effects. The implications of the model are explored by means of simulation techniques. The third part deals with the current situation in Europe. Chapters VII and VIII use empirical analyses to demonstrate regional gaps and differences in economic and innovative activity across regions. Summary and conclusions Chapter II gives a (selective) survey of the literature on diffusion of knowledge. The chapter reviews the perspective on knowledge spillovers in various strands in the literature. Whereas traditional neo-classical growth models assume knowledge to be immediately everywhere available, cumulative causation theories adopt the idea that knowledge will stay in the geographical place it originates and will not spread at all. ‘In between’ are imperfect diffusion theories, which assume knowledge to spread gradually over time and space. From these contributions, we obtain elements for a regional model on knowledge spillovers. One element is taken from technology gap theory, which assumes that technological distance plays an essential part in determining the growth rates of countries (the technology gap literature mainly considers countries, although some regional models exist). A second element is taken from the geographical literature on agglomeration effects and growth poles. In this literature, geographical distance plays a very important role in that increasing returns to location are assumed. The geographical agglomeration of industrial activities leads to a set of advantages that induces further agglomeration. Among these advantages are the availability of skilled labour and intermediate goods suppliers, as well as the easy transmission and discussion of new ideas. In this book, the focus lays on this last factor. Knowledge spillovers are assumed to take place more easily across smaller geographical distances. In Chapter III, evolutionary theory is presented as an apt way to model regional knowledge spillovers. The merits of an evolutionary approach lie in the acknowledgement of the heterogeneity of agents. We assume that reality can be best approximated by a model that takes into account that regions differ in several respects. Regions are assumed to differ in