Please use this identifier to cite or link to this item:
Rosenfeld, Martin T.W.
Year of Publication:
45th Congress of the European Regional Science Association: "Land Use and Water Management in a Sustainable Network Society", 23-27 August 2005, Amsterdam, The Netherlands
Throughout the literature in regional economics, most authors agree that spatially concentrated industrial activities are important for regional economic growth. Agglomeration economies, which may occur in the context of spatial concentration and "clusters", may lead to lower costs of production and may reduce transaction costs of all kind, e. g. information costs, including the costs for R&D activities. There is much less agreement on (and: knowledge about) the empirical identification of existing spatially concentrated economic activities in the real world. For the last decade, the discussion on spatial concentration has been dominated by praising the benefits of economic "clusters". Many case studies on regions with economic "clusters" are describing how the mechanisms of these specific "clusters" work. But there have been only a few attempts, so far, to look at a greater region or even at an economy as a whole, in order to find out – with empirical data which allow to compare between the sub-regions – for all sub-regions what their specific „competences" in the field of spatially concentrated industrial activities are and whether there is a typical spatial pattern of concentrated industrial activities. The proposed paper is presenting the empirical results of a study on spatially concentrated industries in the Eastern part of Germany, where the current regional policy scheme, which de facto follows the "watering can principle”, is under debate. In the first part of the paper, the possible dimensions and the "ingredients” of economic "clusters" and industrial agglomerations will be discussed. One important dimension is the existence of intra-regional value-added chains for certain products. But, at the moment, it is not possible to identify such value-added chains by using existing empirical data at the more aggregated levels of an economy. Therefore, our study and the paper are concentrating on just three main dimensions of economic "clusters” and industrial agglomerations: (1.) Particular high spatial concentrations of certain industries are identified (based on employment data by NACE 2 digit level) for all East German counties. (2.) The existence of business networks (and their main characteristics, e. g. industry classification) in the East German regions is discovered by a broad exploration of internet sources and an inquiry among regional development agencies. (3.) Spatially concentrated innovation activities are recorded for each region by using data on patent applications (by IPC classes). It will be discussed in the paper what is implicated with these three dimensions and their operationalization with empirical data. Finally, the findings from the first three steps of research are synthesized for showing how sectoral concentration, business networking and innovation competencies in the individual regions are overlapping. The findings reveal that spatially concentrated industries are mainly located in and around the largest East German cities in Saxony and Thuringia, and in Berlin and its hinterland. In contrast, a number of less densely populated, rural or former industrialized areas in the northern and central parts of East Germany have no or only some elements of spatially concentrated industries. The presented method for identifying spatially concentrated industries could be applied to other regions and economies for bringing more light into the debate on economic "clusters". With regard to regional policy in East Germany, one may conclude from our findings that the present use of the "watering can principle” had not been able to stimulate economic agglomerations in economically weak peripheral regions. It could be a better strategy to support the existing "clusters” and industrial agglomerations.
Appears in Collections:
Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.