Structural Change and Economic Dynamics 20 (2009) 163–182 Contents lists available at ScienceDirect Structural Change and Economic Dynamics journal homepage: www.elsevier.com/locate/sced Geographic concentration of innovative activities in Germany Dirk Fornahl a,∗ , Thomas Brenner b a BAW Institute for Regional Economic Research, Wilhelm-Herbst-Str. 5, 28359 Bremen, Germany b Philipps University Marburg, Chair of Economic Geography and Location Research, Germany article info Article history: Received November 2008 Received in revised form May 2009 Accepted May 2009 Available online 15 May 2009 JEL classification: R11 R12 O30 D83 Keywords: Spatial concentration Innovation Technology Concentration indices abstract The geographic concentration of industries has attracted much attention in recent eco- nomic and geographic literature. One mechanism employed to explain the emergence and comparative advantage of industrial agglomerations is based on the relationship between industrial agglomeration and local knowledge production and diffusion, and the result- ing innovation activities. This paper analyses this relationship by identifying geographic concentrations of innovation activities and examining different causes for the emergence of these concentrations. The paper applies different concentration measures to patent data for German regions. We analyse 43 technological fields separately to identify which of these technologies tend to cluster in geographic space. The results are discussed in light of the- oretical predictions of why specific technological fields concentrate while others do not. These explanations include the concentration of industrial activities, the role of dominant firms, dependence on scientific knowledge, and local interactions. © 2009 Elsevier B.V. All rights reserved. 1. Introduction The geographic concentration of industries has been repeatedly studied in the literature (Ellison and Glaeser, 1997; Braunerhjelm and Carlsson, 1999; Lafourcade and Mion, 2005; Brenner, 2005; Sternberg and Litzenberger, 2004; Alecke et al., 2006). Ellison and Glaeser (1997), for example, identified that in general, industries show differ- ent degrees of geographic clustering, but in the majority of industries an agglomeration of firms in a few regions can be observed. This paper takes up the core issue of geographic concen- tration but is focusing not on industrial but on innovative activities. Studies examining such spatial distributions of innovation activities are rare. Such a lack of research on ∗ Corresponding author. Tel.: +421 206 99 30; fax: +421 206 99 99. E-mail addresses: d.fornahl@baw-bremen.de (D. Fornahl), thomas.brenner@staff.uni-marburg.de (T. Brenner). the existence of concentrations of innovative activities is particularly interesting, since such activities are a central element linked to the development of industrial clusters which are seen as a core driver of regional development. In the literature it is argued that firms that are located in industrial agglomerations are more innovative than firms outside such regions (Audretsch, 1998). This would imply that innovation activities follow the concentration pattern of industries but are more pronounced. At the same time, it is argued that innovation activities cause firms’ success and growth (Smolny and Schneeweis, 1999). As a consequence, industrial activity should follow the spatial distribution of innovations. In addition, many arguments for the emer- gence of geographic concentrations of industries are based on mechanisms, such as knowledge spillovers, accumu- lation of technological knowledge, and cooperation, that function rather on a technological than an industrial level. We can conclude that innovation activities should also be geographically concentrated, probably causing geo- 0954-349X/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.strueco.2009.05.001