In this paper, we analyse the disaggregated pattern of German software firm locations. We utilise a comprehensive data set of three million street level geocoded firm obser-vations in Germany as well as detailed geographic data from crowdsourced and official sources. We show that this data can be used to detect information on firm location de-terminants which are superimposed when aggregated spatial units are analysed. In a preliminary exploratory spatial data analysis we find evidence that software firms locate mainly in urban environments but prefer sub-urban locations - especially when rents are high and the local population distribution is polycentric. These findings are confirmed in a subsequent regression analysis. We find strong and significant effects for both absolute and relative population density, urban centrality and firm to population ratio, while controlling for infrastructure, quality of life, science & education, and geocoding quality.

Kinne, Jan (2017), Standortmustermodellierung der deutschen Softwarebranche, in: Kolbe, Thomas H.; Bill, Ralf; Donaubauer, Andreas Wichmann, Berlin, Offenbach, 24-28.


Kinne, Jan


Location Analysis, software industry, geographic data, microgeography, location factors