Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/123955 
Year of Publication: 
2013
Series/Report no.: 
53rd Congress of the European Regional Science Association: "Regional Integration: Europe, the Mediterranean and the World Economy", 27-31 August 2013, Palermo, Italy
Publisher: 
European Regional Science Association (ERSA), Louvain-la-Neuve
Abstract: 
Innovation arises from multilevel interactions. Many factors at various levels of aggregation chip in. Not only individual characteristics, resources and capabilities of firms, but also framework conditions within which they operate matter for their success in the innovation process. Admittedly this has been recognized for a long time in the various ?contextual? perspectives on geography of innovation, including the concepts of innovative (or creative) milieu, technology districts, learning regions, collective learning or localized learning, which have largely converged into the study of regional innovation systems over the recent years. However, quantitative empirical research in this tradition remains scarce and continues to use models that are severely restricted to handle multilevel hypotheses like these. The aim of this paper is to develop the multilevel approach to spatial analysis of innovation. Using a highly representative micro dataset of small and medium size firms derived from pooling multiple waves of Community Innovation Survey in the Czech Republic, we quantitatively assess the hypothesis that the regional milieu directly affects firm?s innovativeness in terms of the intensity on innovative sales and that the framework effects differ for different kinds of firms. The main contribution of this paper is four-fold. First and foremost, we investigate the potential endogeneity of the predictors vis-a-vis the regional random effects, i.e. the unobserved heterogeneity across regions. Second, we noticeably extended the set of regional indicators that feed into the factor analysis of regional framework conditions, including data on scientific articles, patents, university research, public research institutes, etc. Three, we delve more deeply into the issue of what are the most relevant regional units of the analysis. Finally, the empirical analysis is based on an extensive micro dataset, which provides far more representative evidence at a detailed regional level, as compared to previous studies on this topic. Since these surveys use random sampling, the overlap of respondents is fairly limited, which hinders the creation of longitudinal data, but on the other hand if merged together they provide a highly representative regional sample. The results indicate that the quality of regional innovation milieu is a highly significant predictor of the innovativeness of firms and that the regional differences mediate effects of the firm-level predictors, namely size, foreign ownership and internal innovative efforts of the firms.
Subjects: 
innovation
regional milieu
multilevel modeling
cross-level effects
Czech Republic
Document Type: 
Conference Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.