ABSTRACT
Tools for Smart Specialisation Strategy (RIS3) development to assess at which territorial level innovation is concentrated and how innovation systems compare across regions and sectors remain scarce. The triple helix indicator is a data-driven tool that can serve in RIS3 development to assess synergy in innovation systems based on readily available data sets. For Germany, we find strong decentralization with innovation system synergy concentrated at regional and local levels. Innovation systems in manufacturing are less regionalized than in knowledge-intensive services. An East–West divide persists. RIS3 policy-making should account for decentralization by building capacity for region-specific RIS3 development at lower governance levels.
ACKNOWLEDGEMENTS
We thank Michael Wyrwich for the preparation of data on patent applications in Germany.
DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.
Notes
1. For a detailed description, see Fritsch and Wyrwich (Citation2020a) and Sanders et al. (Citation2020).
2. Firms are classified as world market leaders if they are either among the top three suppliers of their product worldwide or the leading supplier in the European market.
3. The system of innovation approach represents a general framework for an analysis of innovation processes that accounts for many factors highlighted by innovation theories such as endogenous innovation (Romer, Citation1990) or innovative milieu (Ratti et al., Citation1997). For details see, Chaminade et al. (Citation2018) and Asheim et al. (Citation2019).
4. For an explanation and instruction, including a routine, see https://leydesdorff.net/software/th4/.
5. NACE is the abbreviation of Nomenclature Statistique des Activités Économiques dans la Communauté Européenne which represents the industry standard classification system used in the European Union.
6. NUTS stands for Nomenclature of Territorial Units for Statistics.
7. For 869,340 records, a classification by their NUTS region was missing in the data. For 790,889 of these records, we could add the respective NUTS code by matching the address information based on Eurostat’s NUTS/ZIP code correspondence tables (Eurostat, Citation2020). The remaining 78,451 (approximately 2%) records were excluded from the data set (cf. the section on empirical data).
8. Other empirical analyses confirm a leading role of Saxony among the former East German states with regard to innovation behaviour (EFI Commission, Citation2020).
9. Information on the number of regional patent applications was harvested from the OECD RegPat database (Maraut et al., Citation2008). Patents are assigned to the region where the inventor has her or his residence to avoid a bias from counting any patent invented at any establishment in a given firm to count towards the region of the firm’s headquarter. If a patent has more than one inventor, the count is divided by the number of inventors, and each inventor is assigned his/her share of that patent. Data on regional private sector R&D employment are from the German employment statistics, which covers all employees subject to compulsory social insurance contributions (for details, see Spengler, Citation2008). R&D employees are defined as those with tertiary degrees working as engineers or natural scientists.
10. We did not find a statistically significant correlation between synergy and the patent ratio at lower levels of governance. A possible reason for this finding is that at the NUTS-3 level, for example, it is unlikely that inventors live and work in the same district given the prevalence of commuting into cities for work.
11. The level of governmental regions (NUTS-2) is expected to allow only limited cross-regional comparisons of synergy generation because some federal states (NUTS-1) are not subdivided in governmental regions (NUTS-2), as discussed in the results and discussion section. These findings both reflect and refine the results obtained by Leydesdorff and Fritsch (Citation2006) who found that differences in the synergy generated in formerly Eastern and Western were significant in the early 2000s at the NUTS-1 level, but not at NUTS-2.
12. The 96 German planning regions (Raumordnungsregionen) represent functionally integrated spatial units comprising several districts (NUTS-3 regions).