ICT Research networks and regional competitiveness: an analysis of the 7th Framework Program

ABSTRACT

Using Social Network Analysis and the novel concept of bridging centrality this paper investigates the links generated by the research networks created by two of the European Union-funded Programs, the 7th Framework Program (FP7) and the ICT action of the Competitiveness and Innovation Framework Program (CIP) and their impacts on regional competitiveness. Our findings show a positive impact on the competitiveness of both networks. Those regions with higher capacities to establish bridging paths tend to show higher levels of competitiveness. Moreover, our results indicate that these effects differ between research and diffusion networks. Whereas the FP7-related network impacts all the dimensions of competitiveness, the CIP influences competitiveness only through the innovative channel. Our results highlight the importance of facilitating intra- and inter-regional collaboration networks between core and peripheral regions in Europe.

KEYWORDS:

• EU Framework Program
• R&D networks
• regional competitiveness
• Information and Communication and Technologies (ICT)
• Social Network Analysis

Acknowledgements

This research has received funding from the C_innovA - GITPA (Chair for the innovation studies, supported by the GITPA - www.gitpa.es) under grant agreement No CATI-2020-05.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 https://ec.europa.eu/digital-single-market/

2 A vast literature collects the relevance of brokerage role in the generation and diffusion of innovation (Rodan and Galunic 2004; Burt 2009, among others).

3 Formally, the measure is normalized as follows:${\displaystyle \frac{\left(\mathrm{B}{\mathrm{C}}_{\mathrm{i}}-\mathrm{B}{C}_{\mathrm{m}\mathrm{i}\mathrm{n}}\right)}{\left(\mathrm{B}{C}_{\mathrm{m}\mathrm{a}\mathrm{x}}-\mathrm{B}{C}_{\mathrm{m}\mathrm{i}\mathrm{n}}\right)}}$

4 Several alternative measures of innovation and human capital have been checked. The two variables finally considered show low correlations between them and hence, limit potential problems of multicollinearity.

5 The results from the estimations without the variable GDP and robustness analysis are available from authors upon request.