Managing Long Distance and Localized Learning in the Emilia Romagna Life Science Cluster

Abstract

The paper provides some empirical evidence of the effectiveness of the “open innovation” model in the life science cluster of Emilia Romagna (a region of Italy), comparing the network of R&D collaborative activities in public research organizations (PROs) and the network linked to R&D collaborative activities in private firms. By presenting the main results of a field research in the life science sector in Emilia Romagna, we are contributing to the recent debate focused on the crises of the old “closed innovation” model and the rise of the “open innovation” model. Our survey consists of both primary data deriving from face-to-face interviews with researchers and entrepreneurs, and secondary data extracted from the Internet, the PubMed database, and from the European Patent Office. Our work is based on the analysis of a representative sample of 30 research groups in PROs, 2173 scientific articles published by the interviewed scientists, and a representative sample of 78 private firms.

Notes

1. We derived this archive from the firms that have participated in the programmes of research support for innovation launched in Emilia Romagna in recent years (PRIT and PRAI), see Belussi and Di Bernardo (2006).

2. See Appendix, Table A1, for a short description of the sub-sectors of activity of the enterprises inserted in the sample.

3. We excluded firms involved only in assistance services, because they are not firms endowed with innovation and technological capabilities.

4. The district enterprises are located in a small bunch of municipalities, which show high contiguity, like Camposanto, Cavezzo, Concordia, Finale Emiliano, Medolla, Mirandola, S. Felice, S. Possidonio and S. Prospero.

5. See Appendix, Table A2, for a short description of the sub-sectors of activity of the research laboratories inserted in the sample.

6. We choose the criterion of belonging to faculty, for the problem of data aggregation, because many departments are founded by different faculties, and have various names.

7. The software used to analyse relational date is Ucinet 6 (Borgatti et al., 2002). See also Wasserman and Faust (1994).

8. This information is derived from an accurate screening of PubMed, which is to our knowledge the most complete collection of biomedical articles. It allows access to more than 11 million citations of scientific journals and links to the full-text articles. This powerful life science database has been created and daily updated by the National Centre for Biotechnology Information (NCBI), a division of the National Library of Medicine (NLM), belonging to the National Institute of Health (NIH), USA. NCBI is updating the archive about all international research in molecular biology, biochemistry and genetics to provide information to the community of researchers and doctors. N_PUBINT is the number of international publications of the interviewed researcher of the single research laboratory in the sample signed in during the period 2000–2004.

9. We aim to use some proxies to reflect the networking activities of firms and PROs: the number of direct collaborations enacted by each single firm or PRO (firm-to-firm, firm-to-PRO, PRO-to-PRO); and the number of external sources of knowledge for innovation.

10. Innovative and absorptive capacity is represented by two proxies (RD and TF). The information on the amount of research and development investments (RD) is derived from the EPO Database. We decided to integrate this information with the entrepreneurs’ declarations, to avoid the risk of not considering the patents not signed by the firm, but by the individual researcher regularly employed by the firm. Total funds (TF) is the sum of research funds (public and private) plus labour costs.

11. N_PATENTS measures the number of patents owned by the single firm in the sample; the information comes from face-to-face interviews with entrepreneurs, controlled with the EPO. Earlier studies have suggested, and assumed, that patents are a fairly good indicator of the inventive output of the research department of a firm and a measure of the “output” or “success” of R&D (Bound et al., 1982; Hausman et al., 1984; Crépon & Duguet, 1996), although they have only been able to prove a simultaneity in the year-to-year movements of patents and R&D, which appear to be dominated by a contemporaneous relationship (Hall et al., 1986). See also Zimmermann and Schwalbach (1991).

12. See for instance the research findings reported by Dahlander and McKelvey (2005) for the Gothenburg population of biotechnology enterprises where the authors found that 43% of the firms were involved in firm-to-firm relations, measured through formal arrangements.

13. See Appendix, Table A3.