![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
We rely on a novel database of Spanish author-inventors to explore the relationship between the past patenting experience of academic authors and the scientific impact (citations received and journal prestige) of scientific articles published during 2003–2008 in journals listed in SCOPUS. We also study how such a relationship is affected by differences across academic affiliations, distinguishing between public universities and different types of non-university public research organisations. Our econometric estimations show that scientific impact is positively associated with having authors with past patenting experience as inventors at the European Patent Office. Exceptions are the articles of authors affiliated to new independent public research centres, not tied to the civil service model and oriented to do research that is both excellent and use-inspired. These are also on average the most cited articles.
Acknowledgements
We would like to thank Francesco Lissoni, Natalia Zinovyeva, Luis Sanz-Menendez, Koen Jonkers, Félix de Moya, Katrin Hussinger, Bart Van Looy and three anonymous referees for very useful comments and suggestions. We are also very grateful to Félix de Moya and Elena Corera from the SCImago Group for their help with the SCOPUS data. This work has also benefited greatly from comments of participants at the ESF-APE-INV “Scientists and Inventors” workshop, Katholieke Universiteit Leuven (10–11 May 2012) and the EPIP Conference “IP in motion”, Leuven (27–28 September 2012). We acknowledge funding by the Spanish National R&D Plan (Ministry of Science and Innovation grant CSO2009-10845).
Notes
1 They also find that the article citation rate of papers in patent-paper pairs weakly declines after the patent in the pair is granted.
2 This terminology follows the classification presented in OECD (Citation2011b) and is consistent with that used in Arnold et al. (Citation2010), distinguishing between scientific research institutes and government laboratories.
3 This type of institutes may be known in other countries as “centers of excellence” or “multidisciplinary research centers” (Bozeman and Boardman, Citation2003; OECD, Citation2011b).
4 See www.csic.es.
5 E.g. CIEMAT, Instituto de Salud Carlos III, INIA, IGME, INTA, CIMA.
6 The average share of tenured researchers and professors in public universities with respect to the total was 47 per cent in the academic year 2010–2011. For some universities, it was above 70 per cent (Polytechnic University of Madrid), and for others was around 20 per cent (University Pompeu Fabra). Source: http://www.mecd.gob.es.
7 For a description of the Catalan IRI model and a list of centres, see http://cerca.cat/en/general-characteristics/.
8 In the case of contract research or consulting projects developed by public researchers but funded by firms, the title of the patents can be de facto retained by the contractor. This can happen either as a result of negotiations between the contractor and the administration of the public institution or because the research was the result of consulting work carried out by the researcher without the knowledge of its institution. Maraut and Martinez (Citation2013) show that the share of academic inventions that are business owned differs across institutions and it is lower for CSIC than for universities.
9 Some extracts from the mission statements of Spanish IRIs are included as supplementary material.
10 A journal may be classified by SCOPUS in more than one field. For instance, 43 per cent of all articles in our database classified in the field “Biochemistry, Genetics and Molecular Biology” have it as their only field, but 17 per cent are also classified in “Medicine” and 9 per cent in “Chemistry”.
11 The SJR is calculated based on citations to articles published in a given journal in the previous 3 years, correcting for accumulated prestige and size of the journal (http://www.scimagojr.com/). It takes positive continuous values, with the maximum being in the range of 10–20, but the values themselves only make sense in the context of comparisons amongst journals, “they are just a way to put journals in order” (http://www.journalmetrics.com/faq.php).
13 A summary of the matching methodology is available as supplementary material.
14 Life-cycle aging effects lead researchers to be, in general, more productive during the early stages of their academic career (Levin and Stephan, Citation1991).
15 We use the author identifier available in SCOPUS to link articles in t with articles in t-1 by the same author. Our visibility measure is equal to zero either when the authors of a given article have no articles published in the previous year or when their previous year's publications have not received any citation.
16 The one-year lag is in line with Jensen et al. (Citation2011), who use the number of publications and the total number of citations received by faculty in the previous year as “proxies of faculty quality”. Since we only have publications of Spanish authors between 2003 and 2008, taking one-year lag to build a measure of visibility enables us to keep articles with publication years 2004–2008 for the analysis.
18 Additional tables with descriptive statistics for other relevant samples are included as supplementary material.
19 This can be related to the nature of the research they carry out, which may require larger research teams working in the same laboratories, or to their ability to establish larger research groups at their own institutions probably because they have access to more resources.
21 A similar figure for journal rank is provided as supplementary material.
22 Scientific fields, publication year and authors' institution dummies are jointly significant according to Wald tests in all regressions. Likelihood ratio tests that negative binomial outperforms Poisson models and Vuong statistic shows indifference of zero-inflated negative binomial and binomial regression for the estimation of the number of citations.
23 The marginal effect of the academic inventor dummy diminishes substantially when we include total visibility of Spanish authors, rather than the visibility of Spanish authors who are not academic inventors, indicating that a large share of the academic inventor's effect can be attributed to their visibility.
24 Results for universities, CSIC and MOC authors are available as supplementary material.
25 Coefficients are not different according to Chow tests.
26 Coefficients are different only with a 5 per cent level of significance according to Chow tests.