Academic Patenting in Europe: a Reassessment of Evidence and Research Practices

Academic Patenting in Europe: A Reassessment of Evidence and Research Practices

Francesco Lissoni, GREThA-Université Bordeaux IV & CRIOS-Università Bocconi, Milano

Correspondenceaddress: Francesco Lissoni, GREThA - UMR CNRS 5113, Université Montesquieu - Bordeaux IV, av. Léon Duguit, 33608 Pessac cedex, France. Email:

Abstract

European universities and public research organizations (PROs) contribute substantially to their countries’ overall inventive activity, but are far from being exclusive owners of the related intellectual property. Business companies and, to a lesser extent, individuals appear to play a major, sometime dominant role. . This special issue offers a selection of papers addressing issues of measurement, commercialization, and ownership of such academic patents in Europe. Measuring the extent of the phenomenon requires identification of the academic inventors, a data mining operation that imposes technical as well as procedural challenges for social scientists. The heterogeneity of ownership models poses the question of whether ownership is related to the patents’ quality and/or successful commercialization. Further questions concern the identity and business models of firms holding academic patents in their portfolios, and the economic and legal factors that explain a university’s or PRO’s choice of whether to maintain the intellectual property of its staff’s inventions. The papers in this special issue discuss these topics at length and rely on data collected following a joint methodology, made available to readers for use and extension.

Key Words: academic patenting, inventor data, intellectual property, APE-INV

JEL Codes: I23, O31, O34

Acknowledgements:

Funding for all the activities of the APE-INV Programme has been provided by several member organizations of the European Science Foundation (ESF) and by the University of Strasbourg). The ESF’s Humanities and Social Sciences Unit has provided continuous assistance and showed great patience to me. The APE-INV Chair was hosted by KITES (now CRIOS) – Bocconi, and could have not accomplished its tasks without the assistance of Michele Pezzoni (APE-INV External Coordinator), Monica Coffano, and Sabrina Miraglino. The Department of Informatics, Systems and Communication (DISCo) of the University of Milano-Bicocca kindly agreed to host the project’s data server, which was managed by Lucio Leone, under Andrea Maurino’s direction. Members of the APE-INV Steering Committee have both provided scientific advice and managerial help at different stages of the programme. The APE-INV workshops and final conference were made possible by the contribution of too many people to be all named here (for a complete list: Gianluca Tarasconi provided technical advice, moral support, and cheeky humour throughout the project and beyond. This special issue was made possible Susan Lees’ brilliant editing and handling of the refereeing process, and by editor-in-chief Mark Lorenzen’s patience and guidance. No less than 30 referees contributed to select and improve the papers, with remarkable timeliness.

Academic Patenting in Europe (APE) is a research theme, the name of a programme funded by the European Science Foundation (ESF), as well as a policy issue. As a research theme, it is now about 10 year old, and a number of surveys exist, which can provide the reader with a complete overview of progress so far (Geuna and Rossi, 2011; Lissoni, 2012 and 2013). This Introduction will focus on pointing out a number of methodological issues that emerged during this decade, justifying the implementation of a dedicated research programme funded by the European Science Foundation, and illustrate a number of substantial outcomes of this programme. The Introduction will also discuss two sets of policy issues: On the one hand, universities’ handling of their staff’s inventions, and on the other, the economics of research based on large data archives, with special emphasis on the problem of data sharing.

As a research theme, APE originates in the early 2000s, after the appearance of several studies documenting the impressive growth of patent filings by US universities throughout the 1980s and early 1990s. Such growth stood in contrast with the (apparently) near-to-zero figures of most European countries. At a time when European policy-making on S&T was still dominated by the Green Paper on Innovation of 1995 and the “European Paradox” paradigm (EC, 1995; Dosi et al., 2006), the contrast was quickly seized as a piece of evidence on European universities’ lack of engagement in technology transfer activities. Much attention was then dedicated to propose or produce some European version of the most celebrated piece of US legislation in the field, the Bayh-Dole Act of 1980 (Mowery and Sampat, 2005: Grimaldi et al., 2011).

A number of European researchers in economics and related social sciences, however, decided to pay attention not only to official statistics, but also to also history and institutions. They noticed that US universities’ IP dealings date back to the early 20th century, when a number of not-for-profit institutions started diffusing IP awareness and expertise among university administrators (Apple, 1989; Mowery and Sampat, 2001; Mowery et al., 2002; George, 2005). And that such a long history combined with the autonomous status of both private and public US universities, which allowed them to dispose of their financial assets and negotiate IP rights with both their faculty and sponsors (Geiger 1986, 1993, and 2004). On the contrary, while European academics had been patenting, at the individual level, throughout the late 19th and early 20th century (Guagnini, 2011; Becker, 2012; MacLeod, 2012), their universities abstained for long from getting involved, for two reasons. First, several countries’ IP legislation had been for long characterized by the “professor’s privilege”, a norm that exempts academic personnel from the duty to disclose and relinquish their inventions to their employers (Lissoni et al., 2009; von Proff et al., 2012; Damsgaard and Thursby, 2013). Second, European universities never enjoyed the same degree of financial and managerial autonomy of their US counterparts (Ben-David, 1977; Clark, 1993; Estermann, 2009). Therefore, they had neither the incentives nor the administrative tools and skills necessary to take title of and commercialize their professors’ inventions. The latter were then tacitly encouraged to deal personally with IP matters, either patenting in their own name or trading their IP to business companies in exchange for sponsorship.

These historical observations had two consequences. At the data collection level, they led early researchers on APE to adopt a definition of “academic patent” based on the inventor’s identity, that is to classify as academic not only the patents owned by universities, but also those authored by academic scientists. These researchers engaged in heroic, but uncoordinated and often duplicative data-mining efforts to assemble information on both inventors and individual academics. Their results revealed the extent of a phenomenon until then misunderstood, and made once and forever clear that, at least in terms of relative contribution to domestic patenting, no gap, or a very limited gap existed between academic patenting in Europe and in the US. The papers presented in this special issue confirm all the results obtained by this early literature and extend them in space (by adding new countries to the list for which evidence is available) and in time (by producing, in a few cases, the first reliable time series data, which are necessary to spot any trend in the phenomenon).

This change of perspective also led to the emergence of both a new set of research questions, and of a new awareness of the technical and organizational challenges posed by inventor-based data mining. It is these questions and challenges that APE-INV, the programme funded by the European Science Foundation, decided to take up. The programme is a Database Harmonization project ( whose objectives can be summarized as follows: to share expertise and methods for name disambiguation (aka entity resolution), by borrowing the necessary techniques from information technology and adapting them to patent documents;

1. to produce a number of open access datasets for APE research purposes, namely:

1.1an inventor database, comprising all inventors (academic and non-academic), listed on EPO patents from 1978 onward;

1.2a collection of country datasets on academic patents, produced with a common methodology and therefore comparable;

1. to encourage data sharing and coordination of data-mining efforts, in order to avoid wasteful duplications and the propagation of ill-informed research results, based on obsolete or amateurish data-mining techniques (on bias due to poor quality inventor data, see Raffo and Lhuillery, 2009).

Objective 2 was met by organizing the NameGame workshop series, with multidisciplinary contributions from both social scientists and IT scholars, whose papers or presentations can be downloaded from the APE-INV website. All together, these contributions form a state-of-the-art set of methodologies, to which one may wish to add Lai et al.’s (2011) application to USPTO data.

As for the datasets produced to meet objective 1, they are also available from the APE-INV website. The importance of inventor data goes well beyond their application to the identification of academic patents, as they are widely used for research on the geography of innovation and social networks (survey by Moreno and Miguelez, 2012). As for the academic patent country datasets (which we can provide only in anonymized form, for privacy reasons), they are used by papers in this special issue, but all authors encourage their further use both for reproducing and extending their results.

The papers presented in this special issue can be read from different angles. Taken together, they offer a cross-country view of the APE phenomenon, with special emphasis on the issue of ownership of academic inventions, and its relationship to patent value. Six out of seven papers are country studies, which offer up-to-date evidence on Belgium (Flanders), Germany, Italy, Spain, Sweden and the UK. Besides confirming the importance of academics as contributors to each country’s inventive efforts, the studies document a trend away from the European traditional ownership pattern (academic patents being assigned, for the largest part, to business companies) towards a more US-like one, with universities reclaiming exclusive or joint intellectual property of their staff’s inventions. The studies on Belgium ( paper by Callaert et al.), Germany ( paper by Schoen and Buenstorf)and Italy (paper by Lissoni et al.), taken together, show a convergence of European IP laws and academic institutional structures towards the US model, most notably through the abolition of the professor’s privilege (in Germany) and the increasing autonomy granted to universities (in Italy). Whether this trend should be evaluated positively is open for discussion. The evidence on Belgium suggests that university-owned academic patents receive more forward citations than firm-owned ones, which may imply that universities are good at retaining control of promising inventions and/or at commercializing them. The same, however, does not apply to Germany, for which the paper by Schoen and Buenstorf find that academic patents owned by general universities are less cited than those invented in the same universities, but assigned to industry (while no difference exists in the case of technical universities). These results contradict previous studies on Germany, and are in line with what recently found by Lissoni and Montobbio (2013) for universities in France, Italy, Denmark, and Sweden, and by Sterzi (2012) for the UK. The case of Belgium, however, seems more in line with that of the Netherlands. We can then speculate that universities’ experience and flexibility in handling IP issues is at stake here, with universities having a longer tradition and more autonomy (such as the Flemish and the Dutch ones) being better positioned to handle IP. This does not exclude within-country heterogeneity, with some universities being in the position to integrate their IP policies in a general strategy for technology transfer and commercialization, rather than just deciding case by case on whether to reclaim IP ownership. Finally, it remains to be seen to what extent universities may choose to go for patent ownership for signalling purposes, and whether such a choice, if confirmed, may be influenced by the increasing pervasiveness of research evaluation exercises conducted by national governments, through their university and research agencies.

The paper by Giuri et al. goes beyond the use of patent citations in trying to assess universities’ ability in commercializing their staff’s inventions, and investigates the extent of licensing and spin-off creation, as opposed to patent sales. It also compares universities with another important source of inventions stemming from public funded research, namely Public Research Organizations (PROs). It finds that university-owned patents are more likely of PRO-owned ones of being licensed, which suggests that universities may be better at commercializing their staff’s inventions that PROs. But it also finds that licensing and patent sales are alternatives, with the former being more common in countries where universities have a higher tendency to reclaim ownership of academic patents, It remains to be assessed whether the licensing schemes used by universities perform better than sales either in promoting the application of the academic inventions, and/or in producing net profits for the university.

Two papers investigate the business-owned academic patents, which until recently had attracted scant attention. The paper by Ljunberg et al. studies the forward citation rate of Swedish academic patents owned by large companies, and find that it depends on whether the patented inventions are related to the companies’ core technologies. The paper by Lawson investigates the property structure of business companies holding UK academic patents in their portfolios, and finds that an unexpected large number of small companies are indeed university-funded start-ups. This challenges the dichotomous distinction (university- versus firm-owned academic patents) used until now in the field. Further research is required to assess to what extent the phenomenon is common also in countries that, differently from the UK, have little or no tradition of universities’ involvement into entrepreneurial ventures.

Finally, the paper by Martinez et al. come back to an issue that has attracted considerable attention in the recent past, namely the relationship between academics’ inventor status and their scientific productivity. They do so by considering, along with universities, a heterogeneous set of Spanish PROs. Their results generally confirm the existing evidence, namely that academic inventorship is associated to high scientific quality, but also find an exception, which concerns a new generation of research institutes recently created by the Spanish government with the mission of performing applied research in collaboration with the private sector.

Future research will have to further investigate the issue of academic patent ownership and its relationship with commercialization of research results. This ought to be made by paying great attention to the institutional arrangements concerning university autonomy and individual universities’ experience and managerial models. This will require shifting the researchers’ efforts from data collection at the inventor and patent level towards investigating universities’ organizational arrangements, financial conditions, and ties to industry. Yet, data collection on inventors and patent cannot be discontinued, as updating existing datasets will be necessary to keep monitoring existing trends. In order to reconcile these two necessities, time may have come to institutionalise the inventor data collection efforts, with the intervention of national or international agencies. This brings us to the issue of data sharing as a research model: social and economic research on innovation is turning into big science, and if policy makers wish to obtain sensible advice they need to move in the direction of assisting the change. Some encouraging signs in this direction are available, witness the creation of the EPO Worldwide Patent Statistical Database and the rapid growth of its user community ( or the StarMetrics initiative, launched by several US funding agencies for both evaluation and research purposes ( In that respect APE-INV is part of a general trend that will hopefully break the isolation in which most social scientists work when it comes to large scale data collection, of which inventor data are an important part.

References

APPLE, R. D. 1989. Patenting University Research: Harry Steenbock and the Wisconsin Alumni Research Foundation. Isis, 80, 374-394

BECKER, M. C. 2012. Entrepreneurial imprinting and organizational persistence: The case of carl zeiss. Preprint: Danish Institute for Advanced Study ed. Odense, Denmark: University of Southern Denmark.

BEN-DAVID, J. 1977. Centres of learning: Britain, France, Germany, United States, New York NY, McGraw-Hill.

CLARK, B. 1993. The Research foundations of graduate education: Germany, Britain, France, United States, Japan, Berkeley CA, University of California Press.

DAMSGAARD, E. F. & THURSBY, M. C. 2013. University entrepreneurship and professor privilege. Industrial and Corporate Change, 22, 183-218.

EC 1995. Green paper on innovation. Brussels: European Commission.

ESTERMANN, T. & NOKKALA, T. 2009. University Autonomy in Europe - Exploratory Study. Brussels.

GEIGER, R. L. 1986. To advance knowledge: The growth of American research universities, 1900-1940, Oxford University Press (2004 edition: Transaction Books).

GEIGER, R. L. 1993. Research & relevant knowledge: american research universities since World War II, Oxford University Press (2004 edition: Transaction Books).

GEIGER, R. L. 2004. Knowledge and money: Research universities and the paradox of the marketplace, Stanford University Press.

GEORGE, G. 2005. Learning to be capable: patenting and licensing at the Wisconsin Alumni Research Foundation 1925–2002. Industrial and Corporate Change, 14, 119-151.

GRIMALDI, R., KENNEY, M., SIEGEL, D. S. & WRIGHT, M. 2011. 30 years after Bayh–Dole: Reassessing academic entrepreneurship. Research Policy, 40, 1045-1057.

GUAGNINI, A. 2011. Imprenditorialità scientifica e scienziati imprenditori in Gran Bretagna tra Otto e Novecento. In: LISSONI, F. (ed.) L'imprenditorialità accademica: Un analisi multidisciplinare. Roma: Carocci.

LAI, R., D’AMOUR, A., YU, A., SUN, Y., TORVIK, V. & FLEMING, L. 2011. Disambiguation and co-authorship networks of the US Patent Inventor Database. Harvard Institute for Quantitative Social Science, Cambridge, MA, 2138.