The determinants of academic patenting by Italian universities

Abstract

This work investigates the determinants of patenting at Italian universities. We test the effect of public and external funds, in terms of commercial activity by universities, towards the propensity of universities to file patents at the European Patent Office. We then estimate the effect of several variables referring to the contextual characteristics of both university and localisation. The results suggest that, while in the southern regions of Italy commercial income bears a negative effect on the patenting activity of universities, in the other regions we register an important positive effect. We moreover find that the adoption of internal university regulations relating to patenting and the fact of being embedded in an innovative region are factors that exert a positive influence on the university rate of application for patents.

Keywords:

• academic patenting
• university funding
• regional context
• technology transfer

Acknowledgements

A previous version of this paper was presented at the ‘Workshop on Innovation Dynamics’, University of Minho, Portugal (2013). Comments from the audience were very useful. We are also grateful to two anonymous referees. All remaining errors are those of the authors.

Notes on contributors

Ugo Rizzo is a post-doctoral researcher at the Department of Economics and Management of the University of Ferrara. He obtained a MSc at Manchester Business School and the title of Doctor Europaeus (PhD) at the University of Ferrara. His main research interests regard university–industry relations.

Laura Ramaciotti is an associate professor at the Department of Economics and Management of the University of Ferrara. She was director of the Technology Transfer Office of the University of Ferrara from 2004 to 2010, one of the top Italian universities in technology transfer activities according to the Italian Research Evaluation Committee.

Notes

1. Regulation that allows professors to keep the property of the invention they patent.

2. We searched for patents in which the university appears among the applicants. In order to avoid missing data, we searched both with the Italian and with the English university name (when present), with each of the different university names (for example ‘Univeristà Federico II di Napoli’), and with a combination of different ‘university’ suffixes (e.g. uni or univ).

3. We excluded from the sample the online universities.

4. The patenting experience is calculated through the following equation: where K is the stock of patents, δ is the depreciation rate, and INV is the number of new patents in year t.

5. The number of patent applications include the university applications.

6. Italy is divided in 120 NUTS 3 regions, called Provinces, which are small geographical areas. We therefore decided to include also the NUTS 3 bordering regions because considering only the NUTS 3 regions would be restrictive in terms of university influence.

7. The inflated part of the model tests the relationship of the independent variables toward the probability of having a zero as outcome in the dependent variable.