New indicators on patenting activities can support European innovation policy

New indicators on patenting activity in Europe, produced by the InnoS&T research project, increase understanding of the pathways by which inventions can lead to economic growth. Until now there has been no way to make direct and detailed comparisons of science and technology patenting activity across all regions and industry sectors in Europe, or explain to what extent scientific research leads to technological progress.

The number of patented inventions has exploded in the last three decades, and patenting technological advances has several benefits for society. When applying for a patent, an inventor includes ‘disclosure information’ on the science behind an invention and how it can be replicated. Sharing this knowledge is valuable as it can save time and form a basis for future developments. Patenting also provides a financial incentive to inventors who own a monopoly and can license or sell their technology.

Around 42% of patents taken out in Europe are unused by their holders, according to InnoS&T’s survey. Although this may be for strategic reasons, it has important implications for policies, such as decisions to fund public research organisations (PROs) or encourage links between science and industry.

The survey (PatVal-EU II) investigated 22,533 European Patent Office patents issued between 2003 and 2005 in 20 European countries, including some New Member States, for whom patenting is a relatively recent activity. Data from the US, Japan and Israel were also included. Patent holders were asked about the invention process, their motivations to patent, the use and value of the patent, and the links between science and patented inventions.

The researchers then used algorithms to develop indicators on the linkage between scientific research and technology innovations, based on citations of scientific research within the patents, by European region, industry sector, firms and research organisation.

Key questions investigated were: which factors determine the rate of commercialisation of inventions; to what extent did science contribute to this technological progress; are differences in technological performance between countries related to the scientific intensity of technology, based on citations with scientific references and university patents; how can policy support the production of valuable inventions and their commercial exploitation; and how can policy help remove impediments in the markets for technology.

Summary of main findings:

  • Around 53% of patents are used for commercial purposes, i.e. in a product, service or manufacturing process, 8% are licensed and 5% are sold.
  • Significant differences were found across Europe in the tendency to use patents commercially in new processes or products. In Austria and Denmark the use of patents is above average, in Germany, the UK, Japan and the US it is average, and in France and Spain it is below average.
  • The patent trade (sale and licensing) is above average in the UK and the US, and below average in France, Germany and Japan. Technology-based entrepreneurship varies, with the tendency to use patents to start a new firm above average in the UK and the US, and below average in France, Germany and Japan.
  • A higher science intensity of the national patent portfolio (the number of citations to science in patent applications) was found to relate positively to a country’s technological performance. However, only 29% of patents taken out by public research organisations are used, as compared with 55% for firms.
  • About 42% of patents remain unused, of which half are due to firms taking out patents for strategic reasons to block their rivals’ investments in related inventions. Other reasons for inventors not commercialising their inventions include applicants not possessing the assets to produce and market them or the small value of many inventions.
  • Larger firms (5000+) are less likely to use their patents commercially than smaller firms. Many large firms own unused patents, a large number of which have been shelved for blocking reasons. Patents resulting from collaboration with customers and users are more likely to be used.
  • Time saved for future developments by the information disclosed in patent applications is relatively low, varying from 2.7 hours in the telecoms sector to 99.8 hours in organic chemicals.
  • The proportion of female inventors in the sample was very low, at 5%.

Policy recommendations:

  • Create more favourable conditions for the creation of new technology‐based firms which contribute to economic growth.
  • Encourage markets for technology, such as online marketplaces, to enable use of unused patents by licensees in a better position to extract value from them.
  • Identify the technologies, the type of patent owner and patent characteristics that are most associated with significant misallocation of patent rights.
  • For early stage inventions, compulsory licensing policies should not be applied, and arguably inventions based on basic research should not be patented at all.
  • Alternatives to compulsory licensing, such as deferred examination of patents - allowed in some countries like Japan, Germany and the UK - would stimulate better exploitation of ‘sleeping’ patents.
  • Provide further encouragement for women to study science at university and take an increased share in technological invention.
  • Stimulate academic entrepreneurship through incentives to universities to become involved in patenting.

The researchers point out that it is important to encourage patent ‘breadth’ based on genuine innovations – i.e. the generality and applicability of a patent to several technological classes – and discourage those that are aimed at creating strategic defences. Understanding the extent of the ‘inventive step’ – the effort necessary to bring an invention to commercial production – is a good guide to future value, so that policies that encourage inventive steps are likely to enhance an invention’s economic value.

InnoS&T - Innovative S&T indicators for empirical models and policies: combining patent data and surveys (duration: 1/4/2008 – 31/3/2011). FP7 Socio-economic Sciences and Humanities, Activity 6 “Socio-economic and scientific indicators”, Research area 6.2 “Developing better indicators for policy”. Collaborative project (small and medium scale focused research project).

See: http://www.innost.unibocconi.it

Contact: Alfonso Gambardella, alfonso.gambardella@unibocconi.it