| ||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © The Authors 2007. Published by Oxford University Press.
The value of intellectual property rights to firms and society
* Oxford Intellectual Property Research Centre and St Peter's College, Oxford, e-mail: christine.greenhalgh{at}economics.oxford.ac.uk
** Oxford Intellectual Property Research Centre and Harris Manchester College, Oxford, email: mark.rogers{at}hmc.ox.ac.uk
 |
Abstract |
|---|
Economists view intellectual property rights (IPRs) as policy tools for encouraging innovation, but they recognize that they can also inhibit competition. There are many types of IPRs and institutions concerned with their administration. We begin by outlining how these complex and varied rights are supposed to work and how they interact with other characteristics of firms and markets. We then survey the available literature on patents, trade marks, and copyright to assess the value of these IPRs to firms and the costs to firms of acquiring and defending their rights. The paper concludes with suggestions for topics requiring further research to inform public policy better.
Key Words: innovation • intellectual property • patents • trade marks • copyright
Some parts of this article draw on an earlier review by Dixon and Greenhalgh (2002) and thus the authors wish to acknowledge the helpful contribution of Padraig Dixon.
1 For an explanation of the legal frameworks relating to these rights see Cornish (1999).
2 See details in Table 1.1 of the Gowers Review of Intellectual Property (HM Treasury, 2006, p. 20).
3 In addition, the Japanese patent applications have been scaled down by a factor of three since each Japanese patent is commonly thought to represent around one-third of a US patent (i.e. the Japanese system breaks down an invention into more discrete stages).
4 In common with many other econometric studies, these authors do not use patent renewals data (unlike the costs approach to patent value) to construct actual patent stocks; rather, they use an assumed depreciation rate. Clearly, citation weights pose further problems in that the importance of any patent is revealed slowly throughout its duration and some citations may be yet to come, necessitating either the estimation of unobserved future citations or the adoption of a fixed time cut-off.
5 There is thus a problem of mis-specification in that Bloom and Van Reenen estimate a gross output production function (using real sales as their dependent variable), but do not include as regressors any variables related to the use of intermediate inputs.
6 These results are based on panel estimators that control for persistent, time-invariant differences between firms. Further analysis of these firm-level persistent differences shows that they are associated with the presence of R&D and IPR activity for both high- and low-tech firms. Firms that are never R&D or IPR active have persistently lower productivity. Hence, Greenhalgh and Longland suggest that firms need continually to renew their intangible assets stocks to improve both their production technology and product offering.
7 Their study covers all patents applied for in the UK, France, and Germany for the period from 1950 to 1979, but with no breakdown for industrial sectors in any of the countries. Schankerman (1998) presents estimates of the private value of patent protection among different technology fields.
8 The ratio of patent grants to applications was only 33 per cent in Germany compared with 83 per cent in the UK and 93 per cent in France, (Schankerman and Pakes, 1986, Table 1). This suggests that patents accepted in Germany are likely to be of a somewhat higher value than those in the other two countries.
9 German data were used because, as noted above, the German system of patent renewal is particularly rigorous in rejecting applications of low inventive output and because of its highly progressive renewal fee schedule.
10 In related work, Harhoff et al. (1999) showed that within the class of patents renewed to full term in both the USA and Germany, citations of the patent rise with estimated value, although this relationship is somewhat noisy.
11 Boltho and Toniolo (1999) contains these and other statistics on growth in GDP per capita.
12 This is not a new issue; for example, in the nineteenth century, US book publishers were quick to copy new books by Charles Dickens. In fact, the USA was not a full signatory to all international copyright agreements until 1989 (Khan, 2004).
13 For example, there was a serious sample selection problem arising from the non-random nature of samples, since the sample of firms studied typically only included those firms that reported R&D. A further issue was the extent to which studies managed successfully to control for firm characteristics other than size. This creates problems for empirical work since the intensity of R&D varies across firms (Cohen et al., 1987). A related problem is the need to control for industry effects. Since firm size and innovation are likely to be affected by attributes of the overall industry, such as the level of technological opportunity and regulatory considerations, studies that use inter- and intra-industry data need to control for the industry-level effects in order to obtain unbiased estimates of the effect of the specific influence of firm size on innovation. Moreover, as Cohen (1995) notes, controlling for industry effects in firm-level data can be extremely difficult, given that many large firms are often composed of smaller units that operate in separate industries.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. S. Siegel and M. Wright Intellectual property: the assessment Oxf. Rev. Econ. Policy, December 1, 2007; 23(4): 529 - 540. [Abstract] [Full Text] [PDF]
|
|