رشد و کارآیی تولید از صدور مجوز مالکیت معنوی دانشگاه

Licensing activity of US universities has increased substantially in recent years. We examine this increase focusing on the ‘productivity’ and ‘changes in productivity’ of licensing activity. While it is generally acknowledged that there has been a dramatic increase in licensing, there is little understanding of the licensing process as it relates to university characteristics, nor is there evidence on the extent to which this is the result of increased resources devoted to commercialization or whether it stems from other factors. This paper employs data envelopment analysis (DEA) combined with regression analysis to examine the overall productivity of university licensing activity as well as the productivity of individual universities. We find that licensing has increased for reasons other than increases in overall university resources. Given input levels, universities are today more commercially productive than they were in the recent past; we propose reasons for this shift and analyze factors associated with the shift and factors associated with relative levels of commercialization.

It has been suggested in a number of venues that university resources are not fully exploited as a source of economic growth and competitiveness and recent public policy has been aimed at increasing the commercial impact of universities.1 For reasons of greater US ‘competitiveness’, the federal government has encouraged greater interactions between universities and the private sector. The Bayh–Dole Act of 1980 changed the nature of ownership of inventions developed under federally funded programs. With passage of the Act, universities could elect to retain title to such inventions, but they are required to file patent applications on those inventions. The Act also encourages the technology transfer activities of universities. The federal government has not been alone in encouraging university licensing activity. State governments have looked increasingly toward public universities as sources of economic development.2 Have these efforts made a difference? According to results reported in the AUTM Licensing Survey (Association of University Technology Managers, 1997), as well as AUTM press releases, university licensing activity has increased over the 6-year period during which AUTM has collected data. Using the sample of universities that participated in all years of the survey, we find that the number of license agreements rose by nearly 70% and the amount of royalties received (in real terms) more than doubled between 1991 and 1996. Prior to 1980, fewer than 250 patents were awarded annually to universities; currently, over 1500 patents are awarded annually. The number of universities actively engaged in technology transfer has increased eight-fold since 1980 to now more than 200. In this paper, we examine closely this increase in licensing activity. In particular, our focus will be on the ‘productivity’ and ‘changes in productivity’ of licensing activity by US universities over a 6-year period. While it is generally acknowledged that there has been a dramatic increase in university licensing and patenting, there is little understanding of the licensing process as it relates to university characteristics, nor is there evidence on the extent to which this is the result of increased resources devoted to commercialization or whether it stems from other factors. This paper employs data envelopment analysis (DEA) combined with regression analysis to examine the overall productivity of university licensing activity as well as the productivity of individual universities. We find that licensing activity has increased for reasons other than increases in overall university resources. That is, given input levels, universities are today more commercially productive than they were in the recent past; we propose reasons for this change. The regression analysis allows us to relate efficiency to university characteristics. We find that private universities tend to be more efficient in commercialization than public, while universities with medical school are less likely to be efficient. The latter result is particularly interesting given that the majority of university licenses are in the life sciences. We also examine efficiency as related to the size and quality of research faculty in biological sciences, engineering, and physical sciences. Our measures of faculty input in biological sciences and engineering are significantly related to efficiency, while they are not for physical sciences; we propose reasons for this difference. In our study of the licensing of university intellectual property (IP) we consider the levels and changes in the levels of the ‘outputs’. 1. Sponsored research agreements between universities and industry. 2. License agreements which permit the use of university IP by private sector firms. 3. Royalty payments received by universities in exchange for the use of IP. 4. Disclosures by faculty to their central administration of potentially commercializable innovations. 5. University patent applications. Each of these outputs is an integral part of a university’s efforts to obtain commercial rewards from the creation of IP.3 We settle on these five outcomes as commercialization outputs based on interviews and survey results reported in Thursby et al. (2001). Each is viewed as important in measuring the success of a university’s technology transfer. For want of better terms we use ‘commercialization’, ‘commercial activity’ and ‘licensing activity’ as succinct ways of referring to the process by which universities produce these five outcomes.

The federal and state governments have encouraged universities to commercialize the results of their research. Universities espouse an increasing willingness to engage in such activities and private sector firms are looking more carefully at university IP. We examine university commercial activities in order to shed light on the activities of individual universities as well as the overall direction of such activities. Commercial activities include industry sponsored research and royalties as well as numbers of invention disclosures, licenses executed and new patent applications. University productivity or efficiency with regard to such activities varies not only according to the capabilities of the faculty and staff, but also according to university preferences in the use of their resources. We use DEA to measure the relative efficiency of each university in our sample. Both contingency table analyses and regressions are used to relate the efficiency scores to the levels of all commercial outputs as well as to the levels of inputs. We also examine the changes in productivity of university commercial activity over time. Our major results are the following. • We find substantial evidence of inefficiencies. As we note, this inefficiency may stem simply from university preferences for or specialization in outputs unrelated to licensing activity (such as basic research and teaching) rather than from competencies in licensing. • There has been substantial growth in commercial activities of US universities. We attribute this both to a changing environment within universities regarding commercial activity as well as an increasing desire of industry for university technologies. • This growth has stemmed primarily from a growth in commercialization by all universities rather than a ‘catching up’ by the inefficient institutions. • We find that biological sciences and engineering are more important to licensing activity than are the physical sciences. We attribute this to the more applied nature of engineering and the better market opportunities and orientation toward markets of biological sciences. • We find that, on a number of dimensions, the smallest schools tend to be more like the largest schools than the mid-range schools. • The lower the research quality of a university the more efficient the university tends to be in commercial activity. This we interpret as resulting from greater specialization in basic research of the higher quality research faculty. • We find that private universities are more likely to be efficient and universities with medical schools are less likely to be efficient. It has been suggested in a number of venues that university resources are not fully exploited as a source of economic growth and competitiveness. Our results clearly show that this criticism is decreasing in its relevance, though the criticisms of those who feel that universities have gone too far in commercialization activities may be increasing in relevance. What we have not been able to fully determine is the source of the growth in university commercial activity. To what extent is it due to a shift of resources away from basic research to more commercializable research?11 To what extent does it follow from an increasing willingness and ability of universities to license their IP? To what extent is this growth due to a greater demand by industry for university IP? Answers to these questions are the subject of future research.