TIIP Newsletter

Universities are an important source of technical change. By the end of the 1990s, they accounted for about 50 percent of all basic research and almost 5 percent of all domestic patent grants in the U.S. (National Science Board, 2000). Academic research has had significant effects on the economy by increasing the productivity of private sector R&D and the growth in total factor productivity (Jaffe, 1989; Adams, 1990). There is ample empirical evidence that these benefits occur through knowledge transfers from academia to the rest of the economy, which sometimes take the form of licensing university-owned inventions to private firms. Such technology licensing activity has grown dramatically in the past two decades.[1]

Given the importance of university research for long-term growth and productivity, it is critical to understand what drives this research, and the technology licensing activity that follows. Is it a purely intellectual pursuit, as many commentators claim, or do economic incentives help shape scientists’ research interests? In this paper we take a first step in examining these issues by presenting econometric evidence on the role of economic incentives in university research and licensing outcomes.

Specifically, we examine how the share of license royalties received by academic inventors affects the number and licensing value of inventions in universities. In the United States, university intellectual property policies typically grant the university exclusive control rights over inventions. However, in all U.S. research universities the cash flow rights from licensing inventions are shared between the inventor and various parts of the university according to specified royalty sharing schedules. We show that there is substantial variation in these royalty sharing arrangements across U.S. research universities and use this cross-sectional variation to estimate the effect of royalty sharing arrangements on inventive output.

We focus on two outcomes: license income and the number of inventions disclosed by faculty scientists to the university technology licensing offices (TLOs). Our analysis is based on data from the Association of University Technology Managers, combined with information on the distribution of royalty shares, which we collected from university websites. The sample covers 102 U.S. universities between 1991 and 1999. We develop a simple model in which scientists can allocate effort to new research projects, to improving the quality of existing projects, and to other responsibilities, like teaching. Scientists attach private value to royalty income, publications, and teaching, and face different implicit incentives for these activities set by the university. The model predicts that a rise in the inventor's share of royalties will increase both the number of inventions and total license revenues.

The novel aspect of our data is the information on the distribution of license income between the university and the inventor(s).[2] One striking feature of these data is the great variation in inventor shares. In about half the universities surveyed, these royalty shares vary with the level of license income generated by an invention—we call these non-linear royalty schedules. Among the 58 universities using linear royalty schedules, the average inventor’s share is 41 percent, but there is substantial variation across universities. The minimum inventor royalty share among these universities is 25 percent, and the maximum is 65 percent. The royalty shares in the 44 universities with non-linear schedules display even larger cross-sectional variation within each license income interval. For these universities we compute an “expected” royalty share by weighting the share in each income interval by the probability of observing license income in that interval. The mean royalty share is 51 percent in universities with non-linear royalty sharing, and there is huge variation. The minimum expected inventor share among these universities is 20 percent and the maximum is 97 percent.

The other striking feature is that inventor royalty shares are either constant or decline with greater license income per invention-- royalty retention is regressive (equivalently, the university `tax' on inventors is progressive). On average, they start at 54 percent in the lowest interval and decline to 30 percent for inventions generating over $1 million. This feature holds for every university in our sample. It is also interesting to note that royalty shares are not related to observed university characteristics such as faculty size, university quality and the number of TLO professionals per faculty.

Our econometric work yields two key results. First, we show that academic research and inventive activity in universities respond to variations in inventors’ royalty shares. Controlling for a variety of other determinants, including university size and quality, and the level of R&D funding, we find that universities with higher royalty shares generate higher levels of license income. This finding implies that the design of intellectual property rights and other forms of incentives in academic institutions can have real effects on the direction of research. We also explore whether the incentive effects of royalty sharing induce greater effort by scientists, as our model predicts, or whether they affect the distribution of scientists among universities. We find evidence of the latter, but we cannot pin down the relative contribution of effort and such “sorting” using the available data.

Second, we show that the response to incentives and the effectiveness of technology licensing offices are much larger in private universities than in public ones. In private universities, the incentive effect is strong enough to produce a “Laffer effect,” whereby raising the inventor’s royalty share would increase the license revenue actually retained by the university. This finding implies that private ownership is important in the university sector. Why this is so, however, remains an open question for future research.

References

Adams, James (1990), “Fundamental Stocks of Knowledge and Productivity Growth,” Journal of Political Economy, 98(4), 673-702

Jaffe, Adam (1989), “Real Effects of Academic Research,” American Economic Review, 79(5), 957-970

National Science Board (2000), Science and Engineering Indicators (Washington D.C.: National Science Foundation)

[1] Part of this rapid growth in university innovation and licensing activity is due to the passage of the Bayh-Dole Act of 1980 (Patent and Trademarks Amendments Act, PL 965-17), which gave universities the right to patent and a mandate to license discoveries made with federally-sponsored research to the private sector. By the year 2000, nearly all American research universities had established, or expanded, a technology licensing office and introduced explicit intellectual-property policies and royalty-sharing arrangements for academic scientists.

[2] The inventor retains a given percentage of net licensing income from the license and the rest is allocated to the inventor’s lab and department, and to the university. The criterion we use for identifying the inventor’s share is that the inventor must gain either cash-flow rights or must have direct control rights over the income (e.g. lab research money).


		Incentives and Invention in Universities NBER Working Paper No. 9727, May 2003 Saul Lach (The Hebrew University of Jerusalem and NBER) and Mark Schankerman (London School of Economics and CEPR) FULL TEXT (fee required)
		--Summary by Mark Schankerman
		Universities are an important source of technical change. By the end of the 1990s, they accounted for about 50 percent of all basic research and almost 5 percent of all domestic patent grants in the U.S. (National Science Board, 2000). Academic research has had significant effects on the economy by increasing the productivity of private sector R&D and the growth in total factor productivity (Jaffe, 1989; Adams, 1990). There is ample empirical evidence that these benefits occur through knowledge transfers from academia to the rest of the economy, which sometimes take the form of licensing university-owned inventions to private firms. Such technology licensing activity has grown dramatically in the past two decades.[1] Given the importance of university research for long-term growth and productivity, it is critical to understand what drives this research, and the technology licensing activity that follows. Is it a purely intellectual pursuit, as many commentators claim, or do economic incentives help shape scientists’ research interests? In this paper we take a first step in examining these issues by presenting econometric evidence on the role of economic incentives in university research and licensing outcomes. Specifically, we examine how the share of license royalties received by academic inventors affects the number and licensing value of inventions in universities. In the United States, university intellectual property policies typically grant the university exclusive control rights over inventions. However, in all U.S. research universities the cash flow rights from licensing inventions are shared between the inventor and various parts of the university according to specified royalty sharing schedules. We show that there is substantial variation in these royalty sharing arrangements across U.S. research universities and use this cross-sectional variation to estimate the effect of royalty sharing arrangements on inventive output. We focus on two outcomes: license income and the number of inventions disclosed by faculty scientists to the university technology licensing offices (TLOs). Our analysis is based on data from the Association of University Technology Managers, combined with information on the distribution of royalty shares, which we collected from university websites. The sample covers 102 U.S. universities between 1991 and 1999. We develop a simple model in which scientists can allocate effort to new research projects, to improving the quality of existing projects, and to other responsibilities, like teaching. Scientists attach private value to royalty income, publications, and teaching, and face different implicit incentives for these activities set by the university. The model predicts that a rise in the inventor's share of royalties will increase both the number of inventions and total license revenues. The novel aspect of our data is the information on the distribution of license income between the university and the inventor(s).[2] One striking feature of these data is the great variation in inventor shares. In about half the universities surveyed, these royalty shares vary with the level of license income generated by an invention—we call these non-linear royalty schedules. Among the 58 universities using linear royalty schedules, the average inventor’s share is 41 percent, but there is substantial variation across universities. The minimum inventor royalty share among these universities is 25 percent, and the maximum is 65 percent. The royalty shares in the 44 universities with non-linear schedules display even larger cross-sectional variation within each license income interval. For these universities we compute an “expected” royalty share by weighting the share in each income interval by the probability of observing license income in that interval. The mean royalty share is 51 percent in universities with non-linear royalty sharing, and there is huge variation. The minimum expected inventor share among these universities is 20 percent and the maximum is 97 percent. The other striking feature is that inventor royalty shares are either constant or decline with greater license income per invention-- royalty retention is regressive (equivalently, the university `tax' on inventors is progressive). On average, they start at 54 percent in the lowest interval and decline to 30 percent for inventions generating over $1 million. This feature holds for every university in our sample. It is also interesting to note that royalty shares are not related to observed university characteristics such as faculty size, university quality and the number of TLO professionals per faculty. Our econometric work yields two key results. First, we show that academic research and inventive activity in universities respond to variations in inventors’ royalty shares. Controlling for a variety of other determinants, including university size and quality, and the level of R&D funding, we find that universities with higher royalty shares generate higher levels of license income. This finding implies that the design of intellectual property rights and other forms of incentives in academic institutions can have real effects on the direction of research. We also explore whether the incentive effects of royalty sharing induce greater effort by scientists, as our model predicts, or whether they affect the distribution of scientists among universities. We find evidence of the latter, but we cannot pin down the relative contribution of effort and such “sorting” using the available data. Second, we show that the response to incentives and the effectiveness of technology licensing offices are much larger in private universities than in public ones. In private universities, the incentive effect is strong enough to produce a “Laffer effect,” whereby raising the inventor’s royalty share would increase the license revenue actually retained by the university. This finding implies that private ownership is important in the university sector. Why this is so, however, remains an open question for future research. References Adams, James (1990), “Fundamental Stocks of Knowledge and Productivity Growth,” Journal of Political Economy, 98(4), 673-702 Jaffe, Adam (1989), “Real Effects of Academic Research,” American Economic Review, 79(5), 957-970 National Science Board (2000), Science and Engineering Indicators (Washington D.C.: National Science Foundation) [1] Part of this rapid growth in university innovation and licensing activity is due to the passage of the Bayh-Dole Act of 1980 (Patent and Trademarks Amendments Act, PL 965-17), which gave universities the right to patent and a mandate to license discoveries made with federally-sponsored research to the private sector. By the year 2000, nearly all American research universities had established, or expanded, a technology licensing office and introduced explicit intellectual-property policies and royalty-sharing arrangements for academic scientists. [2] The inventor retains a given percentage of net licensing income from the license and the rest is allocated to the inventor’s lab and department, and to the university. The criterion we use for identifying the inventor’s share is that the inventor must gain either cash-flow rights or must have direct control rights over the income (e.g. lab research money). This work is licensed under a Creative Commons License.