1. Bayh-Dole and Entrepreneurship Reconsidered: University versus Inventor Ownership* Martin Kenney Dept. of Human and Community Development UC Davis & Berkeley Roundtable on the International Economy & Donald Patton Dept. of Human and Community Development UC Davis Presented in North Central Regional Center for Rural Development Webinar October 18, 2011

2. Motivation for Paper Appreciative modeling exercise suggested inventor ownership should be effective (Kenney and Patton 2009) Entrepreneurship and technology transfer long time US policy goals

3. Motivation In response to increasing questioning in the academic literature, a 2010 National Research Council Report stated: “Arguments for the superiority of an inventor driven system of technology transfer are largely conjectural. There is certainly anecdotal evidence of faculty dissatisfaction with the technology licensing office-dominated model as well as evidence of faculty entrepreneurial success independent of such offices, but there is no systematically collected evidence that inventors have knowledge and skills superior to those of technology transfer personnel and their service providers in the various components of IP acquisition, management, and licensing.”

4. Methodology Find inventor-ownership university –Cambridge, Stanford, Wisconsin – all changed –Only Anglo-Saxon pure inventor ownership univ. -- Waterloo, Canada Collect all technology-based startups –Internet search, documents, interviews, TLO offices Extremely strict definition about firms to be included –Decisions made by both authors

5. Hypothesis With various controls, we expect greater entrepreneurship at inventor-ownership university, i.e., Waterloo But conditioned by: 1. University academic ranking 2. Academic field 3. University R&D expenditures 4. Number of professors

6. Population and Data Waterloo versus University of Wisconsin, Madison; University of Michigan, Ann Arbor; University of Illinois, Urbana- Champaign; UC Davis; UC Santa Barbara Spin-offs, University/Field ranking, R&D Expenditures, Number of Faculty

7. Spin-offs Academic Ranking R&D Expenditures* Faculty Size Attribute Description Number of Spin- offs, 1957-2010 2010 SJTU Ranking $ millions, 2005-2008 avg. 2005-2008 avg. Total BMS, CS&EE, EPS faculty UWM140177922,195 UMAA88227243,193 UIUC72254622,120 UCD40465632,038 UCSB3732163559 Waterloo138151-20095963 Sources: various * Total for BMS, CS&EE, and EPS; all others excluded Overview of the Data

8. Cumulative Number of Spin-offs by University, 1957-2009

9. OverallWorldRankingNatural Sciences and Mathematics Eng., Tech. and Computer Science Life and Agric. Sciences Medicine and Pharmacy Computer Science UWM1717231120 52-75 UMAA22217248 18 UIUC2523418n.a. 13 UCSB32191551-75n.a. 44 UCD46464621n.a. 50 Waterloo151-200n.a.52-75n.a.n.a.76-100 2010 Shanghai Jiao Tong University Global Academic Rankings, Overall and Selected Technology Categories

10. BiomedicalCS&EEEPSTotal UWM 2,1991048663,169 (274.9)(20.8)(216.5)(186.4) UMAA 2,0561936472,896 (228.4)(32.2)(323.5)(170.4) UCD 1,810573862,253 (452.5)(14.3)(77.2)(173.3) UIUC 6424407661,847 (642.0)(88.0)(127.7)(153.9) UCSB 91173388652 (15.2)(43.3)(194.0)(54.3) Waterloo 7197212381 (71.0)(6.1)(26.5)(15.2) Total R&D 6,8691,0643,26511,198 2005-2008Spin-offs29402796 R&D$M per Spin-off (120.9)(116.6) R&D Expenditures, Total and Per Spin-off by University and Technology, 2005-2008 Data from 2005-2008 236.9 26.6

11. MBSCS&EEEPSTotal UWM 1,3851556552,195 (173.1)(31.0)(163.8)(129.1) UMAA 1,7901721,2313,193 (198.9)(28.7)(615.5)(187.8) UCD 1,396995432,038 (349.0)(24.8)(108.6)(156.8) UIUC 1,0232768212,120 (1023.0)(55.2)(136.8)(176.7) UCSB 14889322559 (24.7)(22.3)(161.0)(46.6) Waterloo 232165566963 (232.0)(10.3)(70.8)(38.5) Total Faculty 5974956413811068 2005-2008 Spin- offs 29402796 Faculty per Spin- off (153.3)(115.3) University and Technology Field, Number of Faculty and Spin-offs per Faculty Spin-offs 2005-2008; number of faculty 2006 except Waterloo which is 2011 20623.9

12. Decreasing as one moves toward origin Spin-offs per Faculty and R&D Expenditures ($ million) Waterloo UCSB

13. BMSCS&EEEPSTotal Total%Total%Total%Total% UWM7847.3%4314.0%1936.8%14035.7% UMAA3767.6%3839.5%1338.5%8851.1% UIUC887.5%4035.0%2458.3%7248.6% UCD2630.8%633.3%837.5%4032.5% UCSB128.3%1729.4%825.0%3721.6% Waterloo812.5%941.1%368.3%1383.6% Total16946.7%23818.1%10831.5%51530.3% University Spin-offs, Number and Percentage Licensed by University and Technology Category, 1957-2010

14. Discussion Inventor ownership proved to be conducive to entrepreneurship –University of Wisconsin had inventor ownership in the past Surprising differences in levels of entrepreneurship between study fields EE&CS appear far more efficient at generating startups EE&CS have no need for TLOs and are hard to police Some US universities are experimenting, e.g. North Carolina Express Agreement, for inventors wishing to establish their own firm.

15. Discussion (continued) TLOs would survive and even thrive as SERVICE organizations not bureaucratic control operations US needs to experiment with alternative models

16. Limitations and Further Research with Database Only one inventor-ownership university Role of student entrepreneurship can be explored in our database Extend to private universities and foreign universities

17. Thank you mfkenney@ucdavis.edu