Abstract
University spin-offs are defined as firms founded by university employees. Using a large database on venture-backed start-up companies, I describe the characteristics of university spin-offs and investigate whether they perform differently than other firms. I find that venture-backed university spin-offs are concentrated in the biotechnology and information technology industries. Moreover, a spin-off tends to stay close to the university, suggesting that technology transfer through spin-offs is largely a local phenomenon. Multivariate regression analyses show that university spin-offs have a higher survival rate but are not significantly different from other start-ups in terms of the amount of venture capital raised, the probability of completing an initial public offering (IPO), the probability of making a profit, or the size of employment.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
It is well known that the popular Internet search engines Yahoo! and Google both grew out of Stanford. However, they were founded by students rather than university employees and thus, by my definition, are not considered university spin-offs in this paper.
Lowe and Gonzalez-Brambila (2007) is perhaps the only study analyzing a relatively large database that contains 150 academic entrepreneurs in 15 academic institutions. They refer to academic firm founders as “faculty entrepreneurs,” and their study focuses on such entrepreneurs rather than their companies.
See Pirnay et al. (2003) for a typology of university spin-offs.
Data on companies founded to exploit MIT intellectual property during 1980–1996 show that a university inventor was the lead entrepreneur in about only one-third of the companies (Shane 2004, pp 6–7).
VentureOne defines a venture capital firm as “a professional, institutional venture capital limited partnership that generally manages over $20 million in assets and invests in privately held companies” (VentureOne 2000). Once a company receives some investment from venture capital firms, it becomes a “venture-backed company” and enters the VentureOne database. VentureOne then tracks the company’s financing from all sources, including bank loans and IPO. While I do not count bank loans or money raised through an IPO as venture capital, I do include equity investment made by non-VC corporations or “angel investors” as venture capital in my calculations.
For a more detailed description of the VentureOne data, see Zhang (2007).
For VentureOne’s research methodology, see http://www.ventureone.com/ii/research.html (accessed on December 23, 2005).
As I will argue below, it is more appropriate to drop all the firms founded before 1990 in the regression analyses. So the Pearson’s χ2 tests are also conducted for the two subsamples excluding firms founded before 1990. The hypothesis that these two subsamples arise from random sampling is again rejected.
Gompers et al. (2005), who study spin-offs from public companies using data extracted from the same VentureOne database, try to supplement the data by searching for the missing founder information using alternative sources such as Google and Lexis–Nexis. However, I cannot do the same. When I obtained the data from VentureOne, the company was very much concerned about researchers’ practice of tracking down the founders. So they replaced all company names and founder names with identification numbers. This makes it impossible for me to complement the founder data using alternative sources like Gompers et al. did. Note that Gompers et al. search alternative data sources only to construct a more complete sample. It is not meant to and cannot solve the sample selection problem. Because information about successful entrepreneurs tends to be more easily available, their searches have the same sort of sample selection problem as the original VentureOne database does.
An academic entrepreneur’s specialty is not always identifiable using the VentureOne data. For example, a firm founder’s biographical sketch might read like this: “professor, Stanford University.” In this case, I left the “specialty” field blank. In other cases, more information is available, such as “professor, Department of Computer Science, Stanford University,” or simply “professor of physics, Stanford University,” which enables me to identify this person’s specialty.
To a great extent, what I choose to present in this section is determined by the availability of data. There are many other characteristics of academic entrepreneurs and their firms that would be interesting to explore. For example, do academic entrepreneurs enter the industry that is most closely related to their academic field? Do young professors have a higher tendency to become academic entrepreneurs than older ones? The VentureOne data are not suitable for answering these questions because the information is either incomplete or entirely unavailable.
For example, Herbert Boyer, cofounder of Genentech in 1976, remained a professor of biochemistry at UCSF until 1991; Richard Newton helped found a number of design technology companies including Cadence, Synopsys, and Simplex Solutions but never left UC Berkeley; Phillip Sharp, a cofounder of Biogen in 1978, is still a professor at MIT.
There are 23 entrepreneurs whose biographical information contains university names, but they were listed as “research assistants,” “Ph.D. students,” or “post-doc fellows” and did not hold formal job positions at universities. I excluded these founders from the group of academic entrepreneurs shown in Table 2. One could argue that these individuals should also be counted as academic entrepreneurs. However, it is very possible that many of the founders do not consider these as important job experiences and therefore do not include them in their biographical sketches. For example, there are simply too few post-docs in the sample, which seems to be a result of underreporting. Thus, leaving them out is better than including them as academic entrepreneurs because the latter definition could lead to more serious measurement errors.
Although the existing market demand for more effective drugs is obvious, biotechnology may have some other applications that are unknown today. If some of these applications are carried out in the future, it is likely that non-academic entrepreneurs, rather than university professors, will make it happen.
See, for example, Leland and Pyle (1977) for a formal discussion of the informational asymmetries between entrepreneurs and investors.
VentureOne data include foreign researchers who founded firms in the United States but do not include U.S. researchers who started businesses overseas. Thus it is impossible to measure the net flow of academic entrepreneurs between the United States and the rest of the world using the VentureOne data.
In corporate demography literature, a firm is considered as not surviving if it is acquired by another firm (see, for example, Carroll and Hannan 2000). Here I define “survival” more broadly and count acquired firms as surviving firms. An acquired firm, although it has lost its identity, is likely to retain its personnel and technology and continue to operate. From an economic point of view, it is still alive.
The VentureOne data only indicate whether a firm has ever made a profit; there is no information about the amount of the profit or loss.
Employment here refers to the employment level at the end of the sample period (fourth quarter of 2001). This comparison of employment is done only for firms that were still alive and privately held by the end of 2001. Employment information of other firms is not “current” because VentureOne would stop updating it if the firm was out of business, went public, or was acquired by another firm.
Some of the performance variables clearly measure desirable outcomes from a start-up’s (or its founder’s) perspective. For example, one can rather confidently assume that a start-up’s founder wants to build a surviving and profitable firm. However, raising the most VC may or may not be a start-up founder’s goal. After all, the founder has to give up a share of the firm in exchange for VC investment. In some cases, it might be a better strategy for the entrepreneur to rely more on alternative sources of capital. One can only say that all else being equal, start-ups receiving more VC tend to be the more successful ones.
For example, total VC investment in the U.S. amounted to $88.9 billion in 2000, compared to only $6.8 billion in 1995. VC investment in the late 1990s is often described as “too much money chasing too few good ideas.”
In all cases where the independent variable is a dummy variables, probit models are also estimated, which give qualitatively similar results as the logit models.
For survival analysis, hazard models are preferred over logit models. However, for most of the start-ups that were out of business, the VentureOne data indicate only that they no longer existed by the end of 2001 but do not specify an exact exit date. Thus it is impossible to construct the time-to-event variable to estimate standard hazard models. For similar reasons, logit models instead of hazard models are estimated for IPO analysis, as will be shown below.
The odds of survival are defined as the ratio of probability of surviving to the probability of not surviving. Under a logit model, it is simply \( e^{{\alpha {\text{ + }}x\beta }} \). Thus the ratio of these two odds (for non-spin-offs and spin-offs) is \( e^{{{{(}}x{_{1}} - x{_{2)}}\beta }} {{ = }}e^{{ - {{1*0}}{{.485}}}} {{ = 0}}{{.616}}\).
Results in Tables 7–10 together point to an interesting phenomenon: University spin-offs have a higher chance to survive despite relatively lower financial support from VC (the latter result being persistent in Tables 7–9 although in most cases statistically insignificant). This may also be explained by the incubatory support university spin-offs tend to receive from parent universities and other local organizations. In addition, this result is also consistent with the theory that the asset parsimony strategy and growing by bootstrapping often help a start-up to succeed (Hambrick and MacMillan 1984; Venkataraman 2003).
The measure of profitability used in this study might appear to be an imperfect one because a firm that loses money for many years does not necessarily make less profit over a longer period of time. For example, a biopharmaceutical company may lose a large amount of money for many years. However, once it sees profit, it can also continuously generate a large amount of profit for many years, which makes it more profitable than many other companies over its whole lifetime.
Economists have developed various strategies (such as the use of instrumental variables or regression discontinuity design) that rely on richer data and better research designs to correct this kind of endogeneity bias. However, the VentureOne data are not rich enough to allow for such sophisticated econometric analyses.
In principle, I can estimate Heckman-type sample selection models by adding a selection equation to explain what types of firms tend to have founder information available. And indeed I tried that. However, VentureOne data contain limited information on firm characteristics and most of such characteristics are discrete variables. Similarly, most of the control variables in the main equation are also discrete variables. As a result, the likelihood function of the sample selection model is not smooth and its maximum is difficult to find. This is particularly true for discrete performance outcome variables. I tried many different specifications for the selection equation and tried both probit and linear probability models for the main equation, but Stata could only complete the maximum-likelihood estimation for the simplest specifications of the equation for survival and IPO. For profitability, the estimation is impossible under any specifications. These exercises are not very informative about the magnitude of selection biases.
There is no lack of anecdotal accounts of start-up location choices. Through my personal communications with start-up founders, I find that quality of life, quality of labor pool, and availability of capital are the most frequently cited factors that affect the location of a start-up.
References
Agarwal, R., Echambadi, R., Franco, A. M., & Sarkar, M. B. (2004). Knowledge transfer through inheritance: Spin-out generation, development, and survival. Academy of Management Journal, 47, 501–522.
Biotechnology Industry Organization. (2004). Laboratories of innovation: State bioscience initiatives 2004. Washington, D.C.
Carroll, G. R., & Hannan, M. T. (2000). The demography of corporations and industries. Princeton, N.J.: Princeton University Press.
Chiesa, V., & Piccaluga, A. (2000). Exploitation and diffusion of public research: The case of academic spin-off companies in Italy. R&D Management, 30, 329–340.
Clarysse, B., Wright, M., Lockett, A., Mustar, P., & Knockaert, M. (2007). Academic spin-offs, formal technology transfer and capital raising. Industrial and Corporate Change, 16, 609–640.
Cochrane, J. H. (2005). The risk and return of venture capital. Journal of Financial Economics, 75, 3–52.
Cooper, A., & Folta, T. (2000). Entrepreneurship and high-technology clusters. In D. L. Sexton & H. Landstrom (Eds.), The Blackwell handbook of entrepreneurship. Malden, MA: Blackwell Business.
Di Gregorio, D., & Shane, S. (2003). Why do some universities generate more start-ups than others? Research Policy, 32, 209–227.
Druilhe, C., & Garnsey, E. (2004). Do academic spin-outs differ and does it matter? Journal of Technology Transfer, 29, 269–285.
Etzkowitz, H. (2002). MIT and the rise of entrepreneurial science. London: Routledge.
Feldman, M. (1994). The university and economic development: The case of Johns Hopkins university and Baltimore. Economic Development Quarterly, 8, 67–76.
Feldman, M., & Desrochers, P. (2003). Research universities and local economic development: Lessons from the history of the Johns Hopkins university. Industry and Innovation, 10, 5–24.
Gompers, P., & Lerner, J. (2000). Money chasing deals? The impact of fund inflows on private equity valuations. Journal of Financial Economics, 55, 281–325.
Gompers, P., Lerner, J., & Scharfstein, D. (2005). Entrepreneurial spawning: Public corporations and the genesis of new ventures, 1986–1999. Journal of Finance, 60, 577–614.
Hambrick, D., & MacMillan, I. (1984). Asset parsimony: Managing assets to manage profitability. Sloan Management Review, 25, 67–74.
Kaplan, S., Sensoy, B., & Strömberg, P. (2002). How well do venture capital databases reflect actual investments? Manuscript, University of Chicago.
Kenney, M. (1986a). Schumpeterian innovation and entrepreneurs in capitalism: A case study of the U.S. biotechnology industry. Research Policy, 15, 21–31.
Kenney, M. (1986b). Biotechnology: The university-industrial complex. New Haven, CT: Yale University Press.
Kenney, M., & Goe, W. R. (2004). The role of social embeddedness in professorial entrepreneurship: A comparison of electrical engineering and computer science at UC Berkeley and Stanford. Research Policy, 33, 691–707.
Klepper, S. (2001). Employee start-ups in high-tech industries. Industrial and Corporate Change, 10, 639–674.
Klepper, S., & Sleeper, S. D. (2005). Entry by spinoffs. Management Science, 51, 1291–1306.
Klofsten, M., & Jones-Evans, D. (2000). Comparing academic entrepreneurship in Europe—The case of Sweden and Ireland. Small Business Economics, 14, 299–309.
Leland, H. E., & Pyle, D. H. (1977). Informational asymmetries, financial structure, and financial intermediation. Journal of Finance, 32, 371–387.
Link, A. N., & Scott, J. T. (2005). Opening the ivory tower’s door: An analysis of the determinants of the formation of U.S. university spin-off companies. Research Policy, 35, 1106–1112.
Lockett, A., Siegel, D., Wright, M., & Ensley, M. D. (2005). The creation of spin-off firms at public research institutions: Managerial and policy implications. Research Policy, 34, 981–993.
Lockett, A., & Wright, M. (2005). Resources, capabilities, risk capital and the creation of university spin-out companies. Research Policy, 34, 1043–1057.
Lowe, R., & Gonzalez-Brambila, C. (2007). Faculty entrepreneurs and research productivity. Journal of Technology Transfer, 32, 173–194.
Lowe, R. A., & Ziedonis, A. A. (2006). Overoptimism and the performance of entrepreneurial firms. Management Science, 52, 173–186.
McQueen, D. H., &Wallmark, J. T. (1982). Spin-off companies from Chalmers University of technology. Technovation, 1, 305–315.
Nerkar, A., & Shane, S. (2003). When do start-ups that exploit patented academic knowledge survive? International Journal of Industrial Organization, 21, 1391–1410.
O’Shea, R. P., Allen, T. J., Chevalier, A., & Roche, F. (2005). Entrepreneurial orientation, technology transfer and spinoff performance of U.S. Universities. Research Policy, 34, 994–1009.
Pirnay, F., Surlemont, B., & Nlemvo, F. (2003). Toward a typology of university spin-offs. Small Business Economics, 21, 355–369.
Roberts, E. B. (1991). Entrepreneurs in high technology: Lessons from MIT and beyond. Oxford: Oxford University Press.
Rogers, E. M. (1986). The role of the research university in the spin-off of high technology companies. Technovation, 4, 169–181.
Rosenberg, N., & Nelson, R. R. (1994). American research universities and technical advance in industry. Research Policy, 23, 323–348.
Samson, K. J., & Gurdon, M. A. (1993). University scientists as entrepreneurs: A special case of technology transfer and high-tech venturing. Technovation, 13, 63–71.
Shane, S. (2004). Academic entrepreneurship: University spinoffs and wealth creation. Northampton, MA: Edward Elgar.
Shane, S., & Stuart, T. (2002). Organizational endowments and the performance of university start-ups. Management Science, 48, 154–170.
Siegel, D. S., Wright, M., & Lockett, A. (2007). The rise of entrepreneurial activity at universities: Organizational and societal implications. Industrial and Corporate Change, 16, 489–504.
Smilor, R. W., Gibson, D. V., & Dietrich, G. B. (1990). University spin-out companies: Technology start-ups from UT-Austin. Journal of Business Venturing, 5, 63–76.
Steffensen, M., Rogers, E. M., & Speakman, K. (2000). Spin-offs from research centers at a research university. Journal of Business Venturing, 15, 93–111.
Stuart, T. E., & Ding, W. W. (2006). When do scientists become entrepreneurs? The social structural antecedents of commercial activity in the academic life sciences. American Journal of Sociology, 112, 97–144.
Venkataraman, S. (2003). Entrepreneurship: Creating something new and of enduring value with very limited resources. In The portable MBA (4th ed.). New York, NY: Wiley.
VentureOne Corporation. (2000). The VentureOne venture capital sourcebook. San Francisco, California.
VentureOne Corporation. (2001). Venture capital industry report. San Francisco, California.
Wright, M., Lockett, A., Clarysse, B., & Binks, M. (2006). University spin-out companies and venture capital. Research Policy, 35, 481–501.
Wright, M., Vohora, A., & Lockett, A. (2004). The formation of high-tech university spinouts: The role of joint ventures and venture capital investors. Journal of Technology Transfer, 29, 287–310.
Zahra, S. A., Van de Velde, E., & Larraneta, B. (2007). Knowledge conversion capability and the performance of corporate and university spin-offs. Industrial and Corporate Change, 16, 569–608.
Zhang, J. (2003). High-tech start-ups and industry dynamics in Silicon Valley. San Francisco: Public Policy Institute of California.
Zhang, J. (2006). A study of academic entrepreneurs using venture capital data. Working Paper No. 2006.01, Public Policy Institute of California, San Francisco.
Zhang, J. (2007). Access to venture capital and the performance of start-ups in Silicon Valley. Economic Development Quarterly, 21, 124–147.
Zucker, L. G., Darby, M. R., & Torero, M. (2002). Labor mobility from academe to commerce. Journal of Labor Economics, 20, 629–660.
Author information
Authors and Affiliations
Corresponding author
Additional information
This paper has benefited from comments by Maryann Feldman, Amy Ickowitz, Donald Siegel, Xue Song, an anonymous referee, and participants at the Technology Transfer Society’s Conference at the Kauffman Foundation in Kansas City, MO, the 9th Uddevalla Symposium at George Mason University, Fairfax, VA, and the Conference on Technology Commercialization and Entrepreneurship at the Kauffman Foundation in Kansas City, MO.
Appendix: Geographic definition of industrial clusters
Appendix: Geographic definition of industrial clusters
Following the tradition established by regional institutions such as the Joint Venture: Silicon Valley Network, I define Silicon Valley as Santa Clara County and adjacent cities in Alameda, San Mateo, and Santa Cruz Counties.
City | Zip Code |
|---|---|
Santa Clara County | |
All cities | All zip codes |
Alameda County | |
Fremont | 94536–39, 94555 |
Newark | 94560 |
Union City | 94587 |
San Mateo County | |
Atherton | 94027 |
Belmont | 94002 |
East Palo Alto | 94303 |
Foster City | 94404 |
Menlo Park | 94025 |
Redwood City | 94061–65 |
San Carlos | 94070 |
San Mateo | 94400–03 |
Santa Cruz County | |
Scotts Valley | 95066–67 |
Other regions are more loosely defined using area codes.
Region | Area code |
|---|---|
San Francisco Bay Area | Silicon Valley, plus 408, 415, 510, 650, and 925 if not already in Silicon Valley |
Boston | 508, 617, 781, 978 |
New York | 201, 212, 347, 516, 646, 718, 732, 845, 908, 914, 917, 973 |
Seattle | 206, 253, 360, 425 |
Washington, D.C. | 202, 240, 301, 571, 703 |
Rights and permissions
About this article
Cite this article
Zhang, J. The performance of university spin-offs: an exploratory analysis using venture capital data. J Technol Transf 34, 255–285 (2009). https://doi.org/10.1007/s10961-008-9088-9
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10961-008-9088-9