Employment Growth from Public Support of Innovation in Small Firms

8 Conclusions

103 8 Conclusions The goals of the Small Business Innovation Research (SBIR) program, set out by Congress in the Small Business Innovation Development Act of 1982, did not include the growth of employment. In Chapter 2, we reviewed the legislated goals for the program, both in the 1982 Act and in subsequent reauthorizations. The goals include the stimulation of technological innovation, the use of small businesses to meet federal R&D needs, the encouragement of participation in technological innovation by disadvantaged persons, minorities, and women, and the increase in private sector commercialization of innovations derived from federal R&D. Although employment growth could of course be a concomitant of successful attainment of the legislated goals for the SBIR program, such growth has never been a stated objective of the SBIR program. In this book, we analyzed, albeit in an exploratory manner within an original empirical framework, a rich and unique set of data collected by the NRC to investigate whether such public support of innovation has in fact stimulated employment growth for the firms that have received SBIR awards. Although SBIR projects are commercially successful roughly half of the time (Link and Scott 2010), possibly in many cases the employment growth that would accompany the sales from successful projects may not be observed in the firms performing those projects, but instead will occur elsewhere in the economy. Conceivably, the employment growth to support sales from technologies developed with SBIR funds will be realized in other firms that purchase or license the technology or have manufacturing agreements with the firm that performed the SBIR program research. The NRC data do not allow us to document the employment growth at firms other than the award recipients, but we are able to ask if employment growth for the award recipients is less when the recipients have made commercial agreements allowing other firms to use the technologies created by the SBIR awards. Chapter 4 addressed the question of whether the technology created in an SBIR Phase II project has itself typically added substantially to 104 Link and Scott the employment in the firm conducting the project. The evidence shows that typically the answer is no—on average the SBIR projects add only a small number of employees. The average number of employees retained because of the technology developed in SBIR projects is only one or two for the random samples of the SBIR projects for the five agencies —the DoD, NIH, NASA, DOE, and NSF. However, the range for employees retained because of the technology developed in the Phase II project is wide. Exploiting that range of outcomes, the data reveal that for many types of commercial agreements (regarding the Phase II technology) between the SBIR firm and other firms or investors, the number of employees retained is less, ceteris paribus, supporting the belief that the jobs created when the SBIR Phase II technology is used in production actually occur at other firms that have entered into agreements to use the technology. Chapters 5 and 6 turned from the employment effects directly associated with the use of the SBIR project’s technology to the overall trajectory of the SBIR firm’s employment in order to address the dynamic question of whether the SBIR award helps the recipient firm overcome the initial hurdles that small, entrepreneurial firms face, thus facilitating a more permanent and possibly longer-term employment growth. The previous literature on entrepreneurial firms has been unable to address the question because of the lack of a comparison group to which the growth of firms receiving SBIR awards could be compared. We developed in Chapter 5 a model of the growth of small entrepreneurial firms, and with this model we estimated the growth for each of the SBIR award recipients using only information about each firm prior to when it received the SBIR Phase II award in the sample. That model is then used to predict the award recipient’s ultimate employment if it had not received the Phase II award. That predicted employment absent the award then provides a baseline for comparison to address the more dynamic question. The answer, as seen in Chapter 5, is that for four of the five agencies ’ samples of projects, there are substantial average employment gains that are due to the SBIR award. However, these average gains, despite being substantial in absolute amount, are not statistically significant . Nonetheless, there are a number of individual cases where the actual employment for the firm exceeds the predicted employment by a statistically significant amount...