The fountain of knowledge: an epistemological perspective on the growth of U.S. SBIR-funded firms

This premise is reflected in the peer review process used to evaluate applications for SBIR awards. For details of the process, see: https://​www.​nap.​edu/​catalog/​11989/​an-assessment-of-the-sbir-program. We use information on SBIR awards in the empirical analysis below, but application review scores are confidential for each project.

The origin of this paper traces to our earlier writings on sources of knowledge and entrepreneurial behavior. See, for example, Audretsch and Link (2019).

Baron et al. (1996, 529) interpreted, from their background as eminent sociologists, Stinchcombe’s pioneering reasoning in the follow way: “… founding conditions become imprinted on organizations and mold their subsequent development.”

We have written about the institutional background of the SBIR program many times so duplication of text is inevitable. For example, see Link (2014).

See, https://​www.​presidency.​ucsb.​edu/​documents/​industrial-innovation-initiatives-message-the-congress-administration-actions-and

Purpose (3) was changed in the 1992 reauthorization of the program as: “[To] foster and encourage participation in innovation and entrepreneurship by women and socially or economically disadvantaged persons.” See Link and van Hasselt (2019) on gender and the patenting performance of SBIR-funded firms.

There are conditions under which an agency may increase Phase I and Phase II awards by as much as 50%.

The NRC is the operating arm of the National Academies of Science, Engineering, and Medicine. The Council was created in 1916 by President Woodrow Wilson to “bring into cooperation government, educational, industrial, and other research organizations with the object of encouraging the investigation of natural phenomena, and increased use of scientific research in the development of American industries, the employment of scientific methods in strengthening the national defense, and such other applications of science as will promote the national security and welfare.” See, http://​www.​nasonline.​org/​about-nas/​history/​archives/​milestones-in-NAS-history/​organization-of-the-nrc.​html

The survey instrument used in the 2005 survey is not consistent in its use of the word company and the word firm. When the survey question used the word company, we changed it to the word firm for consistency in the text of this paper.

The survey instrument does not define the growth metric of interest, but the juxtaposition of this survey question to other questions on the survey imply that growth refers to employment growth.

The project-specific and program-specific questions are the same for each of the five agencies that participated in the 2005 survey. See, for example, Appendix A in the National Research Council’s final report to Congress: https://​www.​nap.​edu/​catalog/​11989/​an-assessment-of-the-sbir-program.

The NRC survey about firm growth had categorical responses to the question about growth. The four categories of responses are “less than 25%, 25% to 50%, 51% to 75%, and more than 75%.” Responses to this survey question follow a uniform distribution. Note that the mean value of Growth in Table 2 is about 50%. We imputed a mean value to each category: 12%, 37.5%, 63.5%, and 88.” This was done for the ease of interpreting the regression results in Table 3. However, if one estimated our reduced form model using an ordered probit specification, the results are identical to the values in Table 3 in terms of algebraic signs and levels of significance. These results, and the calculated marginal effects, are available from the authors on request. Finally, the implications from the regression results in Table 3, using data from all funding agencies and controlling for fixed agency effects, are the same if the underlying model was estimated separately by funding agency. These results are available from the authors on request.