Conference on Entrepreneurship and Economic Growth

Supported by the Ewing Marion Kauffman Foundation
Manuel Adelino and David T. Robinson, Organizers
October 14-15, 2016

Do University Programs Promote Entrepreneurship?

Though many universities have been incorporating entrepreneurship-related courses and programs to their curriculums, relatively little is known about the impacts of such university initiatives. Yong Suk Lee and Chuck Eesley examine two fundamental questions that pertain to entrepreneurship education: Do university entrepreneurship programs affect the decision to become an entrepreneur, and how do startups by students perform?

If university education programs improve entrepreneurial ability, such programs could result in higher rates of entrepreneurship as well as better firm performance. However, if students tend to overestimate their entrepreneurial ability ex-ante, and university entrepreneurship programs enable students to better assess their own entrepreneurial ability, education programs could reduce the average entrepreneurship rate among students. Hence, ex-ante it is not clear how university entrepreneurship programs would affect the rate of entrepreneurship and the performance of newly created businesses.

This research empirically examines whether university entrepreneurship programs affect entrepreneurial choice and startup performance. The researchers use the Stanford University Innovation Survey, which asks about entrepreneurship-related activities of Stanford degree-holders, to examine the entrepreneurship consequences of Stanford University's two major entrepreneurship programs founded in the mid-1990s: the Stanford Technology Venture Program (STVP) at the Engineering School and the Center for Entrepreneurial Studies (CES) at the Business School. The OLS regressions indicate that program participation is positively and significantly related to various measures of entrepreneurial activity.

However, estimating the causal impact of entrepreneurship programs is challenging because program participation is not random — more entrepreneurial students are more likely to participate in the entrepreneurship courses offered by the university. The researchers utilize the fact that the initiatives were implemented at the school level, i.e., the business school and the engineering school, and that primarily students in each school were affected by the respective entrepreneurship programs. This enables them to use the introduction of each school's program as an instrument for program participation.

There is strong evidence of a first stage — the introduction of university entrepreneurship programs increases student participation primarily for students in the respective schools. The participation rate is substantially higher in the business school's entrepreneurship program than in the engineering school's program. Cross-enrollment into the business school's entrepreneurship program by students from the other schools is minimal, though less so for the engineering school's program. The researchers use a difference in difference framework to estimate the first stage and reduced form using students from the other schools as the control group. Inference hinges on the assumption that student characteristics do not change differentially within each school after the introduction of the entrepreneurship programs. They find no evidence that demographic characteristics, such as race, gender, and nationality, or parental entrepreneurship, an important determinant of entrepreneurship, change in each school after program introduction. However, as the figure indicates, there is no detectable differential trend in the entrepreneurship rate across the different schools after program introduction.

The 2SLS estimates indicate that the business school program decreases, or at most has no effect on entrepreneurship. The engineering school initiative has no effect on entrepreneurship. The negative effects from the business school program imply that entrepreneurship courses could be helping students to better identify whether they are of the entrepreneurial type or not, and ultimately might deter certain students from pursuing entrepreneurship.

This study also examines whether program participation affects the characteristics and performance of startups after graduation. In particular, it analyzes whether program participation affects the timing to founding after graduation, the probability of survival, revenue, number of employees, and whether the startup successfully exited through an IPO or acquisition. The main finding here is that the business school initiative seems to increase the probability of survival and startup revenue of surviving firms. Again, no significant impact on firm performance from the engineering school program is found.

Overall, the results imply that business school entrepreneurship programs may decrease the entrepreneurship rate but improve the quality of entrepreneurship in terms of survival and firm revenue. The programs enable students to better identify their potential as entrepreneurs, and ultimately results in better startup quality. The null effect of entrepreneurship programs on entrepreneurship is quite robust. However, the researchers note that the effects found here are average effects at the school level over several years. There still may have been specific teachers, courses, or programs that had significant influences on entrepreneurship.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us