Browsing by Author "Kojola, Sami"

This thesis studies possible peer effects in entrepreneurship. Peer effects encompass any effects that membership in a given social group can enact on a person. Peer effects have been studied in many fields of human action, including educational attainment, worker productivity and substance use of minors. Peer effects have a special interest as a peer effects can produce social multipliers. If people react to actions taken by their peers, even small variations in individual behavior can cause large variations at the aggregate level. If peer effects in entrepreneurship are persistent, they would offer a possible explanation for the observation in previous research, that areas with large number of small firms produce an abnormally high level of entrepreneurship. This is interesting for regional developement, as higher levels of entrepreneurship have been linked to stronger subsequent economic and employment growth. I use Finnish employer-employee records to study, if particularly entrepreneurial peer groups produce more entrepreneurs. I have two environments in which I study the effects: workplaces and colleges. Because of the Finnish two tier system of colleges the college population is split in two, students of universities of applied science and students of traditional universities. All estimations are done with a discrete hazard rate model, that estimates if the probability of becoming an entrepreneur is higher in groups with other entrepreneurs. My results give some support to the hypothesis that an entrepreneurial social environment raises the number of entrepreneurs. However, all results are supportive of the hypothesis. Although there is a positive correlation between coworker entrepreneurial experience and the propensity to become an entrepreneur, the models detect a certain level of unobserved homogeneity within the coworker groups, that could explain at least part of the correlation. In the universities of applied science there is also a positive correlation between classmates setting up firms. However, the study of students of traditional universities does not produce statistically significant results.