Schooling and entrepreneurship: Evidence from a regression discontinuity design

Does an additional year of formal education affect the decision to become an entrepreneur? Using human capital theory as a conceptual lens, we explore three channels through which it might: productivity, certification, and health impacts. To test the mechanisms and uncover whether there are causal relationships between these variables, we exploit two exogenous changes to British compulsory schooling laws that generated sharp across-cohort differences in years of education. Using a fuzzy regression discontinuity design, we estimate that the reforms significantly reduced self-employment. We go on to explore which channels best explain this finding, and discuss implications for scholars and policymakers. • A Regression Discontinuity Design is used to obtain causal estimates of the relationship between high school education and adult entrepreneurship. • The study exploits as a natural experiment legal reforms to the national minimum school leaving age in Britain. • Estimates indicate a negative relationship between an additional year of schooling and engagement in self-employment. • The findings cast doubt on the notion that the additional schooling affected self-employment through productivity or credentialing channels. • Policy implications might be most applicable to developing countries where governments are exploring whether to raise school leaving ages.