Working Papers

Self-employment dynamics and Unemployment risk

I study the role of labour market outcomes and unemployment risk on self-employment dynamics. Entry to self-employment from paid work is lower during downturns as well as in industries that are experiencing higher increases in unemployment. In line with empirical evidence highlighting the uncertainty and risk associated with business creation, I build a model where transitions to entrepreneurship are risky. Failure leads to the individual being unemployed. Risky business creation stresses the importance of hiring and the job finding rate, in contrast with previous literature emphasizing the importance of displacement and job separations. I structurally estimate my model to fit the US economy during the Great Recession and decompose the change in unemployment risk for workers into these two components: I find that the fall of the job finding rate dominates the rise of the job separation rate and drives the dynamics of entry into self-employment. Higher unemployment risk can explain the 16% decrease of the self-employment rate between 2007 and 2011, and my findings suggest that it also leads to significantly worse entrepreneurial quality.


Unemployment Insurance and Self-employment during the Great Recession

I study the impact of large changes in the duration of unemployment benefits during the Great Recession in the United States. At the micro-level, I show that unemployment benefits extensions were on average not associated with lower entry to self-employment. However, this result masks substantial heterogeneity among the self-employed: Those who work less than 20 hours per week on their business were 30% less likely to enter following eligibility to extended benefits, consistent with entry by necessity in a slack labour market. In addition, I also show that individuals about to exhaust their benefits were less likely to enter if they had expected a future increase in the duration of unemployment insurance. At the macro-level, while there was a large variation in the number of UI weeks available across states and over time, I find that the extension and termination of benefits did not affect state-level self-employment rates.

[draft coming soon]

Work in Progress

Housing Wealth and Business Dynamism

Productivity and Misallocation in the Caucasus and Central Asia Region (with Benedicte Baduel and Gaelle Pierre)