Abstract
dc:description.abstractThis thesis consists of three chapters, which broadly connect workers’ occupational choices both to their wealth and to the impact they have on government-provided unemployment insurance. The first chapter documents differential post-displacement earnings losses dependent on wealth level. Earnings declines after job losses have been documented in a literature emanating from the seminal work of Jacobson et al. (1993). This chapter extends those results by documenting differences between high-wealth displaced workers and low-wealth displaced workers. Using data from the PSID, it is found that poorer workers have deeper and more persistent post-displacement earnings losses relative to their peers. The hypothesis that the results are connected to occupation switching – as documented by Huckfeldt (2022) – is tested. While occupation switching seems to be negatively correlated with wealth in the general population, this correlation loses statistical explanatory power when the sample is restricted to those displaced through forces deemed exogenous to their ability. Finally, participation in contingent work is connected to wealth. Lower-wealth workers are found to be more likely to participate in these shorter term work arrangements. The second chapter introduces occupational dynamics to the question of optimal unemployment insurance. Baily (1978) and Chetty (2006) frame the question of unemployment insurance as a trade-off between insurance and incentives; higher benefits offer better protection against income loss but weaken job-search incentives. This chapter extends the debate by introducing occupations and highlighting the risk created by heterogeneous post-displacement income losses, between workers who return to their original occupation and those who switch. Huckfeldt (2022) shows that occupational switchers – workers who lose their job and find a job in an occupation different from the one they held pre-displacement – suffer from larger and more prolonged post-displacement earnings losses when compared to occupation stayers. What is the optimal level of unemployment benefits when these heterogeneous post-displacement losses are introduced to a model of the labor market? A directed search-and-matching model is constructed and calibrated to the U.S. economy to answer this question. The model includes heterogeneous losses through occupation switching, resulting in an increased variance of outcomes for displaced workers. The result is an optimal government benefit replacement rate of 16.9% when adjusted for take-up rate, or 46.2% when the adjustment is removed. This is higher than the 15.0%, or 41.0%, representing the current U.S. policy. The final chapter extends the second chapter to assess the impact of increased prevalence of gig employment. Over the recent years, we have seen an increase in the share of contingent employment in the U.S. and other countries. Moreover, the way in which gig jobs are found has been altered through technological improvement. Electronically mediated employment is on the rise – workers can use phone applications to find jobs with companies such as Uber, Lyft, and Doordash. Then, if workers have access to these easy-to-find jobs, does this impact the optimal provision of unemployment insurance? A search-and-matching model is constructed to include gig jobs. These are constructed to be easily-available as well as offering non-pecuniary benefits representing the flexibility that workers in these jobs value. On the other hand, they offer a lower wage rate than traditional employment on average, they are less stable, and offer no means of human capital appreciation over time. The optimal policy in this scenario is found to give a take-up-adjusted replacement rate of 15.7%, 1.2 percentage points lower than the counterfactual which excludes gig employment. Thus, the availability of electronically mediated gig jobs does represent a market-based partial substitute to publicly-funded unemployment insurance programs.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ellul, Christian
Subjects
dc:subject × 3Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Ellul, Christian. (2026). Occupational Choices and Unemployment Insurance. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/282425.