Essay 1: “Self-employment experience and labor market integration of young immigrants: A field experiment in Sweden.” In many countries, immigrants turn to self-employment to enter the labor market, and policymakers often present it as a stepping stone to regular employment. However, credible evidence on whether self-employment improves employment prospects remains scarce. This paper tests the stepping-stone hypothesis using a correspondence experiment in which fictitious applications were randomly assigned to signal self-employment, wage employment, or unemployment. Applicants with self-employment experience were more than twice as likely to receive interview requests as otherwise identical unemployed applicants. Interview callback rates for self-employed and wage-employed were similar, suggesting that when wage employment is hard to find, self-employment can serve as an alternative route into regular employment for young immigrants.

Essay 2: “Equal Pay Laws and the Gender Earnings Gap: Evidence from Swedish Administrative Data.” (co-authored with Lina Aldén,  Anne Boschini and Samuel Mann). Gender-based equal pay laws are intended to narrow earnings disparities, yet evidence on their effectiveness remains mixed. We study Sweden’s 1979 Equal Pay Act using population-wide administrative data and a propensity-score matched event-study that compares women’s annual earnings to those of observably similar men within narrowly defined job cells. We find that the reform modestly reduced the within–job-cell gender earnings gap. The gains are concentrated among workers who change jobs, with little evidence of changes in relative earnings growth among job stayers. We also find that women’s job mobility declined relative to men’s over this period, and that women who moved were less likely than comparable men to enter male-dominated job cells, where relative earnings gains were largest. Taken together, the results suggest that equal pay laws can improve within-job pay equity, but that limited access to higher-paying job transitions may constrain their broader effects on gender earnings convergence.

 Essay 3: “Work Training for Disadvantaged Job Seekers: Evidence from Sweden.” Work integration social enterprises (WISEs) provide work training for job seekers with disabilities or long-term unemployment, groups who often struggle to reach the regular labor market. This study examines a unique Swedish work training program launched in 2022 as a nationwide reserved procurement mainly delivered by WISEs. Participation is not random; participants are matched to similar non-participants using linked population registers. The study uses a stacked regression model in a difference-in-differences framework to estimate the effect on positive labor income over a two-year period. The estimated intention-to-treat effect is negative and statistically significant: a gap of 7-10 percentage points relative to the matched control group persists over this period.