Improving labor market outcomes is central to enhancing individual well-being. Two key strategies available to individuals are investing in skills and migrating to more dynamic labor markets. In sub-Saharan Africa, where a rapidly growing population confronts limited job creation and predominantly informal, precarious employment, it is crucial to assess how policy can support these strategies. Training programs may better align workers' skills with labor market needs, though evidence on their effectiveness remains mixed. Moreover, persistent regional income disparities suggest potential for welfare-improving migration, yet many individuals remain immobile. This dissertation investigates the effectiveness of interventions designed to promote these two labor market strategies through four randomized controlled trials conducted in Ghana and Uganda. Chapters 1 and 2 evaluate two skills training programs. While neither program significantly increased employment, they enhanced - at least partially - job quality, occupational matching and mental health. The findings highlight the importance of aligning the design and goals of interventions with labor market realities. Both chapters also show that stakeholders hold overly optimistic beliefs about program effectiveness and only selectively update them in response to the evaluation results - highlighting an important barrier to evidence-based policy learning. Chapter 3 studies the role of internal psychological constraints, specifically self-efficacy, in shaping skills investments. A random subset of women invited to participate in the training program evaluated in Chapter 1 received a testimonial-based campaign designed to boost self-efficacy. The campaign successfully increased self-efficacy and, in turn, improved training completion among women who started the training. Effects are strongest among women with a low initial propensity to complete the training. These findings suggest that addressing both internal and external constraints can enhance the impact of traditional development interventions. Chapter 4 turns to the second strategy: migration. The urban, working-age study population was found to substantially overestimate regional income differentials. Providing accurate income information affected destination preferences and increased migration intentions among those who had underestimated potential income gains but reduced them among those who overestimated. While peer discussions about migration increased, actual migration behavior did not change, suggesting persistent structural and social barriers. Together, the findings contribute to the literature on skills training and behavioral public policy by providing new evidence on the effectiveness of labor market and migration interventions, and by highlighting the central role of beliefs in shaping behavior. They also offer actionable insights for policies aligned with the Sustainable Development Goals (SDGs) on education (SDG 4), decent work (SDG 8), and reduced inequalities (SDG 10).

Georg-August-Universität

DOI: 10.53846/goediss-11393