Economic and behavioral effects of safety net programs

Abstract

This thesis contains three independent but interrelated articles. Essentially, in this work, I empirically investigate the economic and behavioral effects of safety net programs. To do so, I use econometrics techniques allowing for causal inference, combined with high-quality administrative data sets. Below follows a brief description of the three articles that make up this thesis. I start by studying the behavioral effects of a large-scale Conditional Cash Transfer program, the Brazilian Bolsa Família (BF), on voting outcomes. Using a unique database on BF beneficiaries and where they vote, I explore random variation in program coverage among polling stations, a highly disaggregated level of observation with fewer than 400 registered voters. The findings indicate that the cash transfers positively affected voter turnout. I also find a positive effect on the support for the incumbent party that implemented and expanded the program. This positive effect comes from more people participating in the election but also from voters switching their choices, conditional on turnout. The electoral rewards to the incumbent candidate are mostly led by those beneficiaries who entered more recently in the program and the amount of money transferred matters for voting behavior. In the second paper, we examine the effectiveness of cash assistance targeted to disadvantaged youth. We exploit an exogenous variation in the provision of cash transfers from BF Program in Brazil to credibly identify how an additional year of exposure at the critical age of 18 impacts on educational, labor market, and economic self-sufficiency outcomes. We do not find evidence of significant effects of additional exposure to the program on educational attainment and economic self-sufficiency. However, we observe a small (but still positive) impact on school enrollment, which is mostly driven by male beneficiaries. We also find effects on formal labor supply only for men. For them, we show that one additional exposure to the program decreases the probability of working in the formal sector by 5.38 percentage points during the extra year of exposure. Five years later, this pattern reverses to an increase in participation in the formal labor force. Lastly, the third paper studies the effects of a maternity leave extension on labor market outcomes of women in Brazil, using detailed information on workers and firms in the formal labor market. Taking advantage of the exact leave taking dates and the staggered implementation of the extended leave policy across firms, our analysis compares outcomes within firms before and after the eligibility cutoff. While eligible women could have extended their leave period by 50% (from 120 to 180 days), take up only increases by 13 p.p. Also, employment effects are confined to the maternity leave extension spell, with no permanent effects on employment in the long run. Taken together, our findings indicate that this policy privileges a selected group of workers, while it is not able to retain them in the workforce.