Research
Working Papers
Direct and Spillover Effects of Generative AI: Evidence from the Judiciary
Payroll Tax, Employment and Labor Market Concentration (with Raphael Corbi and Renata Narita) Revise and resubmit, Labour Economics. IZA Discussion Paper 18838 · Winner of the Applied Microeconomics prize, Brazilian Econometric Society, 2024.
How much employment can be generated by decreasing payroll taxes? We examine this question by exploring the staggered rollout of a large payroll tax reform in Brazil. Using administrative matched employer-employee data, we find an increase of 5 percent on employment due to both firm growth and firm entry, no impact on wages and a significant increase in profits. Moreover, employment effects are driven by less concentrated labor markets, consistent with predictions from an oligopsony model.
The Chosen, the Cool and the Central: Peer Education and Teen Pregnancy in Brazil (with Emily Breza, Eliana La Ferrara, Pedro Rosa Dias and Victor Orozco)
We combine insights from network theory and social psychology to target a program aimed at reducing teen pregnancies among Brazilian high school students. The program trains a group of peer educators in each school, who then disseminate information among other students. We propose a model of diffusion and adoption under school-level externalities, extending the framework of Banerjee et al. (2019) to allow certain nodes to have differential effectiveness in the transmission of information; notably, "popular" students may be more influential than others. We conduct an RCT with 88 schools to test the effectiveness of three strategies for selecting peer educators: (i) selection by school administrators, (ii) most connected students, as measured by eigenvector centrality, and (iii) most popular students, based on survey responses. We find that (ii) and (iii) significantly increase contraceptive use, reduce teen pregnancy, and increase intended post-secondary school enrolment, relative to a control group. By estimating the model parameters we show that pure centrality is particularly effective in shaping the spread of information and knowledge, while popularity plays an important role in converting those knowledge gains into the adoption of contraception.
The Electoral Effects of Conditional Cash Transfers
The literature on the electoral effects of social policy expenditures reports generally positive impacts on incumbent electoral support. Studies such as Imai et al. (2020) suggest these results may be confined to discretionary spending, whereas programmatic expenditures are less likely to increase political support. In this paper, I exploit a quasi-exogenous shock to access to Bolsa Família, the conditional cash transfer program in Brazil, to assess whether expanded access increased support for the incumbent, and how this interacts with local political alignment. Despite a large shock that increased the number of CCT recipients by around 13%, estimated electoral gains are close to zero and statistically insignificant. I then analyze a second shock that excluded more than 700,000 families from the program in the first semester of 2013 through eligibility checks. The exclusions are correlated with higher shares of blank votes and lower support for the Workers' Party, providing some evidence of asymmetric electoral effects.
Political Brokers and the Allocation of Social Benefits: The Case of Bolsa Família (with Caio Pedro Castro, Thea How Choon and Giovanna Marcolongo)