Judicial Subversion: The Effects of Political Power on Court Outcomes. (with Henrik Sigstad). [Latest version: May 2020]
Are politicians in power treated more leniently in court? We show that candidates in Brazilian local elections charged with corruption are 43 percent less likely to be convicted if they narrowly win the election. There are small differences in the lawyers representing winners and losers, the effect is larger for members of powerful parties, and judges convicting mayors are more likely to be promoted by seniority than by merit. This favoritism could lead to an adverse selection of politicians in elected offices: Politicians charged with corruption are more likely to run for election and to become the next mayor.
Slave resistance, cultural transmission, and Brazil’s long-run economic development. [Latest version: February 2020]
I show that ethnic territories connected to slave resistance, called quilombos, have a robust positive relationship with local economic development in Brazil. To understand how quilombos can affect economic activity in the long run, I propose a new mechanism where initial religious beliefs and African iron-working and other high-valued skills are perpetuated in the long run through cultural-religious intergenerational transmission. First, I divide the Brazilian territory in virtual municipality cells of approximately 11 x 11 kilometers, which makes possible an extensive use of ﬁxed eﬀects, and show that cells with more quilombos have more economic activity proxied by nightlights. Second, in order to analyze the mechanisms through which quilombos can aﬀect economic development and improve identiﬁcation I employ a randomization inference approach with alternative spatial configurations of counterfactual quilombos. I then show that proximity to quilombos is related to more high-skilled and metal-related occupations and a wide array of cultural-religious outcomes, such as higher cultural activities, community trust, and collective action.
Estimating a Behavioral New Keynesian Model. (with Joaquim Andrade and Pedro Cordeiro). [Latest version: December 2019]
This paper analyzes identiﬁcation issues of a behavorial New Keynesian model and estimates it using likelihood-based and limited-information methods with identiﬁcation-robust conﬁdence sets. The model presents some of the same diﬃculties that exist in simple benchmark DSGE models, but the analytical solution is able to indicate in what conditions the cognitive discounting parameter (attention to the future) can be identiﬁed and the robust estimation methods is able to conﬁrm its importance for explaining the proposed behavioral model.
Work in progress
Geography, slavery, and income in Brazilian municipalities in the 1870s: A spatial equilibrium approach. (with Eustáquio Reis).
Who ensures free and fair elections? Judicial bias in Brazilian electoral courts. (with Moya Chin and Henrik Sigstad).
The long-run effects of the Underground Railroad. (with Felipe Valencia).
Geographic cross-sectional fiscal multipliers over the business cycle. (with Ursula Mello).
Market-wide Disclosure of Teachers’ Quality and the Labor Markets for Teachers. (with Diana Moreira and Raissa Fabregas).
Acesso à terra, escolha ocupacional e o diferencial de produtividade agrícola entre pequenos produtores. 2016. In: J.E.R. Vieira Filho and J.G. Gasques, ed., Agricultura, transformação produtiva e sustentabilidade. Brasília: IPEA.