Working Papers
Dropping Out of Higher Education after Trade Liberalization
with Emanuel Ornelas
Presented at the University of Cambridge - PhD Empirical Micro Workshop and the Trade and Spatial Economics Reading Group (2024); Latin American Economic Association (LACEA-LAMES) meeting in Lima, Peru (2022); the Australasian Trade Workshop (2022);
This paper investigates the effects of Brazil's 1991 trade liberalization on dropout rates from higher education. We use a shift-share instrument combining tariff cuts and employment across industries in a region to measure local economy exposure to the trade shock. We find that regions more exposed to the exogenous tariff cuts experienced a permanent relative increase in dropout rates of 7% twenty years after the trade reform. Moreover, we compute different regional tariff shock measures for adult and young workers to separate income and substitution effects from the trade shock. We find that the net effect is mainly explained by the impact on adult workers, suggesting that the income effect more than offsets the substitution effect in determining the impact of tariff cuts.
Trade Liberalization and Intergenerational Mobility
Presented at the University of Cambridge - Trade and Spatial Economics Reading Group (2024);
This paper examines the impact of Brazil's 1991 trade liberalization on intergenerational mobility (IGM). To measure the local economy's exposure to the trade shock, we employ a shift-share instrument that combines tariff cuts and employment across industries within a region. Our findings indicate that regions with higher exposure to the exogenous tariff reductions experienced a permanent relative decline in upward occupational mobility and a long-term decrease in both absolute income and relative educational mobility compared to regions with less pronounced effects. Additionally, we analyze IGM measures separately for son-father and daughter-mother pairs to account for gender heterogeneity, where daughters suffer higher reductions in income and educational mobility. We find that the overall effect is primarily driven by negative consequences in the labor market, characterized by an increase in the proportion of low-skilled occupations and a decrease in earnings levels following the trade shock.
Trade Shock and Slums: Changing Rural-urban Migration Opportunities (submitted)
with Luciene Pereira
Presented at the World Congress of the Econometric Society in Seoul (2025); the Latin American Economic Association (LACEA-LAMES) meeting in Recife (2025)
This paper studies the effects of the China trade shock on slum expansion in Brazil during the boom in bilateral trade from 2000 to 2010. Using a shiftshare IV design, which combines the rise in trade value with baseline industry composition across local labor markets for identification, we find that regions connected with greater exposure to export demand experienced relatively lower growth in slums. To explain this result, since many slum dwellers migrated from rural areas concentrated in agricultural and extractive sectors —industries responsible for the largest share of Brazil’s export growth to China— we disentangle the local effects from the migratory component of the economic shock using pre-shock migrant networks. Our findings indicate that increased export demand through migrant networks helped curb slum expansion by redirecting internal in-migration from urban areas facing greater import competition to regions concentrated in the extractive sector and by reducing out-migration from agricultural areas. Furthermore, our mechanism analysis reveals that the slower slum expansion was associated with a relative increase in average earnings for workers in non-tradable sectors and a shift in the workforce toward agricultural and extractive industries in areas more exposed to export demand.
Changing occupations: the effect of trade liberalization on family structure and domestic violence (submitted)
It is well understood that family dynamics affect the economic participation of men and women. But how do economic shocks affect family structures and domestic violence? Using the 1991 Brazilian trade liberalization as an unexpected change in local economic conditions, I find that areas more exposed to tariff cuts experienced a relative decrease in the share of young women who are married and are mothers, and an increase in the share of single women. These effects are particularly evident among women at their early 20s to mid-30s. I disaggregate the trade shock into male- and female-specific labor demand components and find that these effects were predominantly driven by tariff cuts that affected male employment. Additionally, there is a reduction in female domestic homicides. These results are motivated by the changes in occupational choices of men and women. Although gender differentials in wages and employment have consistently reduced, young men adapted to the trade-induced shock by transitioning into self-employment, while young women permanently left the labor market to engage in unpaid work.