Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.

Author ORCID Identifier


Open Access Dissertation

Document Type


Degree Name

Doctor of Philosophy (PhD)

Degree Program


Year Degree Awarded


Month Degree Awarded


First Advisor

Arslan Razmi

Second Advisor

Arindrajit Dube

Third Advisor

Daniele Girardi

Fourth Advisor

Christian Rojas

Subject Categories

Labor Economics | Macroeconomics | Political Economy


Do exchange rate shocks have distributional consequences? Does employment respond to exchange rate shocks? Do political parties matter when it comes to governing cities? Each chapter of this dissertation attempts to answer one of these questions in the Brazilian context. In the first chapter, titled Large devaluations and inflation inequality: evidence from Brazil, I show that prices of tradable goods/lower-priced varieties increase significantly more than the prices of nontradables/higher-priced varieties. These relative price changes may lead to inflation inequality when household consumption baskets are different across the distribution of income. Using Cravino and Levchenko (2017)'s methodology, we show that inflation of poor households in Brazil was at least 11 percentage points higher than of the rich in the aftermath of the 2002 large devaluation. A detailed case study of the City of São Paulo estimates an inflation inequality ranging from 8 to 11 percentage points in the city. In the second chapter, titled Employment effects of real exchange rate shocks: evidence from Brazilian local labor markets, I use local labor markets data over the period 1995-2016 for the Brazilian economy to study how employment responds to real exchange rate shocks. Exploiting regional variation in the intensity of real exchange rate shocks across local labor markets, I find that total employment increases in the short and medium run after a devaluation of the export-weighted real exchange rate. Meanwhile, a devaluation of the import-weighted real exchange rate decreases employment only in the short run. These effects are explained by the responses of the tradable sector as I find no significant effect of both measures of exchange rate shock on the nontradable sector. Within the tradable sector, manufacturing employment responds positively to export-weighted real exchange rate shocks both in the short and medium run, while I find no effect of these shocks on employment in the primary sector. Import-weighted real exchange rate shocks negatively affect both manufacturing and primary employment, but only in the short run. In the third chapter, titled Partisanship and local fiscal policy: evidence from Brazilian cities and co-authored with Daniele Girardi, we study the role of partisanship in shaping local fiscal policy in Brazilian cities in the 2004-2016 period. Using a regression-discontinuity design, we find no effect of left-wing mayors on the size of the city government. We find a modest but robust positive effect of approximately 0.6 percentage points on the social expenditures share, which translates in a small (approximately 1 percent) increase in social expenditure per capita. The impact of left-wing mayors on social spending is stronger for lame-duck mayors and in cities receiving oil windfalls. These results suggest that Brazilian parties attempt to shape the allocation of municipal resources to favor their respective electoral bases but their ability to do so is severely limited by factors such as institutional constraints and re-election concerns.