Current Research


Evaluating the Effects of a Transit Fare Exemption for the Elderly in Brazil

This paper investigates how a policy that grants fare-free public transportation for the elderly affects the travel behavior of its beneficiaries. To identify the policy impacts, I explore the fact that eligibility for exemption is based on age thresholds that vary by gender and by city. Using data from household travel surveys, I estimate the causal effects of free-fare on travel behavior through a Regression Discontinuity Design. The results of my analysis indicate that eligibility for farefree public transportation increases transit ridership among beneficiaries in approximately 27.3%. However, I do not find any significant effects of the policy on mode substitution nor on trip characteristics.

Working Paper



Affirmative Action Expansion in Brazil: Effects on Access to Public Higher Education

With Mary Arends-Kuenning

This paper investigates how affirmative action policies for college admission improved the access of students from disadvantaged groups to public tertiary education in Brazil. We use a difference-in-differences design to explore the fact that between 2004 and 2012 several Brazilian federal universities implemented affirmative action policies at different points in time. Our results indicate that the policies were effective in increasing the shares of Negros and public high-school students in treated universities. However, the policies were not as effective for increasing the enrollment of low-income households. While the impacts of affirmative action policies were larger on prestigious programs, it was mostly negligible in the less competitive ones. Additionally, we observed that race-blind policies had no impact on increasing the participation of racial minorities in federal colleges; meanwhile race-conscious policies did achieve that outcome.

Working Paper



Why Should Developing Country Cities Reduce their Speed Limits? Evidence from São Paulo, Brazil

With Peter Christensen and Amanda Ang

Cities throughout the world are experimenting with more stringent speed limits in an effort to reduce road accidents. The effectiveness of these policies is of particular interest for developing world cities, where a disproportionate share of accident damages occur but also where extant congestion creates heightened concern about speed reduction. This paper empirically evaluates a set of policies that changed traffic speed limits and enhanced its enforcement in São Paulo, Brazil. We exploit the temporal and spatial heterogeneity of policy adoption to examine the effectiveness and impacts of enforcement using measurements of traffic accidents, traffic tickets issued by monitoring cameras, and a panel of repeat observations of real-time trip duration for a representative sample of travelers before and after the new regulations. Our results indicate that speed limit reductions reduced number of accidents by 28.9% on treated road segments while not affecting traffic volume. We find that camera-based enforcement augmented the effect of the speed limit reduction. We find that speed limit increases on major urban highways reduced trip times by 7.5% during off-peak periods, though we find no significant effects during peak hours. We estimate that the social benefits from the reduction in road accidents are at least 2.13 times larger than the costs of longer commutes. The benefits of accident reductions accrue largely to lower income pedestrians and motorcyclists, indicating that speed limit reductions have an important progressive welfare impact in developing country cities.

Working Paper