Raphael Guinâncio Bruce
Doctorate – Essays in Development Economics
Advisor: Prof. Dr. Marcos Yamada Nakaguma
Comission: Profs. Drs. Ricardo de Abreu Madeira, Fernando de Magalhães Papaterra Limongi and Rudi Rocha de Castro
Class: 217, FEA-5
This doctoral dissertation is comprised of three separate research pieces divided by chapters. In the first chapter, I answer the following question: Do people acquire more information when they are encouraged to participate in elections? I present empirical evidence on the effects of compulsory voting laws on the consumption of TV news. In Brazil, literate citizens over the age of eighteen are subject to a number of legal penalties if they fail to turn out on election day. This allows me to identify the causal effect of being under a compulsory voting regime on information acquisition. I find that compulsory voting has a significant and substantial positive impact on the probability of an individual to watch Brazil’s main TV newscast. Additionally, the effect is stronger for individuals with neutral evaluations of the incumbent president. My regression discontinuity results are restricted to young voters who just turned eighteen and are robust to different polynomials and bandwidth lengths. In the second chapter, I investigate if there’s persistence in turnout in the context of the world’s largest democracy under compulsory voting, Brazil. Using individual panel data for all registered voters in Brazil for five different elections I explore the age cutoffs created by the compulsory voting law in order to provide causal estimates of persistence in turnout. Employing a Fuzzy RD approach, I estimate the effect of previous on current turnout, using the exposure to abstention fines as an instrument. Using this strategy, I am able to identify that voting on an election causes an individual to become 46% more likely to vote on the next election. I show, therefore, that there is a high degree of persistence in turn out due to compulsory voting. In the third and last chapter, I investigate if longer school hours have an impact on the homicide rates of teenagers. I show how the full-day high school policy in the state of Pernambuco, Brazil, had a negative impact on the homicide rates among youth in that state. Using an event-study design, I exploit the temporal and spatial variation in the introduction of the program to show that this reduction was focused on individuals in the school-age range of high school students. I find no impact of the policy on homicides for similar age ranges. I also don’t find any impact on mortality rates by other causes. Overall, these results suggest that the introduction of the policy played a significant role in the reduction of lethal crime rates in that state.
*Abstract provided by the author