Essays in political economy

• Ano de defesa: 2012
• Autor(a) principal: Rizzi, Renata
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: Dívida externa; Dívida pública; Electoral system; Foreign debt; Grupos de interesse; Interest groups; Participação política; Political participation; Public debt; Sistema eleitoral
• Link de acesso: http://www.teses.usp.br/teses/disponiveis/12/12138/tde-05032013-195951/

This thesis is divided into three parts. The first one evaluates the institution of compulsory vote, providing new estimates for the effects of the obligation to vote on individuals. The identification strategy relies on the Brazilian dual voting system - voluntary and compulsory - the exposure being determined by the date of birth. Using RD and IV approaches and data from a self-collected survey, we find that the compulsory legislation leads to a significant increase in voter turnout. These changes are followed by a sizable increase in the probability that individuals will express preference for a political party, but not by an increase in political knowledge among the population. Moreover, we find that the first compulsory voting experience permanently affects individuals\' preferences. The second part of the thesis empirically analyses episodes of sovereign debt default. Some of the salient features of the theoretical literature on sovereign debt, including its prediction that almost all defaults should arise in \"Bad Times\", are at odds with the data: over 38% of defaults actually occur in \"Good Times\", as measured by an HP filter. We explore the specific characteristics of each type of default and present econometric evidence that failures to repay foreign debt in good times can, usually, be rationalized by three components: (i) changes in the political environment, (ii) hikes in global interest rates and (iii) instances in which good HP times actually take place under quite poor economic conditions. We also present some suggestive indications that the duration of the episodes does not vary substantially with the type of default that precedes them, but with the environment in which they occur, drawing some important implications for the understanding of economies\' post-default market access. The third part of the thesis looks at the issue of campaign contributions in exchange for political favors (the so called \"pay-to-play\" scheme). I proposes a simple game to model the incentives of political parties and firms from public-revenue-intensive sectors, and test the implications of this model using data on campaign contributions and public contracts from Brazil. The data confirms the pay-to-play hypothesis.