Mayors in the Dock: Judicial Responses to Local Corruption in Brazil

This dissertation proposes an explanation of the variation of judicial responses to political corruption. It asks why some judicial systems are more active than others in punishing corrupt elected officials – i.e., why legal accountability varies across polities and over time. It examines these questions in a subnational comparative research project on Brazil. Particularly, this dissertation analyze how and why the judicial systems of the states of Rio Grande do Sul, Minas Gerais, and Bahia generated so different conviction rates of mayors while enforcing identical national laws on mayoral corruption. Relying on publicly available data on convictions, archival research, in-depth interviews with seventy-five members of the legal community, on-site visits to courts, prosecutors’ offices, and auditing agencies, and review of secondary sources, the findings of this dissertation go beyond the traditional emphasis on judicial independence to address conviction rates in corruption cases. The proposed argument emphasizes two pairs of variables. First, a relatively high level of institutional autonomy from the elected branches of government is necessary not only for the judiciary, but also for other institutions of the system of justice to generate sustained legal accountability results. That is, institutions responsible for case adjudication as well as those responsible for prosecuting, investigating and detecting irregularities require such autonomy to work effectively. This autonomy, in turn, rests primarily on the degree of political pluralism of the political system. Hegemonic or monopolistic political systems inhibit the workings of accountability institutions, whereas plural ones allow greater latitude to them. Second, higher levels of inter-institutional coordination among legal accountability institutions intensify judicial responses to corruption. When courts, prosecutors’ offices and auditing agencies streamline their work, conviction rates are expected to increase. This coordination, though, relies less on formal features of those institutions and more on legal mobilization, or the activism of actors working inside each of these agencies, who build bridges among otherwise isolated institutions. These variables, finally, interact with each other to generate four basic types of legal accountability performances. In increasing order of the intensity of judicial responses to corruption they yield, these types are: constrained isolation, constrained coordination, fragmented autonomy, and coordinated autonomy.