The developing world has provided numerous examples of political elites diverting funds intended for basic public services such as health, schools, and roads. According to Transparency International, Suharto embezzled anywhere between $15-35 billion from Indonesia. Ferdinand Marcos in the Philippines, Mobutu in Zaire and Abacha in Nigeria may have embezzled up to $5 billion each. In the data on Brazil used in the project, there is evidence of corruption in 78 percent of all municipalities, where on average almost 9 percent of federal funds was diverted. This translates into losses of approximately $800 millions per year. Thus, the welfare gains to reducing corruption are potentially large. In spite of this, there is limited understanding of what determines corruption and of how to design policies that can effectively reduce it. The objective of this research is to provide a framework for understanding the decisions of politicians to engage in corrupt activities. This project develops and estimates a structural model of a politician's decisions to provide public goods and engage in corruption over the span of her political career. The proposed research makes two main contributions to the existing literature. First, the proposed framework will allow one not only to capture the various mechanisms by which politicians choose to engage in corruption, but also to assess empirically their relative importance. This lies in contrast to the previous literature whose empirical evidence has been mostly based on simple correlations. Second, the framework will be used to evaluate the impact of anti-corruption policies that have been proposed by both academics and policymakers alike. A key feature of this research project is that it uses objective measures of corruption to estimate the model. Due to the inherent difficulties in measuring corruption, many studies have had to rely on subjective measures of corruption based on either perceptions or self-reported information. This project overcomes this limitation by exploiting measures of political corruption based on information from anti-corruption campaigns in two countries: Brazil and Puerto Rico. In both countries, the federal government audits local governments for their use of federal funds and provides detailed accounts of all the violations that were found to be corruption. These detailed audit reports combined with rich electoral and public finance data are used to estimate the model.
Broader Impacts: According to the World Bank Institute, countries that tackle corruption and improve their rule of law can increase their national incomes by as much as four times in the long term, and reduce child mortality by as much as 75 percent. The implementation of policies that optimally reduce corruption can therefore improve the welfare of millions of people. To design such policies, a sound understanding of the determinants of corruption is a prerequisite. This research project develops and plans to estimate a model that will help quantify the nature and extent of political corruption in local governments. The results of this research agenda will enable policy-makers to better construct appropriate policy as a redress for corruption.
