Essays on violence, money in politics, and electoral system in Colombia

This thesis focuses on the political economy of development, in particular on the causes and consequences of political selection of leaders. It analyses how different electoral systems can determine the type of politician that runs in elections, and the effects of electing different politician types. The first paper studies the consequences of electoral victory of newcomer political parties in a context of newly-introduced local elections and weak institutions. Using a regression discontinuity approach, we find that narrow electoral victories of previously excluded left-wing parties to local executive office in Colombia result in an almost one-standard-deviation increase in violent attacks by right-wing paramilitaries against municipalities that elected-left wing parties. Violence can surge as a de facto reaction of traditional political and economic elites when there is a victory of radically different groups and in the absence of monopoly on violence. This paper shows that the introduction of elections does not necessarily lead to less violence, but can lead to more violence when the electoral winner are newcomers with radically different ideas. The second paper focuses on the consequences of electing a politician funded by donors, and how such politicians may benefit his/her donors disproportionately. Using a novel dataset that uniquely identifies campaign donors and recipients of public contracts during a mayor’s incumbency period in Colombia, I find that barely electing a politician who received donations more than doubles the probability of donors receiving contracts from a mean of 5.9% to 15.5%. Moreover, electing a donor funded politician does not lead to lower local government deficit, nor increasing investment, but it does increase the probability of having a disciplinary sanctions against them, and investigations related to public procurement. I then study the role that campaign contribution limits can play in undermining the role of money in politics. Results indicate that campaign limits lead to less participation of donor funding in campaigns, and as a result reduces the number of contracts that the mayor will reward to donors. This paper shows evidence of the consequences of electing a donor funded politician, and how campaign limits can reduce the influence of campaign contributions. The third paper studies how open or closed lists in proportional representation systems can affect electoral performance and political selection. In open lists, a candidate’s internal party ranking is determined by voter preferences, in contrast to closed lists, in which ranking is predetermined by party preferences. Colombia provides a unique electoral system where parties can field open or closed lists in municipal elections, producing a mixed-list type of electoral system. Using qualitative interviews and constructing a new dataset on politician characteristics and campaign investment, we find that open lists (vs closed lists) produce better electoral returns to the party, induce higher campaign efforts by candidates, and select higher-quality candidates–that are less likely to have registered illegally to vote, and have more political experience. This paper shows that voter discretion – trough open lists – can have a dramatic effect on party performance and on the type of politicians ranked in elections