Essays on Electoral and Intra-Elite Accountability in sub-Saharan Africa

Abstract

Existing macro-level research on electoral accountability in sub-Saharan Africa has found broadly beneficial effects to political competition. At the same time, micro-level work in the clientelism literature has explained the prevalence of vote buying on the sub-continent with a variety of positive-response theories. These essays engage both literatures by looking at both vertical and horizontal accountability. The first paper is motivated by an enduring puzzle in the clientelism literature: why do we observe vote buying under the secret ballot? I argue that we observe widespread vote buying in low-credibility environments because voters attempt to maximize the value of current and future elections by demanding offers as the price of their turnout. Empirical results suggest that elections in low-credibility environments may cause candidates to sacrifice long-term reputation for short-term electoral gain. The second paper examines the effects of promise-making in low credibility environments. Using original observational data, I show that promise-making produces anchoring effects as voters price the expectation of future provision into their evaluations of candidates. I argue that the prevalence of over-promising is due to a time-inconsistency problem facing candidates during the campaign period. Using a conjoint experiment, I demonstrate that politicians are unlikely to suffer electoral costs for over-promising, and in some cases can actually benefit from doing so. The third paper examines horizontal accountability between elites of different ethnic groups during the formation of governments. Existing work on coalition formation in sub-Saharan Africa has focused either on coalition size or on the size of individual coalition members. Neither, however, deal systematically with other coalition-level features such as diversity and intra-coalition income inequality. We attempt to remedy this gap by modelling coalition formation as a two-stage bargaining process. We then test the model with real-world data via structural equation estimation. Results suggest that smaller coalitions are more desirable to potential members, that more income equality makes coalitions more attractive, and that transaction costs, proxied by the number of groups, make coalitions less desirable.