Vote Markets, Latent Opportunism, and the Secret Ballot

The secret ballot is a cornerstone of modern democracy because it protects voter autonomy and allows voters to express their political preferences freely without fear of repercussions. In theory, the secret ballot is supposed to prevent vote buying – the exchange of votes for money or material goods – from operating during elections. Yet, empirical evidence from surveys around the world suggests that vote buying is a common feature of elections in new democracies. Indeed, a fundamental puzzle concerns why political parties use vote buying to mobilize electoral support when the secret ballot allows voters to renege on their commitments and vote as they please. In this paper, we address this puzzle by arguing that voter perceptions of ballot secrecy affect their responses to vote buying offers. Theoretically, we develop a game theoretical model, where voter beliefs in the secret ballot guide their responses to vote buying offers. The model points to different scenarios under which vote buying is effective and where voters will be inclined to defect from or comply with vote buying offers. Empirically, we test the implications of the model using original survey data from a nationwide survey of 3210 respondents in South Africa conducted in the wake of the 2016 municipal elections. Using these data, we analyze how vote (and turnout) buying affects vote choice for the dominant party – ANC – in South Africa, and how this relationship is shaped by voter confidence in the secret ballot. Our results suggest that vote buying is effective mainly when voters doubt that they can cast their ballot in secret. In this way, our paper contributes to the existing literature by analyzing how and why parties operating in the shadow of the secret ballot use vote buying as an important part of their electoral strategies, and under what circumstances such strategies are likely to work.