Where is vote buying effective? Evidence from a list experiment in Kenya

Abstract

Vote buying is common in democracies around the world. Yet relatively little is known about the conditions in which vote buying is an effective campaign strategy, in part because vote buying is challenging to measure. This paper examines the local economic and social conditions in which vote buying influences the behavior of voters in Kenya. I combine data from a nationally representative list experiment conducted after Kenya's 2007 elections with highly disaggegrated census data about local economic and social conditions. While 7 percent report that vote buying influenced their vote when asked directly, the list experiment finds that 23 percent were influenced. I find mixed evidence and statistically weak evidence that vote buying is more effective in the local areas where parties should be best able to monitor voters. Vote buying is, however, most effective where voters lack access to information about politics. I discuss the implications of the results for literature on vote buying, clientelism, and electoral accountability.