Does raising voter expectations improve accountability? A field experiment in Mali (IGC Working Paper)

Abstract

I argue that if citizens systematically underestimate what their government can and should do for them, then they will hold politicians to a lower standard and sanction poor performers less often. A large-scale experiment across 95 localities in Mali in which some voters received information about potential government performance identifies effects of raising voter expectations. Survey experiments on the intent to vote (N=5,560) suggest that people in treated villages are indeed more likely to sanction poor performers and vote based on performance more often. There is also support for the idea that voting is a strategic calculation in which an individual’s actions are contingent on beliefs about others: treatment improved voter coordination and worked better when provided to a majority of villages. A behavioral outcome – the likelihood that villagers challenge local leaders at a town hall meeting – adds external validity to survey findings. Contrary to expectations, increasing voter information appears to decrease politician transparency, at least in the short-run