If someone offered you money – money you may desperately need – to vote for a particular candidate, would you do it?

Vote-buying is endemic in many countries around the world – even in countries where politicians regularly vie for public office through competitive elections. Take, for example, Uganda. In 2016, Yoweri Museveni, the incumbent leader of the National Resistance Movement party in Uganda, was elected president for a seventh term. He first assumed leadership in 1986 and has not lost an election since, though election observers condemned several of these elections as unfair. At the local level, Ugandan elections tend to be fairly competitive. 

Vote buying, a reciprocity system in which political candidates secure votes in exchange for gifts or money, has long been a part of politics in Uganda, a low-income, increasingly autocratic East African country of about 40 million people. Well-funded incumbents often benefit from this practice in national and local elections. In the leadup to the 2016 election, the Alliance for Election Campaign Finance Monitoring (ACFIM), an electoral civil society organization in Uganda, launched a campaign to reduce vote buying. 

Harris Public Policy Professor Christopher Blattman
Harris Public Policy Professor Christopher Blattman

The efficacy of that campaign is the topic of a recent paper – entitled “Eat Widely, Vote Wisely? Lessons from a Campaign Against Vote Buying in Uganda” – from Christopher Blattman, the Ramalee E. Pearson Professor of Global Conflict Studies at the University of Chicago Harris School of Public Policy and The Pearson Institute for the Study and Resolution of Global Conflicts, alongside coauthors Horacio Larreguy, Benjamin Marx, and Otis R. Reid of Harvard University, Sciences Po, and the Massachussets Institute of Technology, respectively. 

What Blattman and his coauthors found was that while the ACFIM campaign did not noticeably slow the practice of vote buying, the longstanding practice of voter reciprocity eroded. Put more simply, many people took the money, but voted as they wished once they were in the proverbial voting booth. 

The ACFIM campaign’s main goal was to introduce a “refusal norm,” or to convince voters that turning down a vote buying offer was the right thing to do. Instead, villagers exposed to the campaign took away an equally disruptive, related idea: take the money and vote for whom they please. That is, they set out to change the reciprocity norm that influences so much vote buying. The evidence suggests that the anti-reciprocity norm shifted more than the refusal norm. 

The ACFIM anti-vote-buying campaign eroded the practice of voter reciprocity.

ACFIM’s efforts to change the norms around vote-buying were multi-pronged. ACFIM activists distributed leaflets describing individual and community moral choices and held informational meetings for villages to collectively renounce the practice of vote-buying. The organization also hosted community discussions allowing for public commentary from constituents in the villages selected for this social experiment, made robocalls, and displayed posters. 

In addition to dissuading voters from accepting gifts, the campaign was designed to serve as a warning to politicians that “village norms had changed, and that their vote-buying transactions might not be welcomed or honored in that village,” according to the paper.

ACFIM began their efforts one month ahead of the February election. They based their operation in 2,796 villages across 918 parishes, defined in the paper as “a collection of 3 to 10 villages...the typical operating unit for the rural vote brokers who work for politicians.” All in all, the report’s authors sampled 1,427 villages comprised of half a million voters. 

“This allows us to estimate average treatment and spillover effects of the ACFIM campaign, and how these effects vary with treatment saturation at the parish level,” they wrote about their methods. “This approach builds on a new strand of empirical work designed to uncover spillover and general equilibrium effects in experimental settings.”

The paper surveyed 28,454 villagers in all 918 eligible parishes of Uganda.

Following the 2016 election, in which Museveni captured 61% of the vote, Blattman and his colleagues surveyed 28,454 villagers in all 918 eligible parishes to assemble post-election analysis on voters’ experiences with vote-buying. At this time, they also expanded their research to include 1,399 nearby villages who had not been part of the ACFIM’s efforts. 

Using a combination of administrative data and qualitative accounts from voters and brokers affected by the ACFIM’s intervention efforts, the researchers sought to measure the impact of the campaign.

“Our key finding is that the campaign had substantial effects on electoral outcomes, even though it did not reduce the extent of vote buying,” they wrote in their report. “In areas exposed to the campaign, vote shares decreased for incumbents and correspondingly rose for challenger (non-incumbent) candidates running in the presidential and parliamentary races.”

They measured a standard deviation of 0.06 in both treatment and spillover villages, which equated to an approximately 10 percent voter share increase for Besigye — Museveni’s main contender — from the 2011 election to the 2016 election. 

ACFIM activists distributed leaflets describing individual and community moral choices and held informational meetings for villages to collectively renounce the practice of vote-buying. The organization also hosted community discussions allowing for public commentary from constituents in the villages selected for this social experiment, made robocalls, and displayed posters. 

This paper documents one of the largest anti-vote-buying campaigns ever evaluated, and while the campaign may not have been effective in terms of diminishing the practice of vote buying, it did weaken the reciprocity norm. Whereas previous campaigns indicate a negative impact on the number of votes sold, ACFIM’s campaign demonstrated a weakening relationship between vote buying and voter behavior. In the long term, the erosion of voter willingness to honor the vote-buying “contract” has the potential to increase the competitiveness of local elections.