New evidence from East Africa suggests generosity declines as balloting approaches.
Elections are moments of intense competition for control of the government and its resources. As last year’s U.S. elections made all too clear, they often polarize electorates and deepen social divisions.
They tend to direct the debate to material benefits, such as government handouts for specific groups, targeted tax-cuts, or local infrastructure investment, and thereby to voters’ self-interest.
For the first time, researchers have demonstrated a link between elections and selfishness. The study draws on extensive data from Kenya and Tanzania.
The researchers gave a small sum of money (approximately equal to an hour’s wage) to more than a thousand individuals in three separate lab rounds. They then asked them to decide how much of this to give away to another randomly selected participant about whom they were given no information.
Participants were recruited from low-income neighborhoods in the capital cities Nairobi and Dar es Salaam to achieve broadly representative samples. Researchers then tested the effects of elections on selfishness in two ways.
Testing the effect of elections on selfishness
Firstly, participants were randomly divided into two groups. One was subtly reminded of upcoming elections before deciding how much money to give away, while the other was not. Members of the election group gave away around 7 percent less than members of the non-election group did.
Secondly, the researchers carried out two lab rounds in Kenya. One around eight months before upcoming elections, and one around 1 month before. In the lab round right before the elections, participants on average gave away 15 percent less than participants in the earlier round.
This decline in generosity was present for each subgroup in the sample, namely across different genders, education levels, ages, intelligence levels and family sizes. Moreover, a large part of the rise in selfishness was due to more than a doubling in the share of participants who kept the entire sum for themselves, from 8 percent to 19 percent.
“The fact that we found such a consistent effect suggests elections may in fact increase selfishness. It is important to note though that these findings are specific to East Africa, and additional research would be needed to investigate whether similar connections exist in other parts of the world,” says Assistant Professor Simon Galle at BI Norwegian Business School, co-author of the paper.
Politics in Kenya and Tanzania are strongly associated with clientelism, rent extraction and corruption. The independent polling and research organization Afrobarometer reports that an overwhelming majority of respondents in both countries believe at least some government officials are corrupt.
“If elections take place in an atmosphere focused on narrow self-interest, clientelistic or corrupt politicians may hold a strong and persistent advantage over candidates who aim to advance the common good,” says Galle.
“From this perspective, our research may help explain why rent extraction remains rampant in young democracies and why democratization does not necessarily help in battling entrenched corruption.”
Source: Kjetil Bjorvatn, Simon Galle, Lars Ivar Oppedal Berge, Edward Miguel, Daniel N. Posner, Bertil Tungodden, Kelly Zhang, Elections and selfishness, Electoral Studies, Volume 69, 2021, 102267, ISSN 0261-3794, https://doi.org/10.1016/j.electstud.2020.102267.
Sign up for our newsletter to get the latest news from BI Business Review.sign up