Presentation Title
Money Machine: Do the Poor Demand Clientelism?
University of Gothenburg
Start Date
Start Time
End Time
296 Gibson Hall (TBC)


The predominant view of clientelism is that the poor are particularly likely to exchange their votes for cash or material goods. The view is often based in a supply-side perspective, which argues that candidates are more likely to offer goods in return for votes to poor because their per-vote cost is lower, they are more likely to act reciprocally (thus overcoming the commitment problem of vote-buying), and less likely to see vote-buying as morally unacceptable. This view implicitly assumes that the poor prefer candidates who offer them one-time incentives for their votes. Yet, is this true? The clientelism literature suggests the poor are likely to prefer candidates who offer to buy their votes over those offering community goods, but research in social psychology finds that the poor may be particularly concerned community welfare and less likely to prefer such candidates. In this paper, we employ a rating-based, conjoint analysis in Malawi, to examine the extent to which voters, and particularly the poor, prefer candidates who promise selective incentives over those who promise goods for the community. We find significant evidence that community interests, and not selfish motivations, drive voters’ behavior. Even the very poor tend to prefer candidates who promise community goods (e.g., a school, clinic, roads), and they respond unfavorably to those who promise an immediate exchange of particularistic goods for votes (or “vote-buying”). This has important implications for the literature on clientelism, which often assumes that the poor prefer candidates who offer selective incentives. It also has important policy implications, as it suggests that offers of vote-buying to the poor may not be the most cost-effective strategy for candidates. Actors engaged in promoting clean elections may find it easier to counter vote-buying if they can provide credible evidence to candidates that vote-buying is an ineffective strategy. For many voters, the expectation that candidates will benefit the community, and not simply the individual, is paramount.


Ellen Lust is the Founding Director of the Programs on Governance and Local Development at Yale University (est. 2013) and at the University of Gothenburg (est. 2015), and Professor in the Department of Political Science at the University of Gothenburg. She received her M.A. in Modern Middle East and North African Studies (1993) and PhD in Political Science from the University of Michigan (1997). She held faculty positions at Rice University and Yale University, and was a visiting scholar at the Institute of Graduate Studies (Geneva, Switzerland) and the Straus Institute in the Law School at NYU. Ellen has conducted fieldwork and implemented surveys in Egypt, Jordan, Libya, Malawi, Morocco, Palestine, Syria, Tanzania and Tunisia. She has authored numerous books and articles, including most recently, Trust, Voice and Incentives: Learning from Local Successes in Service Delivery in the Middle East and North Africa (Washington, DC: World Bank, 2015) in collaboration with Hana Brixi and Michael Woolcock. Her current research is aimed at developing local governance indicators and examining the role of social institutions in governance. She was a co-founder of the Transitional Governance Project, a founding associate editor of Middle East Law and Governance, and has served as an advisor and consultant to such organizations as the Carter Center, Freedom House, NDI, UNDEF, UNDP, USAID, and the World Bank. Foundations such as the Moulay Hicham Foundation, National Science Foundation, Social Science Research Council, FORMAS and the Swedish Research Council have supported her work.