Get access

What Drives the Swing Voter in Africa?


  • The data used for the analysis and a replication file can be downloaded from the AJPS Data Archive on Dataverse: The authors wish to acknowledge insightful comments on earlier versions from four distinctively constructive and helpful reviewers and from the editors, as well as from Leonardo Arriola, Michael Bernhard, Younhok Choe, Gary Cox, Philip Keefer, Ellen Lust, Kristin Michelitch, Susan Stokes, and participants in both the Quality of Government Institute's research group at University of Gothenburg and the Comparative Politics Colloquium at the Department of Political Science, University of Florida. A special thanks is extended to Won-Ho Park and Dominic Lisanti for initial help with the count-models. The data collection was sponsored by a grant from the Africa Power and Politics Programme funded by the UK Department for International Development (DFID). The survey was carried out by Staffan I. Lindberg in collaboration with research officers at Center for Democratic Development-Ghana, and we wish to recognize the excellent work by the 49 fieldwork assistants. Staffan I. Lindberg's time spent on the project has also been partially funded by a grant from Riksbankens Jubileumsfond (co-PIs Bo Rothstein and Sören Holmberg). As always, the content, errors, omissions, and flaws of the text are the responsibility of the authors.


What makes African voters “up for grabs”? Existing approaches to the swing voter have several liabilities. This article introduces a new measure enabling a more comprehensive assessment of swing voting, including the differentiation between clientelistic and collective goods motivations. The issue of swing voting is then brought to an environment where voters are rarely considered persuadable: Africa. Using a count-model estimation technique and original survey data from Ghana's critical 2008 elections, the analysis challenges the near consensus in African politics on clientelism as the only electoral strategy. When voters perceive politicians as providing collective, developmental goods, the efficacy of clientelism as a tool to win over voters is reduced. Many persuadable voters can also be won over by both clientelistic and collective goods, thus contradicting the literature presenting these as mutually exclusive. Finally, the analysis shows that incumbents do better when they provide collective goods even in highly clientelistic environments.