Polls, Turnout and Partisanship in South Africa's First Competitive Election 

• In Event: Experiments on Voter Participation and Partisanship in Southern and East Africa

Abstract

Theory and lab experiments predict turnout is higher in competitive elections where votes are more pivotal, but are ambiguous on the effects of contestation on partisanship: people may swing behind a party which is behind (underdog effects) or swing behind a winning party (bandwagon effects). Lab studies largely accord with theory, but there is limited well-identified evidence on the effect of greater contestation in real elections. I study South Africa’s 2016 municipal elections, where the dominant party, the ANC, suffered a surprise upset, losing three major cities by small margins, including Johannesburg. I survey 2,023 participants from around Soweto, the largest black township in Johannesburg. Pre-election, few believe the ANC will lose or the election will be close. I provide two treatment groups with face-to-face information that the election will be extremely close from a (poorly publicised) IPSOS-Mori poll and examine the effects of knowing an election is more competitive on beliefs, turnout, partisanship and political engagement. One group was told the challenger party was ahead and one that the incumbent party was ahead, so I can examine how the effects of information on competitiveness differ depending on whether the incumbent or challenger is ahead, exploring when and for whom bandwagon or underdog effects dominate. I can also examine the extent and quality of citizen engagement in the election to test the hypothesis that competition improves the quality, as well as extent, of citizen participation.

Author

• Kate Orkin, University of Oxford