Treacherous Coattails: Gubernatorial Endorsements and the Presidential Race in Kenya's 2017 Election

This is the author accepted manuscript. The final version is available from Taylor & Francis (Routledge) via the DOI in this record.Could there be coattail effects in the absence of strong parties? How would these effects manifest in countries with ethnic and personality-based politics? Kenya’s 2017 election presents an opportunity for a theoretical and empirical contribution to the study of coattail effects in such settings. With the newly-created and highly attractive positions of county governors, down-ticket races became a lot more competitive, forcing parties to make difficult choices in terms of campaign focus, the apportioning of resources across the ballots, and how to maintain or forge alliances with local leaders whose networks were key to success in the battlegrounds. Presidential candidates found themselves in a precarious position: endorsing governor aspirants in competitive races could lead to a backlash and cost them votes, failure to endorse could signal lack of confidence in key figures and thus potentially jeopardize all six positions on the ballot. This paper draws on theories of coattail effects in democracies and adapts them to better understand the relationship between the races for governor and president in Kenya’s 2017 election. I argue that coattail effects are conditional on governors receiving clear and public endorsements by the presidential candidates and that effects flow from presidential candidates to governor aspirants in parties’ strongholds, and vice-versa in battleground counties. The incumbent Jubilee party was better able to harness gubernatorial coattail effects because of its ability to field single candidates and entice popular local leaders to either join the ticket or stand down in favour of ticket holders. The findings have broader implications for theories of coattail effects, campaign strategy, legislative fragmentation, and citizen-politician linkages in settings with personality-based politics and weakly-institutionalised parties.The fieldwork for this research was supported by the Max Planck Institute for the Study of Religious and Ethnic Diversity, Goettingen, German