Published online by Cambridge University Press: 15 July 2016
In this analysis of how electoral rules and outcomes shape the internal organization of political parties, we make an analogy to primary elections to argue that parties use preference-vote tallies to identify popular politicians and promote them to positions of power. We document this behavior among parties in Sweden's semi-open-list system and in Brazil's open-list system. To identify a causal impact of preference votes, we exploit a regression discontinuity design around the threshold of winning the most preference votes on a party list. In our main case, Sweden, these narrow “primary winners” are at least 50% more likely to become local party leaders than their runners-up. Across individual politicians, the primary effect is present only for politicians who hold the first few positions on the list and when the preference-vote winner and runner-up have similar competence levels. Across party groups, the primary effect is the strongest in unthreatened governing parties.
We thank Thomas Fujiwara for kindly sharing data on Brazilian municipal elections. For their helpful comments, we thank Daron Acemoglu, Ernesto Calvo, Jens Olav Dahlgaard, Torun Dewan, Mikael Gilljan, Miriam Golden, Shigeo Hirano, Simon Hix, Ulrik Kjaer, Eric Magar, Jenny Madestam, Monica Martinez-Bravo, Jim Snyder, Pär Nyman, Henric Oscarsson, Daniel Markham Smith, Pablo Querubin, Ken Shotts, and David Strömberg; participants in the following workshops: Empirical Political Economics and Political Science Workshop (WATT), Midwest Political Science Association 2013, European Political Science Association 2014, and the NYU-LSE Political Economy Conference 2015; and seminar participants at: Lund University, Uppsala University, Columbia University, Cal Tech, University of California at Berkeley, Gothenburg University, Universidad Pompeu Fabra, Canadian Institute for Advanced Research, London School of Economics, Institute for International Economic Studies, Linnaeus University, and the Swedish Parliament. We are also grateful to the three anonymous reviewers and the APSR editorial team for their thorough and constructive comments on the manuscript. Financial support from the Swedish Research Council and the Torsten and Ragnar Söderberg Foundations is gratefully acknowledged.
Comments
No Comments have been published for this article.