Ballot structure, list flexibility and policy representation

ABSTRACT

There is a growing body of research on the impact of the electoral system ‘ballot structure’ on the behaviour of politicians. We offer a clear, ordinal and rules-based three-way coding (closed, flexible, open) of the electoral systems used in European Parliament elections, taking into account both the ballot type and the intra-party seat-allocation rules. For the notoriously difficult group of flexible list-systems, we show how these operated in the 2004, 2009 and 2014 elections, and introduce an additional behavioural distinction between ‘weakly flexible’ and ‘strongly flexible’ subtypes at the party-list-level. We then illustrate how the type of ballot used in an election can influence individual policy representation by looking at the vote-splits between Members of the European Parliament (MEPs) in the European People’s Party in a vote on tackling homophobia.

KEYWORDS:

• Ballot structure
• electoral systems
• European Parliament election
• parliamentary voting
• personal vote
• representation

Acknowledgements

We would like to thank Stefanie Bailer, Ken Benoit and Dominik Hangartner for comments, Phillip Ayoub for sharing data, as well as Paul Meiners and Ertan Bat for research assistance. Last but not least, we are grateful to many people who answered our questions about electoral rules or helped with obtaining electoral results data. These include Kostas Gemenis, Raimondas Ibenskas, Tom Louwerse, Carolina Plescia, Georgina Sideri-Papagianni, Eftichia Teperoglou, Yannis Theocharis and Galina Zapryanova.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Thomas Däubler is a post-doctoral researcher at the Mannheim Centre for European Social Research, Germany.

Simon Hix is Harold Laski Professor of Political Science, London School of Economics and Political Science, UK.

Notes

1 Appendix A1 provides a comparison of our coding and that by Farrell and Scully (2007).

2 ‘Party’ in this context could refer to the national party or the European party group or both.

3 http://www.europarl.europa.eu/sides/getDoc.do?type=MOTION&reference=P7-RC-2012-0234&language=EN.

4 Some of the absences could be due to MEPs being cross-pressured, which would bias our results downwards. Appendix A6 presents a robustness check based on the sample with ‘did not vote’ coded as abstentions. The key results do not change.

5 This is also clear from comparing the results to a regression model that uses the rules-based three-way classification (Appendix A7). Findings for the category comprising all flexible-list systems are inconclusive in this model, and the error variance at the party-level is considerably higher than in the model with the four-way measure.

6 We alter the public opinion figures, their interactions where indicated by the observed values of the multipliers, and the case-specific electoral systems group variables, while leaving other variables as observed. The party-level intercept is assumed to be zero. Shown are means and 2.5/97.5 per cent quantiles across draws from the mean expected probability across MEPs in each draw.

Additional information

Funding

Thomas Däubler acknowledges funding by Deutsche Forschungsgemeinschaft [grant DA 1692/1-1], ‘The personal(ized) vote and parliamentary representation’.