![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
This article analyses the drivers of individual dissent in floor voting in parliamentary regimes. It focuses on the effect of ideological heterogeneity in legislative parties on individual MPs’ voting behaviour, as well as the different incentives caused by the differing consequences of defection and abstention. Combining individual-level survey and voting data from the Swedish Riksdag, neither of which is subject to selection bias, the study overcomes several limitations of previous research. It shows that MPs’ decisions to dissent are partly driven by ideological differences with their party, but also by the imperatives of maintaining a government majority in a parliamentary regime, along with the level of influence MPs exert on legislation. It also highlights the importance of distinguishing between abstaining from voting and defecting. Merely pooling the two oversimplifies the behaviour of MPs.
Acknowledgements
We would like to thank Elin Naurin and Tamaki Ohmura for providing feedback and suggestions on the manuscript throughout the writing process. Lars Brink was extremely helpful with data collection, for which we are very grateful. Previous versions of this work were presented at the 2013 EPSA Conference in Barcelona and the 2013 ECPR General Conference in Bordeaux, and we thank the discussants and participants for their helpful comments. We would also like to thank the editors and anonymous reviewers at West European Politics, whose comments greatly helped us to improve the article.
Notes
1. Kam (Citation2001) combined recorded votes with roll call data in the British and Canadian Houses of Commons with survey response rates of 53.1% (Canada) and 69% (UK). However, in both these institutions, the ‘standard operating procedure’ (Hug et al. Citation2015) for voting is a voice vote, meaning only the outcome of the vote, rather than the individual MP’s vote, is recorded; as such, selection bias worries in terms of recorded votes are significant in these cases.
2. That is, the individuals and groups in a party that control the nomination process (Hazan and Rahat Citation2006; Pennings and Hazan Citation2001; Rahat et al. Citation2008).
3. As the surveys were clearly stated to be for academic research only, the effect of desirability bias in terms of answers will have been minimised, if not eliminated.
4. Both Ansolabehere et al. (Citation2001) and Coleman Battista and Richman (Citation2011) use data from Project Vote Smart to measure legislators’ policy preferences; however, as the authors acknowledge, these are not ‘true’ measures of policy preferences, but rather electorally induced ones ‒ that is, these surveys measure what legislators want their votes to think they believe, not what they actually believe.
5. Decisions in the Riksdag can be made either by acclamation, electronic voting, the calling of the roll, or by standing. As acclamation requires unanimity, it is used only for non-controversial issues. Both electronic voting and the calling of the roll are automatically recorded, and all MPs have the right to request a recorded vote when voting by standing (Riksdagsordning [Riksdag Rules of Procedure] Citation2014: chapter 11, paras. 8, 9, 10). Interviews with Riksdag civil servants confirm that voting by standing never takes place in practice.
6. Further, in time for the 1998 election, the electoral system was changed to allow personal voting that could alter the order of party lists. Future work will extend the dataset backwards in time, beyond this time.
7. This distinction of course also hinges on whether decisions are made by simple or absolute majority rule. However, with the exception of some rare decisions related to constitutional change, all decisions in the Riksdag are by simple majority.
8. Sweden provides a good example of this. In the 1973 to 1976 term, the two blocks in the parliament split exactly 175:175, meaning that any abstention had the potential to change the outcome of a vote. As deadlock was widespread during this parliamentary term, the Riksdag reduced the number of members from 350 to the current 349 in time for the 1976 election, to prevent a repeat of this situation.
9. If the majority of a party abstained, both yes and no votes were coded as voting against the party line; this occurred in a total of 118 cases in the two terms.
10. In the period covered, two MPs left Vänsterpartiet, and one MP left Moderaterna.
11. As proposal powers in the Riksdag are concentrated in the parties, this measure captures whether a superior policy alternative (from the perspective of the dissenting MP) is available.
12. The official electoral returns of Statistics Sweden were used (http://www.scb.se/Pages/List____250611.aspx).
13. Given the levels of dissent and the fact that party positions will have been stated during debates on the legislation, it is reasonable to assume that MPs know whether a vote will be close or not.
14. As retiring MPs are thanked by the Speaker for their service immediately prior to an election, we could code this based on the Riksdag protocol.
15. It should be noted that free votes are very rare in the Riksdag (Sannerstedt Citation1992). A parliamentary civil servant stated that ‘It is rare I would say. I wonder if it even is once per mandate period’; as such, this should not influence our findings.
16. Voting on legislation in the Riksdag is done on full bills, meaning that each MP only votes on a piece of legislation once. As such, each vote in our dataset is independent of the other votes in it.
17. This was done using the ‘melogit’ command in Stata 14.
18. While Social Democratic governments in Sweden have all been minority governments since the mid-1960s, centre-right parties have formed both majority and minority governments.
19. The agreement was called the 121-point programme, and as the name implies it consisted of 121 policy agreements between the parties.
20. As a final robustness check, we re-ran all models with a term dummy for the 2002–2006 term, to capture any effects of a minority government. The results for all models are substantially identical to those presented in the paper.