We know women are more likely to be elected under proportional representation, but it hasn’t been clear which aspects of PR benefit them. Sona N. Golder, Laura B. Stephenson, Karine van der Straeten, André Blais, Damien Bol, Philipp Harfst and Jean-François Laslier designed an experiment in which people could cast (fake) votes for (real) European Parliament candidates in three different ways. They found that having more women on the ballot does not – as some fear – put off voters.
The vast majority of us can agree that we need more women sitting in parliaments. They are under-represented in legislatures all over the world. But what is the best way to achieve proportional gender representation?
When countries like Canada consider altering their electoral systems, better representation of women in the legislature is a common aim. Unfortunately, the research available to guide reformers as to which electoral system would best accomplish this is mixed. One strategy might be gender quotas, although another potential solution is related to the electoral formula itself.
It is a well-established finding that proportional representation (PR) is associated with greater female legislative representation than single member majoritarian systems, but that finding typically relies on a comparison of the gender composition across different national legislatures – the specific aspects of PR rules that affect voting for female candidates are not fully understood. The existing literature is hampered by a reliance on cross-national data in which individual vote preferences and electoral system features affect one another. In a recently published study, conducted as part of the Making Electoral Democracy Work project, we report the results of a unique experiment that overcomes these usual data limitations and allows us to identify the effects of specific aspects of PR electoral rules on votes for women.
Four questions motivated our research.
First, does having more women on a party list pose a disadvantage to a party’s electoral performance?
Second, does the ability to cast a preferential vote for a candidate, rather than only being able to vote for a party, affect the propensity to vote for women?
Third, is there evidence of same-gender affinity voting (men voting for men, women voting for women)?
Finally, are conservative voters less likely to vote for female candidates?
With these questions in mind, we designed an experiment to enable us to isolate the effects of different PR electoral systems, conducted during the 2014 European Parliament (EP) elections. The study was open to everyone and available in 14 different European languages. Participants in the experiment were given the opportunity to vote for real EP candidates in hypothetical but realistic EU-wide districts (such a system has been proposed). The key element of the study was that we asked each participant to vote three different ways: under closed list PR electoral rules (in this system, used in France, voters vote for a party), open list PR (in this system, used in Latvia, voters can indicate their preferences over one party’s candidates), and open list PR with panachage and cumulation (in this system, used in Luxembourg, voters cast up to two votes for multiple candidates, possibly across party lines). Voters were therefore given the opportunity to support a party, support a party by indicating a preference for a specific candidate, or customise their ballot to support one or more parties and one or more candidates.
We used the existing seven political groups in the EP as parties and randomly chose candidates on the party lists from the actual members sitting in the EP. This led to candidate lists that varied widely in the number of female candidates – from none to 90% of the list (the average was 32%). Although each participant faced a different (randomly generated) ballot, it remained the same under each of the three voting rules. This is an essential feature of the study: because voter preferences were held constant across the three different votes (the same person cast each one), we are able to evaluate the extent to which female candidates were more or less advantaged by the change in electoral system rules.
Our findings are encouraging on two fronts. First, we do not observe any systematic bias against female candidates. In all cases, the proportion of women on the ballot has a neutral or positive effect on party support. This shows that parties are not hurt when they have more women on their lists, whatever the electoral system. Second, we find that the propensity to vote for women increases when voters are given more freedom to express their preferences about the candidates. This effect holds for both male and female voters, although it is particularly strong for female voters. While we anticipated finding some evidence of same-gender voting, we do not observe that at all under closed list PR rules, and not for men under both types of open list PR. Even more encouragingly, the advantage given to female candidates as electoral systems become more open holds true independent of ideology.
Overall, we find that voters, regardless of their gender, support female candidates, and that this support is stronger under open electoral rules. Concerns about voters being put off because of the presence of women on the ballot appear to be unfounded.
This post represents the views of the authors and not those of Democratic Audit.
Laura B. Stephenson is an Associate Professor in the Department of Political Science at The University of Western Ontario, Canada.
Karine van der Straeten is a Senior Researcher at CNRS, and a member of the Toulouse School of Economics and the Institute for Advanced Study in Toulouse.
André Blais is a Full Professor and Research Chair in Electoral Studies at the Université de Montreal.