How legislative gender quotas affect the gender gap in campaign spending: an analysis of the federal and regional elections in Belgium

ABSTRACT

Research on the link between gender and campaign finance in proportional electoral systems suggests that the campaign expenses of female parliamentary candidates are significantly lower than those of male candidates. On the basis of data on 10,436 candidates for nine consecutive elections in Belgium (1991–2014), this article examines whether there is indeed a gender gap in campaign expenses, and in particular whether this coincides with the introduction of legislative quota laws in the Belgian flexible-list system. We distinguish between realistic candidates that run for election from winnable list positions and unrealistic candidates running from lower ranked positions. The results show that, among unrealistic candidates, the gender gap in campaign spending arose again after the introduction of more severe gender quotas. With regard to realistic list positions, however, the significant difference between male and female candidates in the most strict quota phase disappeared, indicating that female realistic candidates were able to catch up financially with their male counterparts. The Belgian experience could provide useful insights for other countries with flexible-list systems regarding the implementation of legislative gender quotas.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Jef Smulders is a PhD fellow of the Research Foundation - Flanders (FWO), affiliated to the KU Leuven, Public Governance Institute. His research interests include political finance, party organization and campaign spending. His work has been published in journals such as International Political Science Review, Government and Opposition and Local Government Studies.

Gert-Jan Put is a senior researcher at the KU Leuven, Research Center for Regional Economics. His research interests include candidate selection, intra-party competition and campaign spending. His work has been published in journals such as Political Behavior, Politics and Electoral Studies.

Bart Maddens is a professor of political science at the KU Leuven, Public Governance Institute. His research interests include political finance, campaign spending, elections and multi-level systems. His work has been published in journals such as West European Politics, Electoral Studies and Acta Politica.

ORCID

Jef Smulders http://orcid.org/0000-0001-6014-2885

Notes

1 See the website of the Inter-Parliamentary Union, with comparative data by country and world and regional averages: http://www.ipu.org/iss-e/women.htm.

2 The only exception being the 1991 federal elections, where voters could endorse the party list as a whole or cast only one preferential vote for a specific candidate.

3 Effective candidates are those who can be directly elected to parliament, on the basis of their individual preferential votes and/or their place on the list. Successor candidates are assigned to a separate and additional list. When an elected MP does not take the seat, becomes a minister, resigns or leaves the seat vacant due to illness or death, a successor candidate will take this seat, based on preferential votes and/or place on the list.

4 The extent to which candidates actually spend these amounts strongly varies between political parties. For instance, while candidates from smaller Belgian parties such as Agalev/Groen barely spend any campaign money at all, their counterparts from CVP/CD&V often spend up to 90% of what is legally allowed (Maddens et al. 2017).

5 These parties are: CVP/CD&V, PVV/VLD/OpenVld, SP/sp.a, Agalev/Groen, Vlaams Blok/Vlaams Belang and VU/N-VA.

6 In order to make expenses comparable across time, all amounts are recalculated in prices of December 2016, on the basis of the Belgian consumer price index.

7 Of all 10,436 observations in our dataset, 17.1% did not spend anything. These observations are mainly candidates of Vlaams Blok/Vlaams Belang and Agalev/Groen, which traditionally spend significantly less than those of other parties.

8 For multiple reasons, we do not expect to see significant differences in terms of gender inequality between federal and regional elections. First, parties competing with each other on both levels are exactly the same. Second, the fairly high degree of level-hopping in the Belgian political arena indicates that politicians themselves rate the federal and regional level as relatively equal (Vanlangenakker, Maddens, and Put 2013).

9 The reported gender differences in campaign spending cannot be reduced to unequal financial support for candidates by political parties. On the contrary, the share of party financial support in the campaigns of female candidates is on average 5% higher, which confirms that reported differences reflect women being less inclined to invest personal resources into their campaigns than their male counterparts (Maddens et al. 2017).

10 We considered presenting a pooled model with a dummy for each gender quota phase in Belgium. Testing the hypotheses implied entering three-way interaction terms between sex, realistic candidate and gender quota phase. As this would substantially complicate interpretation of our findings, we present separate models for each quota phase. In the pooled model, however, we also find that the main term of sex is statistically significant and that the interaction term of sex and realistic candidate does not significantly affect spending levels.

11 For an overview, see: https://www.idea.int/data-tools/data/gender-quotas.

Additional information

Funding

This work was supported by Research Foundation Flanders.