Buying Local Votes: The Effect of Individual Campaign Spending Under a Semi-Open PR System in the Belgian Local Elections

Abstract

This article analyses the effect of campaign spending on the individual result of party candidates in the Belgian local elections. An analysis of data concerning the 2012 local elections in the Flemish region shows that candidates who spend more in absolute terms or outspend their rivals (at the list and the municipality level) obtain a better result, even though the size of the effect is small. Contrary to what was found for national elections, there are indications that spending affects the odds of obtaining a seat in the local council. The spending effect tends to be weaker for candidates holding an executive office in the municipality.

Key Words:

• Campaign spending
• local elections
• semi-open list PR system
• preferential votes
• Belgium
• media coverage
• office effects

Acknowledgements

The authors thank the anonymous reviewers of Local Government Studies for their suggestions. Gert-Jan Put also likes to thank the Research Foundation Flanders for their generous support.

Notes

1. The organisation of the local elections is a competence of the regions. Only the legislation for the Flemish region (Lokaal en Provinciaal Kiesdecreet van 8 juli 2011) is discussed here.

2. This is apparent from various details: the reports are normally filed in one folder by the local party with a single cover letter, and the handwriting on the various forms (or the font type) is often the same.

3. The average number of inhabitants per municipality in our sample (16,355) is comparable to the average for all Flemish communities (without Antwerp and Ghent), that is, 18,300.

4. As these types of political office are not all mutually exclusive, some candidates cumulate a local and higher office. This implies that we distinguish two different office variables in the analysis: holding one of the various local offices (with holding no local office as reference category) and holding an office at a higher (provincial, regional, national and European) level (with holding no office at such a higher level as reference category).

5. In addition, in each analysis we also checked for outliers on the basis of the externally studentised residuals.

6. Reactive spending implies that both the challenger and the incumbent increase their spending in response to opinion polls showing that the challenger stands a real chance of dethroning the incumbent (Green and Krasno 1988, Gerber 1998).

7. This negative effect is not due to the larger number of preferential votes cast in smaller municipalities as this variable is already included in the model. (percentage of voters casting a preferential vote, and average number of preferential vote per ballot form).

Additional information

Notes on contributors

Gert-Jan Put

Gert-Jan Put is a researcher at the Research Foundation Flanders (FWO) affiliated to the Public Governance Institute of the University of Leuven, Belgium. His research interests include political party organisation, candidate selection, geographic representation, legislative turnover and campaign spending.

Bart Maddens

Bart Maddens is professor of political science at the Public Governance Institute of the University of Leuven, Belgium. His research focuses on political party finance and elections.

Jef Smulders

Jef Smulders is a research assistant at the Public Governance Institute at the University of Leuven, Belgium. His research focuses on political party finance, more specifically annual party accounts, as well as on party organisation and party types.