Voter turnout and municipal amalgamations—evidence from Denmark

ABSTRACT

The study utilizes evidence from the Danish 2007 municipal reform to inquire into the relationship between municipal amalgamations and voter turnout, that is, the classical discussion on size and democracy. The Danish municipal reform is particularly suited for investigating the relationship as a large number of units were merged due to reasons thought to be unrelated with the democratic performance while others were left unchanged. This allows us to investigate the relationship in a quasi-experimental setup. The study finds some evidence of a short-term positive effect on turnout and only limited evidence of a medium- to long-term negative effect when comparing amalgamated and non-amalgamated municipalities. However, stronger indications of negative effects on turnout are found when considering the intensity of the amalgamations.

KEYWORDS:

• Voter turnout
• amalgamation
• participation
• quasi-experiment
• jurisdiction size

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. In the analyses, we operate with 67 merged municipalities and 31 non-merged as we include the municipality of Bornholm as a merged one. This municipality was merged from five to one in 2003, and as we will return to, we utilize 1997 as the base year in the analyses.

2. A few old municipalities were divided into several new municipalities in the reform. In the analyses, these are counted as part of the new municipality in which the majority of their constituent became a part. We have tested an alternative operationalization in which we use a proportional weight based on the share of voters who were transferred to the different municipalities (based on the number of eligible voters in the 2001 elections) which yields almost identical results as the ones in Table 1.

3. Ln(20000)-ln(10000) = 0.69 meaning that a coefficient of −3 corresponds to a decrease of about 2 percentage points. p < 0.05.

4. Percent young and older is operationalized as the percent of age 18–29 and 80+ among the population of 18+ year-olds in the municipality. Percent school education is the percentage of 15–69-year-olds with a school education (less than high school). Immigrants percent is the percent of 18+ year-olds in the municipality with a non-Danish background. Small-island municipality is the municipalities Langeland, Ærø, Fanø, Samsø, and Læsø. Note for the 2017 election (held November, 2017) we use education and area data from late 2016/early 2017. For robustness analyses that include elections prior to 1993, late 1990 data on education are used for 1985 and 1989 as this variable does not go further back in the register used.

5. Note that as the unit of analysis is post-reform municipalities, so these are also the basis for the matching.

6. The index is calculated as the sum of the squared proportions of eligible voters from each old municipality in the new municipality.

Additional information

Notes on contributors

Yosef Bhatti

Yosef Bhatti’s research at the Danish Centre of Applied Social Science, Copenhagen, Denmark, focuses on voter turnout, political representation and public reforms. His work has been published in journals including the British Journal of Political Science, Electoral Studies, the Journal of Elections, Public Opinion & Parties, Public Choice, Social Forces and West European Politics.

Kasper M. Hansen

Kasper M. Hansen is Professor of Political Science at the University of Copenhagen, Denmark. His research focuses on political behavior: elections, referendums, voting behavior, turnout, and public opinion. Recent work has appeared in journals including the British Journal of Political Science, Political Behavior, Party Politics, Social Forces, Electoral Studies, the American Journal of Evaluation, Acta Politica, the Journal of Elections, Public Opinion and Parties, International Migration Review www.kaspermhansen.eu and West European Politics.