![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
We study the effects of municipal mergers on voter turnout in a difference-in-differences framework, using data from a wave of municipal mergers in Finland in 2009. Analysing two pre-merger elections and three post-merger elections, spanning a total of 17 years, we find that municipal mergers decrease voter turnout by 4 percentage points in the long run in the relatively small municipalities compared to similar small municipalities that did not merge. As the average turnout rate prior to merging in this group was around 69%, this is a substantial effect. We also find that virtually nothing happens to turnout in the municipalities that were relatively large within their merger. Furthermore, mergers are associated with a decrease in voters’ political efficacy and turnout decreases more in those municipalities that experience larger decreases in efficacy.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Earlier literature also suggests that political considerations have played a role in the merger decisions in different countries (Bhatti and Hansen Citation2011; Saarimaa and Tukiainen Citation2014; Hyytinen, Saarimaa, and Tukiainen Citation2014; Bruns, Freier, and Schumann Citation2015).
2. Currently, there is a plan in place to introduce a new middle tier from 2020 onwards, which will be responsible for health and social care, for example.
3. The 2007 mergers happened in the middle the council term. The 2008 elections are therefore different for these municipalities than for those in the 2009 mergers, because the earlier ones have already experienced some of the effects of the merger on service quality and taxes. Including the 2007 mergers would therefore make the interpretation of the effects difficult.
4. For example, polling districts in 2012 and later might take the name of the pre-merger municipality. In those cases, we assume that they correspond to the pre-merger municipality.
5. The dropped municipalities are somewhat larger in terms of population and have more polling districts than the ones we were able to match across years. However, as we show later, we have enough data that the assumptions for difference-in-differences still hold in our context.
6. Some studies separate between internal and external efficacy (Lassen and Serritzlew Citation2011). We do not have data to do so.
7. Pekola-Sjöblom (Citation2014) reports that the sample of the municipalities in the study were chosen so that they would be representative of all the municipalities in Finland. The number of individual respondents per municipality ranged from 13 to 388. We have checked that our results hold when using only the municipalities with at least a 100 respondents.
8. See Saarimaa and Tukiainen (Citation2014) and Harjunen, Saarimaa, and Tukiainen (Citation2017) for further details on the procedure.
9. Following Harjunen, Saarimaa, and Tukiainen (Citation2017), we use exact matching with respect to number of municipalities in the merger. The matching was based on the following covariates: total population of the merger, median distance of the citizens to the business centre of the largest municipality in the merger, indicator for whether all the partners belong to the same health care cooperation unit, and within-merger heterogeneity in per capita taxable income, expenditures and deficit.
10. We have selected these measures based on previous work by Lassen and Serritzlew (Citation2011) and Koch and Rochat (Citation2017). In our data, these measures are highly correlated (−0.61), but we use both of them in order to maintain comparability with previous literature.
11. We could also conduct the analysis for the three groups simultaneously by also including dummy variables for each group and their interactions. Because this pooled model would be fully saturated, that is, would include all the group dummies and interaction terms with the merger and time dummies, the results would be exactly the same.
12. Tables A1 and A2 in the Appendix show results when splitting municipalities in four groups instead, using the four quartiles of treatment intensity. Our conclusions are similar: municipalities in the highest treatment intensity quartile are the ones experiencing a significant decline in turnout in 2012 and 2017.
Additional information
Notes on contributors
Simon Lapointe
Simon Lapointe is a senior researcher at VATT Institute for Economic Research, Finland. His research interests are political, public and urban economics.
Tuukka Saarimaa
Tuukka Saarimaa is an associate research professor at VATT Institute for Economic Research, Finland. His research focuses on public and urban economics.
Janne Tukiainen
Janne Tukiainen is an associate research professor at VATT Institute for Economic Research, and a visiting associate professor at the Department of Government at the London School of Economics and Political Science, UK. His research interests include industrial organisation and political and public economics.