Using the difference-in-difference and the synthetic control methods, we empirically show that vote participation significantly dropped, also compared to neighboring cantons and the rest of Switzerland. This suggests that (major) centralizing constitutional changes tend to undermine the relationship between citizens and their state.
Conclusions
One of the major expectations has regularly been that a smaller number of communes leads to efficiency gains. However, the literature has shown that the expected economic benefits of merging communes in Switzerland are small and, in some cases, cannot be observed at all. Nevertheless, from 1990 until today, the number of communes in Switzerland has drastically decreased by about 30%. This reduction in the number of communes stands in contrast to the historical stability of the Swiss local government system and underpins the importance of research on commune mergers and their effects.
In the case of the canton of Glarus, where the number of communes has been by far, the most reduced in Switzerland, we find a significant decrease in the share of citizens participating in political decisions on a national level. The descriptive data on vote participation at the Landsgemeinde hints at a falling, rather than a rising, long-term participation rate. This trend suggests that the citizens of the canton Glarus to some extent have a reduced commitment to the three new communes and is consistent with our findings at a national level. This development is worrying as vote participation is a crucial indicator of the quality of participatory democracy and constitutes a crucial source of legitimacy. Such a decline in vote participation indicates a lower attachment of the electorate to democratic politics.
A strong driver is the diminished emotional connection to the place of residence. This unfortunate development may possibly reverse itself in the coming years when the population becomes more accustomed to the three large new communes. However, our analysis indicates that the fall in voter participation has not reversed within eight years after the merger from 25 to three communes. The result based on a DiD analysis in which we compare the development of vote turnout in Glarus to neighboring cantons, and all cantons overall has been supported by the synthetic control method. The empirical results range from a 4.4 to 5.5 percentage point decrease in vote participation in canton Glarus due to the merger. We obtain statistically significant and substantive results for the decision and implementation effect. Even though the differentiation between those effects is difficult due to the short period between the two shocks, we find strong evidence that supports the existence of the implementation effect. Nevertheless, it cannot be excluded that the citizens’ attachment to their political unit improves over a longer time span. This aspect must be left to future research in which the long-run consequences can be observed.
Other (political) institutions and thus different forms of political participation may also have suffered. According to the respective function within the local society, their degeneration might be more or less detrimental to the democratic relationship and, thus, decrease political participation even stronger. In short, social structure matters for political participation (Scheufele et al., 2004).
It remains to be answered how and under which circumstances these effects on citizens’ vote participation can be mitigated by the commune merger’s design, implementation, trust in political decision makers, individuals’ identification with the region, or other mechanisms. Those insights should play a crucial role in future decision-making processes concerning commune mergers. We suggest institutionalizing countermeasures to the residents’ alienation from the political process when communes are merged. For instance, residents should be involved in the decision-making about core aspects of identification value, such as the flag or name of the new commune. This is expected to foster local attachment and mitigate alienation from the political process. Overall, it is vital to substitute the altered opportunities for local political involvement, which we believe will mitigate the decrease in vote participation on the national level.
Radcliff (1994) argues that in the United States, even modest changes in turnout have major implications on party strategy and even presidential elections. A higher turnout is linked to higher participation of Democratic voters. This has implications for the commune merger in Glarus. If abstaining from political participation is systematically correlated to party preferences, the decisions taken will not represent the electorate properly. It might even be exploited by fostering or preventing commune mergers to maximize votes on a cantonal or national level. Incentives may be created, outweighing the interests of the local citizens and reducing their political participation.
Our analysis can be considered a quasi-natural experiment. While the number of independent communes was considered to be excessive for several reasons, the decision in the Landsgemeinde in 2006 to reduce their number from 25 to three was unexpected. Therefore, it was an exogenous shock.
The decision to radically reduce the number of communes in Glarus was taken by the citizens. In other countries, such decisions are taken by the political authorities, the parliaments, and the executive. This would suggest that the implementation effect should be of a similar size, independently of the decision-making actor. However, the decision effect should be smaller where the citizens are directly involved.
It might be argued that Glarus is a small canton with small communes in a small country. This is certainly true but allows us to study the effect of a radical commune merger on vote participation on a national level in detail. A major lesson to be drawn is not to neglect the effects of changes in local democracy on vote participation at the national level.
The consequences are beyond the economic ones, which in the case of Glarus, are rather small, if they exist at all. This means that individuals (and firms) are either less affected by such a drastic change or that they adjust more quickly than is the case for the more deep-seated relationship to the political environment. Our findings support this conclusion. Voters’ rate of political participation stays low after the merger for a substantial number of years. The body of citizens is not solely annoyed in the short run, but there are long-run consequences on how citizens feel and take political action in their political community.
This result should be seriously considered when communes are merged. It may help citizens understand the likely outcomes better, and it may inform governments when they impose reductions in the number of communes from above, as has been the case in other countries. As many countries consider reducing the number of communes in the future; the negative effects on political participation should be considered and mitigated.
