Miller, P. and R. Dassonneville (2016): “High turnout in the Low Countries: Partisan effects of the abolition of compulsory voting in the Netherlands”, Electoral Studies 44, 132-143. DOI: 10.1016/j.electstud.2016.07.009
The literature on turnout and vote choice usually emanates from the presumption that left-wing parties benefit from high turnout rates and the introduction of compulsory voting. Recent studies, e.g. the one of Bechtel et al. (2016) discussed on this blog, also find evidence for this relationship. In their newly published paper Miller & Dassonneville cast doubt on this conventional wisdom by showing that the abolition of compulsory voting in the Netherlands in 1970, which was connected to a sharp decline of overall turnout, led to an increasing vote share of Dutch social democratic parties. Furthermore, the authors find partial evidence that also the Christian democratic and liberal parties could benefit from the 1970 electoral reform. However, they set out that these outcomes for the two latter party families were not directly caused by the abolition of compulsory voting. Finally, the paper finds that the reform led to a significant loss in the share of votes for minor and extreme parties.
The study scrutinizes the Dutch case as a unique testing ground for a quasi-experimental research design. To assess the effect of the respective treatment condition (voluntary voting) in the Netherlands, the analysis includes Belgium as a control case. The two countries depict most similar cases in many respects but show the relevant difference that compulsory voting was abolished in the Netherlands while in Belgium it is in place until today. Also turnout numbers remained quite stable in Belgium – in contrast to the Netherlands. With an elaborate methodology the authors analyze electoral data at the provincial level from these two cases, covering the period between 1946 and 2014. Miller & Dassonneville make use of a fixed effects and a multilevel model to examine the effect of voluntary voting and run a matching model for further validation.
In sum, the article applies an innovative research design bringing some tantalizing results about. However, although the authors make use of several control variables and include Belgium as a control case, it is finally not fully possible to rule out all potential exogenous factors, like e.g. international societal changes or economic developments, which could – at least partially – account for the differences in party vote shares before and after the Dutch reform on compulsory voting. In the end, it seems that the relationship between compulsory voting and the vote share of left parties is quite depending on the studied case, what could explain the contradictory findings in the literature. Consequently, this paper shows that the scientific debate on that topic has not ended yet and further research in this direction could help to shed light on the connection between electoral laws, voter turnout and its partisan effects.
Author: Julian Noseck in December 2016