A cause or a consequence? A quasi-experimental test of the duty-participation nexus

ABSTRACT

Civic duty has been perceived as a key determinant of turnout. That is, while dutiful citizens turn out in large numbers, those who conceive of voting as a choice do less so. The strong correlation between civic duty and turnout might be due to reverse causation, however. Specifically, individuals might adapt their sense of duty according to previous voting behavior. In this article, we leverage Belgium’s compulsory voting system and the age-based discontinuity for the right to vote, and estimate the effects of being treated with participation on civic duty. We do not find “treated” citizens to be more likely to report civic duty than their “non-treated” counterparts. This finding holds across a series of robustness tests, and suggests that civic duty is exogenous to the vote.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 These laws, which are applied to all election levels (Marien, Dassonneville, and Hooghe 2015), determine a fine of up to €10 for first-time abstainers, and up to €25 for second-time abstainers. Those who abstain in four elections over 15 years are removed from the register for 10 years, and cannot receive any nomination nor promotion nor distinction from public authorities. Although these sanctions have become less enforced (Hooghe and Pelleriaux 1998) and no fines have been levied since 2003 (Miller and Dassonneville 2016), most Belgians still vote because of the compulsory voting rules, implying a strictly enforced compulsory voting system.

2 Exploring the differential role of civic duty in compulsory and voluntary systems ourselves, we also find that the meaning and role of civic duty in compulsory voting settings is comparable to what holds in contexts where voting is voluntary (see Figures A1 and A2).

3 16 to 18-year-olds voted in a “mock” election, which in no way compares to the official election. Still, in additional tests, we exclude voters in this election from the analysis, as we discuss below. Importantly, we have no reason to assume that the relation between turnout and duty would be very different in national elections. The discontinuity in citizens’ participation in 2018 Ghent’s local election is large enough to allow a robust assessment of the turnout-duty nexus; indeed, 95.9% of those who were 18 and older report voting in this election. In addition, compulsory voting rules apply in both local and national elections (Marien, Dassonneville, and Hooghe 2015). Finally, Belgian voters do not think of local elections as of ‘second-order’ compared to the national level (Marien, Dassonneville, and Hooghe 2015); consequently, civic duty unlikely differs in national elections (for a discussion of the relation between election order and civic duty, see Goodman [2018]).

4 47 of 877 respondents in the treatment group abstained in the election. As a point of comparison, we also estimate a sharp RD model. The estimates of such a model should be interpreted as the intent-to-treat (ITT) effect of turnout on civic duty.

5 In Belgium, the cut-off for being in a higher grade is January 1st. Consequently, there is no reason to assume sorting around the age of eligibility. Furthermore, individuals in both the treatment and control groups are in the same school year, which means that there is not differential exposure to potential civic education initiatives at school.

6 While recent work from Galais and Blais (2016) shows no evidence of a gender gap in civic duty, thus contradicting past research (e.g. Carreras 2018), we take a more conservative approach and include gender as a control variable in our supplementary tests.

7 For this figure, we chose the number of bins so that “the binned means have an asymptotic (integrated) variability that is approximately equal to the variability in the raw data” (Cattaneo, Idrobo, and Titiunik 2020, 28). Plotting equal-spaced bins provide the same results (see Figure F1 in the Appendix).

Additional information

Notes on contributors

Fernando Feitosa

Fernando Feitosa is a Ph.D. candidate at the Université de Montréal.

Dieter Stiers

Dieter Stiers is a post-doctoral fellow of the Research Foundation – Flanders (FWO) at KU Leuven.

Ruth Dassonneville

Ruth Dassonneville is associate professor in the Department of Political Science at Université de Montréal where she holds the Canada Research Chair in Electoral Democracy.