Compulsory Voting and the Turnout Decision Calculus
| Author | Shane P. Singh |
| Date | 01 August 2015 |
| DOI | 10.1111/1467-9248.12117 |
| Published date | 01 August 2015 |
| Subject Matter | Article |
Compulsory Voting and the Turnout
Decision Calculus
Shane P. Singh
University of Georgia
By altering the turnout decision calculus, compulsory voting should alter the character of the voting population.
Employing survey data across countries and Swiss cantons, I examine how the turnout decision calculus varies across
voluntary and compulsory voting systems. Results indicate that many of the demographic, socio-economic and
political factors known to correlate with turnout play a relatively weak role in motivating electoral participation where
voting is mandatory. Thus, voting populations should be more reflective of the entire electorate in countries with
compulsory voting. I conclude with a discussion of the potential implications of my findings.
Keywords: compulsory voting; voter turnout; participation; voting behavior
Compulsory voting requires citizens to show up to the polls on election day under threat
of a sanction for non-participation. Such compulsion boosts overall turnout levels (Birch,
2009; Blais, 2006; Gosnell, 1930; Robson, 1923; Tingsten, 1937), especially when rules are
strongly sanctioned and enforced (Panagopoulos, 2008; Singh, 2011). Yet while there is a
burgeoning literature on the topic, less is known about the second-order effects of
compulsory laws. Arend Lijphart (1997) argues that compulsory voting irons out demo-
graphic and socio-economic differences in the voting population by inducing less
informed, less interested and poorer groups to come to the polls. Lijphart’s conjecture
implies that the effect of these common forerunners of turnout on the decision to vote will
be weaker where voting is mandatory.
Previous work addresses this conjecture, but findings are inconsistent and often con-
flicting (compare Carlin and Love, 2013; Gallego, 2010; Hooghe and Pelleriaux, 1998;
Irwin, 1974; Jaitman, 2013; Maldonado, 2011; Quintelier et al., 2011; Singh, 2011;
Söderlund et al., 2011). A puzzle arises from these clashing empirical findings, and the first
contribution of this study is an attempt to solve this puzzle through a theoretical discussion
of the possible second-order effects of compulsion and a new empirical examination. I
undertake the broadest empirical analysis used to address this question to date, employing
an extensive cross-national collection of election surveys, as well as electoral survey data
from Switzerland, which employs compulsory voting in select cantons. I also place a novel
focus on the interaction of compulsory voting and politically oriented forerunners of
turnout – perceived efficacy of the democratic process and partisan attachments.
Evidence from multi-level logistic regression analyses indicates that age, income, politi-
cal knowledge, perceived efficacy and partisanship play weaker roles in motivating electoral
participation in compulsory systems. Further, this moderating effect of compulsion tends
to be most pronounced where compulsory rules are routinely enforced and strongly
sanctioned. Compulsory voting affects the character of the voting population by altering
bs_bs_banner
doi: 10.1111/1467-9248.12117
POLITICAL STUDIES: 2015 VOL 63, 548–568
© 2014 The Author. Political Studies © 2014 Political Studies Association
individuals’ turnout decision calculuses, and compulsory voting has the biggest effect on
the decision calculus where rules have teeth. This implies that, in line with Lijphart’s (1997)
original conjecture, mandating participation can iron out disparities in the voting popu-
lation. I conclude with a discussion of the potential implications of this finding.
Compulsory Voting and the Correlates of Turnout: Previous Literature
Compulsory voting is not uncommon – about 30 per cent of the countries in the
democratic world mandate participation on election day (Massicotte et al., 2004, p. 160).
A robustly established consequence of punishing those who abstain is an increase in
participation levels (e.g. Birch, 2009; Blais, 2006; Gosnell, 1930; Robson, 1923; Tingsten,
1937). Moreover, Lijphart (1997) argues that forced voting works to remove demographic
and socio-economic differences in turnout patterns by inducing participation among those
who are inclined to abstain. Lijphart’s argument is plausible, as younger, poorer and less
educated and knowledgeable individuals are relatively unlikely to vote (Almond and Verba,
1963; Blais, 2000; Palfrey and Poole, 1987; Wolfinger and Rosenstone, 1980).
There is some evidence in support of Lijphart’s contention. For example, Galen Irwin
(1974) finds that educational and age-based differences in turnout propensities are more
extreme under voluntary voting, Singh (2011) finds that obligatory voting diminishes
age-based differences in turnout and Aina Gallego (2010) finds the link between education
and turnout to be weaker under compulsory voting than under voluntary voting. Studying
elections to the European Parliament, Peter Söderlund et al. (2011) find that the link
between political interest and turnout is substantially weaker in countries with compulsory
voting, while Marc Hooghe and Koen Pelleriaux (1998) show that compulsory voting in
Belgium helps assure that the socio-economically privileged are not over-represented at the
voting booth. In Argentina, Laura Jaitman (2013) shows that compulsion may iron out
turnout disparities across skilled and unskilled workers, and Ryan Carlin and Gregory Love
(2013) find that partisanship and political interest are relatively weak predictors of turnout
in Latin American countries with enforced compulsory voting. Still, none of these studies
find a consistent suppressive effect of compulsory voting on the salience of the various
predictors of turnout.
Further, Arturo Maldonado (2011) shows that the degree to which various demographic
and socio-economic factors affect turnout propensity varies little across compulsory and
voluntary systems in Latin American countries. In addition, Ellen Quintelier et al. (2011),
in a broader comparative analysis, show that the effects of gender and education on the
likelihood of turning out do not vary across compulsory and voluntary systems. And
interestingly, they show that age actually has a stronger relationship with turnout propen-
sity where voting is required. Thus, whether compulsory rules affect the salience of the
demographic and socio-economic correlates of turnout remains an open question.
The Decision Calculus under Compulsory and Voluntary Rules
All else being equal, an individual’s expected utility – and likelihood – of voting is higher
when the cost of voting is low and when he or she derives some benefit from the act of
voting (e.g. Downs, 1957; Palfrey and Rosenthal, 1983; Riker and Ordeshook, 1968).
These associated costs and benefits, which are often unobservable across individuals, are
TURNOUT DECISION CALCULUS 549
© 2014 The Author. Political Studies © 2014 Political Studies Association
POLITICAL STUDIES: 2015, 63(3)
To continue reading
Request your trial