A global analysis of how losing an election affects voter satisfaction with democracy
| Published date | 01 September 2019 |
| Date | 01 September 2019 |
| DOI | http://doi.org/10.1177/0192512118779213 |
| Subject Matter | Articles |
https://doi.org/10.1177/0192512118779213
International Political Science Review
2019, Vol. 40(4) 518 –534
© The Author(s) 2018
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0192512118779213
journals.sagepub.com/home/ips
A global analysis of how losing
an election affects voter
satisfaction with democracy
Benjamin Farrer
Knox College, USA
Joshua N Zingher
Old Dominion University, USA
Abstract
In this article, we argue that a deeper understanding of citizen satisfaction with democratic elections requires
a global perspective. Regional research found that a gap in satisfaction with democracy emerges after an
election, between those who supported winning parties and those that did not, and also, crucially, that this
gap can be reduced under proportional electoral institutions. In this article we argue instead that these
theories of the winner–loser gap actually apply to only a narrow set of countries. We use a comprehensive
global dataset to show that the predictions of this theory about the effects of proportional institutions
are accurate for Western Europe, but not outside it. Beyond a small cluster of established democracies
in Western Europe, the electoral environment is characterized by more fundamental uncertainty. This
uncertainty alters the incentives created by proportional institutions. We conclude that the winner–loser
gap and ‘losers’ consent’ are concepts that vary systematically around the world. We discuss the implications
of this for democratic stability.
Keywords
Losers’ consent, democracy, elections
Introduction
The term ‘losers’ consent’ refers to voters who continue to endorse their democracy, even after that
democracy declares that their preferred party has lost the election. High levels of losers’ consent
are an unmistakable marker of democratic stability (Dowding and Kimber, 1983; Manin et al.,
1999; Przeworski, 2015). But losers’ consent is never guaranteed. Voters often react negatively in
Corresponding author:
Benjamin Farrer, Department of Environmental Studies, Knox College, 2 E. South Street, Galesburg, IL 61401, USA.
Email: bdfarrer@knox.edu
779213IPS0010.1177/0192512118779213International Political Science ReviewFarrer and Zingher
research-article2018
Article
Farrer and Zingher 519
the wake of elections where their chosen party did not make it into government. Contemporary
academic literature argues that levels of losers’ consent are, in large part, determined by a country’s
electoral rules. Using data from Western Europe, Anderson et al. (2005), demonstrate that the pres-
ence of proportional electoral institutions is positively associated with losers’ consent. This is also
the flagship finding from an important body of literature on institutions and satisfaction with
democracy (Aarts and Thomassen, 2008; Anderson and Guillory, 1997; Banducci and Karp, 2003;
Bernauer and Vatter, 2012; Fjelde and Höglund, 2016; Lijphart, 2012).
In this article we assess whether this institutional argument can be extended around the world,
or whether it is limited to Western Europe. We argue that in Western Europe, although satisfaction
with democracy may vary, there is not a high degree of fundamental uncertainty about the stabil-
ity of democratic institutions (Przeworski, 2015). We suggest that in the presence of such uncer-
tainty – caused by clientelism and by shorter experience with democracy – institutional arguments
may lose much of their vigor. Institutions create incentives for mass and elite actors to behave in
certain ways, but if information is scarce and other players’ strategies are hard to anticipate, then
political actors may not respond to these incentives (Andrews and Jackman, 2005; Boix, 1999;
Fey, 1997; Persson and Tabellini, 2003; Pierson, 2004; Selb, 2012; Shvetsova, 2003). Thus, the
effect of institutions on satisfaction with democracy will be different in established versus emerg-
ing democracies.
The article proceeds as follows. In the first section, we define the concept of losers’ consent. We
then argue that democracy is more established in Western Europe, and that institutional explana-
tions for losers’ consent are less powerful outside this region. The second section outlines our
research design. We introduce additional survey data from Africa, Asia, and Latin America – all
regions with shorter histories of democracy. We also describe how length of experience with
democracy affects how institutions operate differently outside Western Europe. The third section
describes our analysis. We find that the argument of Anderson et al. (2005) holds up well in Western
European democracies, but that electoral rules have a weaker and less consistent effect elsewhere.
The results of our analysis suggest that the effect of institutions is conditional upon how long a
country has been a democracy. The difference between majoritarian and proportional institutions
only becomes pronounced when a country is a well-established democracy. Our final section dis-
cusses the implications of these findings for the study of democratic stability.
Losers’ satisfaction with democracy and the effects of institutions
The concept of losers’ consent begins with the proposition that democratic elections create winners
and losers, and those two groups may have meaningfully different views about democracy.
Anderson et al. (2005) and Anderson and Guillory (1997) provide the most comprehensive assess-
ment of the winner–loser gap in satisfaction. They define losers as individuals who identify with
parties that did not form a government after the election. They found that winners were consistently
more satisfied with democracy than losers, but the size of this winner–loser gap varied by country.
Democracies with more proportional electoral systems had higher overall satisfaction and smaller
winner–loser gaps in satisfaction than more majoritarian systems.
Thus, Anderson et al. (2005) posit that electoral institutions are a key determinant of satisfaction
with democracy. In first-past-the-post electoral systems, electoral districts produce only one win-
ner, whereas in proportional representation (PR) electoral systems, there are more winners; and,
with more seats available, seat shares more accurately reflect vote shares (Cox, 1997; Lublin,
2014). In PR systems, voters who did not support the first-placed party will feel more satisfied with
democracy, because their preferred party is more likely to have gained at least a number of seats.
Anderson and Guillory provide a succinct summary:
To continue reading
Request your trial