Political party strength and electoral violence
| Author | Hanne Fjelde |
| Date | 01 January 2020 |
| Published date | 01 January 2020 |
| DOI | 10.1177/0022343319885177 |
| Subject Matter | Research Articles |
Political party strength
and electoral violence
Hanne Fjelde
Royal Swedish Academy of Letters, History and Antiquities, Uppsala University & Peace Research
Institute Oslo (PRIO)
Abstract
Existing research on the causes of electoral violence has focused on structural determinants and election-specific
characteristics but has paid less attention to the role of political agents that contest elections. This study addresses this
gap by examining the relationship between the organizational strength of political parties and the risk of electoral
violence. The study argues that strong political parties enhance the prospect for peaceful electoral dynamics for two
reasons. First, having strong party organizations reduce incentives for violent electoral manipulation because these
organizations enable more cost-efficient ways to mobilize voters. Second, strong party organizations constrain
political actors from deploying electoral violence, both at the leadership and grassroot levels. The relationship
between political party strength and electoral violence is studied by combining global data on the overall strength
of political parties in the polity with data on violence across all national elections from 1946 to 2010. The statistical
analysis accounts for a number of potentially confounding variables related to formal political institutions and
election-specific characteristics. The results point to a statistically significant and substantively important association
between strong political parties and a reduced risk of violent electoral conflict.
Keywords
elections, electoral violence, political parties
Introduction
As multiparty elections have spread to become the global
norm over the past two decades, violence during electoral
periods has become a significant concern for domestic
and international observers alike. Electoral violence
accompanies about a quarter of all national elections
worldwide (Hafner-Burton, Hyde & Jablonski, 2014).
The ramification of electoral violence on affected societ-
ies goes beyond its significant human and economic toll.
It holds the potential not only to shape immediate elec-
toral outcomes, but also to influence the overall pros-
pects for democratic transition and consolidation.
Against this background, a rapidly expanding aca-
demic literature has turned its attention to the question
of why some electoral processes become marred by vio-
lence whereas others do not. Much of the existing liter-
ature has highlighted the role of proximate factors related
to the electoral dynamics, for example the presence of
electoral observers (Asunka et al., 2017; Daxecker,
2012), the expected margin of victory (Hafner-Burton,
Hyde & Jablonski, 2014), or incumbency effects (Tay-
lor, Pevehouse & Straus, 2017). Less attention has been
paid to the institutional underpinnings of electoral com-
petition. Whereas studies have highlighted electoral rules
(Fjelde & Ho
¨glund, 2016) or strength of constraining
institutions (Hafner-Burton, Hyde & Jablonski, 2014;
Norris, Richard & Martı
´nez i Coma, 2015; Salehyan &
Linebarger, 2015), there are still gaps in our knowledge
of institutional determinants of electoral violence.
In particular, there is a lack of studies that look
beyond the legal and formal outlook of the state to focus
on the meso level, that is on collective actors that operate
within these institutions, including political parties. This
Corresponding author:
hanne.fjelde@pcr.uu.se
Journal of Peace Research
2020, Vol. 57(1) 140–155
ªThe Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0022343319885177
journals.sagepub.com/home/jpr
gap is particularly glaring in light of the growing micro-
level literature that examines violent electoral intimida-
tion at the level of the individual (Burchard, 2015;
Gutie
´rrez-Romero, 2014). Political parties and their
movements are transmission belts between individuals
and political elites. Understanding how they shape the
likelihood of violence could inform our knowledge of
both individual-level and elite-level behavior in this field.
Addressing this gap, this study focuses on how the
organizational characteristics of the political parties that
compete in the electoral arena in fluence the risk that
elections will be violent. I argue that in a polity where
political parties overall are strong, we are more likely to
observe peaceful electoral dynamics for two main rea-
sons.
1
First, strong party organizations reduce the incen-
tives for violent manipulation of the electoral contest
because strong parties have less costly means available
to mobilize voters. Second, strong party organizations
reduce the overall risk of electoral violence by placing
constraints on the ability of political actors to deploy
electoral violence, both at the leadership and the grass-
root level.
I examine the importance of political party strength in
a global, cross-sectional analysis of all national electoral
rounds from 1946 to 2010. The theoretical argument
focuses on how party characteristics shape incentives and
constraints at the level of the organization, but global
data on party strength are only available at the level of
the polity-year. The data, available from Varieties of
Democracy program, thus capture the predominant
characteristics of the most central parties in the polity
(Coppedge et al., 2017). For my independent variable, I
create a composite index of party strength, based on
information about the level and depth of the party orga-
nization; the presence of local party branches; the degree
to which parties have programmatic, rather than cliente-
list linkages to its constituencies; the degree of legislative
party cohesion; and the presence of centralized proce-
dures for candidate selection (Bizzarro et al., 2018;
Tavits, 2013). I combine these with data on electoral
violence from the NELDA dataset (Hyde & Marinov,
2012). The statistical analysis points to a significant rela-
tionship between strong political parties and a reduced
risk of electoral violence.
The article proceeds as follows. The next section out-
lines the research gap. The article moves on to introduce
the theoretical argument and the expectation to be
tested. The next section introduces the data and research
design, before I discuss the results from the empirical
analysis. The final section concludes.
Existing literature
The past decade has seen a surge in the literature exam-
ining the causes of electoral violence. Much of it adopts a
strategic interpretation of the violence: whereas some
might be incidental, much of it is deliberate, and levied
to influence the electoral process (Burchard, 2015; Stani-
land, 2014). As a form of illegal electoral manipulation,
violent electoral strategies are used to reduce uncertainty
about the electoral outcome, for example to disenfran-
chise voters, intimidate, demobilize or even kill opposi-
tion, or to influence vote choice at the ballot.
2
To explain variation in electoral violence, existing lit-
erature has pointed to variables that increase the incen-
tives of political contenders to resort to violent electoral
strategies. Factors that strengthen the competitiveness of
the race and raise the potential payoffs from swinging the
vote at the margin include the closeness of the vote
(Hafner-Burton, Hyde & Jablonski, 2014; Salehyan &
Linebarger, 2015; Wilkinson, 2004); whether the
incumbent is standing for re-election (Taylor, Pevehouse
& Straus, 2017); whether the country has seen a turn-
over in power (Ruiz-Rufino & Birch, 2020); whether
electoral rules produce winner-takes-all outcomes (Fjelde
&Ho
¨glund, 2016); and whether the electoral district in
question is underrepresented in the national votes-to-
seats allocation (Daxecker, 2020). Scholars have also
highlighted variations in the constraints on actors to
engage in violence, such as the institutional limits on the
decisionmaking powers of the executive (Hafner-Burton,
Hyde & Jablonski, 2014); the presence of international
monitors (Daxecker, 2012, 2014; Asunka et al., 2017) or
institutional weakness (Norris, Richard & Martı
´nez i
Coma, 2015; Salehyan & Linebarger, 2015). A growing
micro-level literature presents findings that are consistent
with instrumental interpretations of electoral violence.
Analysis of survey data, for example, suggests that vio-
lence is being used strategically to demobilize swing or
opposition voters, both through suppression and displa-
cement (Rauschenbach & Paula, 2019; Gutie
´rrez-
Romero, 2014; Gonzalez-Ocantos et al., 2020).
3
1
Party strength, in this context, does not refer to the vote share,
membership, or support-base of the party, but to its organizational
characteristics.
2
Post-election violence, in contrast, often erupts to protest electoral
results (e.g. Beaulieu, 2014; Tucker, 2007).
3
See also Bratton (2008), Collier & Vicente (2012).
Fjelde 141
To continue reading
Request your trial