Small versus Statewide Parties: How Subnational Contestation and Decentralisation Influence Presidential Elections in Europe and Latin America
| Author | Karleen Jones West,Jae-Jae Spoon |
| Published date | 01 February 2020 |
| Date | 01 February 2020 |
| DOI | http://doi.org/10.1177/1478929919862432 |
| Subject Matter | Special Issue Articles |
research-article2020
Special Issue Article
Political Studies Review
2020, Vol. 18(1) 145 –159
Small versus Statewide Parties:
© The Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
How Subnational Contestation https://doi.org/10.1177/1478929919862432
DOI: 10.1177/1478929919862432
journals.sagepub.com/home/psrev
and Decentralisation Influence
Presidential Elections in Europe
and Latin America
Jae-Jae Spoon1 and Karleen Jones West2
Abstract
Shugart and Carey were among the first scholars to recognise that there is a relationship between
regional and local – or subnational – electoral forces and the presidential race. Yet because of a lack
of subnational electoral data, this relationship has largely remained unexplored. We elaborate on
Shugart and Carey’s theory to argue that the effects of decentralisation are conditional on a party’s
presence in subnational elections for determining when and why parties enter the presidential
race. Using an original dataset of subnational electoral results and presidential strategies in 17
countries in Europe and Latin America from 1990 to 2013, we find that parties with a small
presence in subnational elections are more likely to compete for the presidency under more
extensive decentralisation. Statewide parties, however, contest presidential elections regardless of
level of decentralisation. These findings have important implications for understanding Shugart and
Carey’s expectation that subnational contestation influences national party systems, presidential
elections and democratic representation more generally.
Keywords
presidential elections, subnational politics, political parties, decentralisation
Accepted: 17 May 2019
Introduction
By the late 1990s, nearly every Latin American country and most countries in Europe had
experimented with decentralisation in one form or another, often as part of their efforts to
democratise (Cheema and Rondinelli, 2007: 8–10).1 From the adoption of decentralisa-
tion reforms (Escobar-Lemmon, 2003; O’Neill, 2003) to their implementation (Vlahos,
1University of Pittsburgh, Pittsburgh, PA, USA
2SUNY Geneseo, Geneseo, NY, USA
Corresponding author:
Karleen Jones West, SUNY Geneseo, 1 College Circle, Geneseo, NY 14454, USA.
Email: kwest@geneseo.edu
146
Political Studies Review 18(1)
2013), decentralisation has affected the electoral fortunes and, correspondingly, the strat-
egies of political parties around the world. For example, in Ecuador, decentralisation
provided the opportunity for the small indigenous-based Pachakutik party to win 68 seats
in local elections upon its formation in 1996, winning 70% of the local races where it
entered candidates (Van Cott, 2003). The party’s presence in local elections provided the
backbone for its entrance in the presidential race both that year and again in 1998.
Shugart and Carey (1992) were among the first scholars to recognise that there is a
relationship between regional and local – or subnational – electoral forces and the presi-
dential race. They argued that even small, localised parties may choose to nominate a
presidential candidate if they believe it would allow them to build enough political
momentum to become a major competitor in future elections. Furthermore, Shugart and
Carey (1992) also recognised that the relationship between local and presidential races
varies with election cycles. When elections are concurrent, parties can present a full slate
of congressional and presidential nominees in order to gain voter attention and increased
recognition, which could then help them in further elections. However, they noted that
when non-concurrent elections are held, parties tend to focus primarily on local issues
over national ones. This may result in the reduction of the salience of national issues to
the voters and thus slow the nationalisation of campaigns and parties as they come to
focus on local and particularistic interests.
Despite Shugart and Carey’s (1992) rich theoretical discussion of the relationship
between subnational contestation and the presidential race, recent research does not provide
a clear answer to the question of how local elections affect presidential elections (Hicken
and Stoll, 2008; Spoon and West, 2015b; West and Spoon, 2013). The question remains – do
subnational elections give parties the experience and resources necessary to compete in
national races (Jones, 1997; Moreno, 2003), leading local parties to run for the presidency
as Shugart and Carey suggest? Or are local parties less likely to expend resources in national
elections, as other research (Chhibber and Kollman, 1998, 2004; Harbers, 2010; Samuels,
2003) has argued? Finally, despite Shugart and Carey’s (1992) emphasis on the relevance of
electoral cycles, there has yet to be an exploration of how election timing affects the rela-
tionship between subnational elections and the presidential race.
In this article, we revisit Shugart and Carey’s (1992) theory of how subnational elec-
toral forces affect the presidential race. We present a more nuanced argument of how
subnational elections shape the presidential race by focusing on how institutions – namely,
decentralisation and the presidency – create incentives for parties with differing subna-
tional electoral portfolios to enter the presidential race. Inspired by Shugart and Carey’s
(1992) assumption that parties use the presidential race primarily to build political
momentum – which we call credibility – we argue that parties with a small subnational
presence are more likely to desire such credibility because it builds their political reputa-
tions and benefits them in future elections. By contrast, statewide parties, by definition,
have extensive presence in subnational elections and therefore enter the presidential race
because they are viable competitors for the office. We argue further that a party’s presi-
dential entry strategy is conditional upon the level of decentralisation. Parties with a small
subnational presence are more likely to enter the presidential race under more extensive
decentralisation because running for the presidency enhances their credibility when com-
peting for powerful and relevant subnational offices. Statewide parties, on the contrary,
are not affected by the level of decentralisation when deciding to enter the presidential
race because they are likely to compete for the presidency regardless. Finally, we control
for Shugart and Carey’s (1992) expectation that election cycles exert a powerful influence
Spoon and West
147
over the decision of subnational parties to enter the presidential race, which is the first
empirical analysis of how election timing affects this decision.
This article also makes a methodological contribution to the study of presidential
elections with our original dataset, which differs from extant research on party nationali-
sation in significant ways. First, the observations in our data are individual parties and
their electoral performance in 17 countries in Europe and Latin America from 1990 to
2013. Gathering party-level data allows us to assess our theory that the impact of decen-
tralisation varies across parties with distinct electoral portfolios. Second, our dataset not
only records parties’ national-level strategies but also contains party-specific data on the
number of subnational districts in which a party ran for all parties in the highest level of
subnational races (governorships or subnational legislative bodies) for the countries in
our sample. We then tracked whether those parties ran a candidate for president in the
concurrent or subsequent presidential election. Because the data are at the party level,
we are able to define whether parties are small or statewide based on their presence in
subnational elections and assess how decentralisation impacts the presidential electoral
strategies of these distinct party types. This is an important contribution because although
there is a growing body of literature on party system nationalisation, only a few investi-
gations have focused on how national party systems form in presidential countries
(Amorim and Cox, 1997; Golder, 2006; Hicken and Stoll, 2008; Morgenstern et al.,
2009), and even fewer have focused on the influence of subnational elections in the for-
mation of national presidential party systems (Jones, 1997; Moreno, 2003; Samuels,
2003). To our knowledge, no one has yet investigated the conditional effects that decen-
tralisation has on individual parties and thus party systems. Our dataset provides us the
opportunity to demonstrate the significance of subnational elections and decentralisation
for arguably the most important of political offices in an increasing number of govern-
ments: the presidency.
The article proceeds as follows. In the next two sections, we review the literature on
decentralisation, multilevel elections and party strategy; define small and statewide par-
ties; and develop hypotheses that highlight our expectations of how these parties respond
to subnational elections. We then discuss our data collection, methodology, and test our
hypotheses about the effects that decentralisation and subnational presence have on the
presidential race. Next, we present our findings, which demonstrate that small and state-
wide parties adjust their presidential strategies differently in reaction to the level of
decentralisation in a...
To continue reading
Request your trial