Predicting Ambiguity
| DOI | 10.1177/1065912917691139 |
| Author | Daniel J. Lee,Christian B. Jensen |
| Date | 01 June 2017 |
| Published date | 01 June 2017 |
| Subject Matter | Articles |
research-article2017
Article
Political Research Quarterly
2017, Vol. 70(2) 301 –313
Predicting Ambiguity: Costs, Benefits,
© 2017 University of Utah
Reprints and permissions:
sagepub.com/journalsPermissions.nav
and Party Competition
https://doi.org/10.1177/1065912917691139
DOI: 10.1177/1065912917691139
journals.sagepub.com/home/prq
Christian B. Jensen1 and Daniel J. Lee1
Abstract
We present a formal model of party competition to explain differences in party ideological ambiguity. Existing works
generally argue that parties are more or less ambiguous depending on whether their supporters are risk-acceptant
or risk-averse. Our model explores more fully strategic choice of ambiguity by considering nonelectoral benefits to
ambiguity—that is, party elite recruitment and retention. In terms of costs, we assume that all voters are risk-averse,
who therefore prefer less ideological ambiguity. Explicitly considering both costs and benefits derives our hypotheses
and highlights the importance of party competition—ambiguity is influenced by the proximity to a party’s closest
ideological competitor. An empirical analysis of twenty-eight European countries supports our hypotheses.
Keywords
ambiguity, electoral competition, party competition
Since the seminal work of Downs (1957), researchers of
moderate voters are risk-acceptant, whereas ideologi-
political parties have studied, both through formal mod-
cally extreme voters are risk-averse. Thus, a party’s
els and empirical analyses, how electoral competition
choice of ambiguity is influenced by voters but not
influences party positioning. Within this framework, the
(directly) by competing parties. In our theory, how-
question asked is how parties choose their ideological
ever, the choice of the level of ambiguity is condi-
location, given voters support the party closest to their
tioned by the ideological proximity to competing
ideal point. Some scholars have also considered the
parties—that is, we show how ambiguity is explicitly
extent to which parties are clear or ambiguous in their
influenced by party competition. Our formal model
positioning. This question of ambiguity actually has its
assumes ambiguity has an electoral cost (risk-aversion
roots in Downs’ original work, which makes these stud-
of voters) but elite membership benefit (diversity to
ies vital to truly understanding the Downsian model. As
pacify intraparty conflict). The main theoretical result
an additional factor that impacts vote choice, ambiguity
is that the level of ambiguity is a function of its ideo-
can influence the party’s choice in the level of ambigu-
logical distance from the closest competing party.
ity (e.g., Rovny 2012; Somer-Topcu 2015) or shifts in
The remainder of this paper is divided into several sec-
positioning (e.g., Adams et al. 2006; Lo, Proksch, and
tions. In the first, we review the existing literature on
Slapin 2016).
position taking and ambiguity, which further clarifies our
We extend the literature by demonstrating both theo-
contribution, and present some real-world examples.
retically and empirically that the level of ambiguity is a
Second, we present a decision-theoretic formal model
strategic choice of the party that is driven by party com-
from which we derive testable hypotheses. We then ana-
petition. That is, much like how parties choose their loca-
lyze data from the Chapel Hill Expert Survey (CHES)
tion to maximize electoral success, we consider how they
project and find support for our hypotheses. We conclude
also choose the level of ambiguity to maximize some util-
with a brief discussion of our findings and their implica-
ity. Although some previous works touch on its causes,
tions for future research.
we find existing explanations incomplete, which we clar-
ify further in the next section.
Our explanation is rooted in party competition, as
1University of Nevada, Las Vegas, USA
the strategic choice of the level of ideological ambigu-
ity is conditioned on characteristics of competing par-
Corresponding Author:
Daniel J. Lee, University of Nevada, Las Vegas, 4505 Maryland
ties. The conventional wisdom is that moderate parties
Parkway, Box 455029, Las Vegas, NV 89154, USA.
are more ambiguous than extreme parties because
Email: dan.lee@unlv.edu
302
Political Research Quarterly 70(2)
Ambiguity and Party Competition:
risk-averse, which captures its costs. The assumption of
Accounting for Costs and Benefits
risk-aversion is quite common in models of electoral
competition with uncertainty (e.g., Berger, Munger, and
There is a great deal of research on the political positions
Potthoff 2000; Downs 1957; Enelow and Hinich 1990;
adopted by parties. Based on a spatial representation of
Grynaviski 2010; Hinich and Munger 1997; Snyder and
politics, researchers assess how electoral competition
Ting 2002). Second, we consider the nonelectoral bene-
influences those positions. Common foci in this literature
fits of ambiguity, in terms of party elite recruitment and
are the incentives to parties to shift their positions (Adams
retention. We elaborate on this point below. But the logic
2012; Adams, Haupt, and Stoll 2008; Schumacher, de
is that adopting a “big-tent” strategy in allowing for a
Vries, and Vis 2013) and voters’ attention to those shifts
wider range of viewpoints (i.e., more ambiguity) helps
(Adams and Somer-Topcu 2009). Still other researchers
pacify party elites from defecting to a different party and
have focused on the differences and competition between
attract new party members. There is indeed considerable
mainstream and niche parties (Ezrow 2008; Meguid
evidence that parties compete with each other for the loy-
2008; Spoon 2011) or between governing and opposition
alty of legislators and other elites (Desposato 2006;
parties (Bawn and Somer-Topcu 2012; Schumacher et al.
Heller and Mershon 2005, 2008; Mershon and Shvetsova
2015).
2007). Furthermore, there is a great deal of research on
Most of the existing work on party positioning starts
internal politics of political parties (cf. Budge, Ezrow,
from the premise that positions on issues are precise and
and McDonald 2010; Carey 2009; Kam 2009). In both
clear to voters. However, a parallel research agenda in
cases, ambiguity might play a significant role. Thus, the
American and, increasingly, comparative politics explic-
sole focus of previous studies on the electoral effects of
itly examines the inherent ambiguity in party positions
ambiguity potentially misses an important aspect of the
(Bartels 1986; Berger, Munger, and Potthoff 2000;
choices confronting parties.
Berinsky and Lewis 2007; Dewan and Myatt 2008; Lo,
We argue that the salience of these costs and benefits
Proksch, and Slapin 2016; Morgenstern and Zechmeister
is a function of the threat (proximity) from ideologically
2001; Rovny 2012; Shepsle 1972; Somer-Topcu 2015;
neighboring parties. In doing so, we highlight the impor-
Tomz and Van Houweling 2009). Rather than treat a par-
tance of party competition.4 That is, rather than simply a
ty’s platform as a specific point in a policy space, politi-
product of the distribution of (risk-acceptant vs. risk-
cal actors might instead view a party’s platform as a
averse) voters, ambiguity is influenced by the competi-
probability distribution, which captures beliefs of the
tion between a party and its ideologically neighboring
party’s true platform.
parties for the support of voters and elites.5
Previous research tends to focus largely on the effects,
Accounting for the costs and benefits also allows us to
rather than causes, of ambiguity—that is, the electoral
make predictions on varying levels of ambiguity, which
consequences of ambiguity (votes as the dependent vari-
prevailing conjectures have difficulty in doing. If a par-
able, ambiguity as an independent variable).1 Furthermore,
ty’s constituency is risk-acceptant and the level of ambi-
to the extent that studies have alluded to causes of ambi-
guity is a strategic choice by the party, then such a
guity, they have limited their attention to its electoral
theoretical model might not predict a “well-defined”
causes—specifically the preferences of voters.2 A com-
party strategy.6 Parties would simply want to choose the
mon conjecture is that voters at the ideological extremes
maximum level of ambiguity. But then what defines the
are risk-averse, which forces extreme parties to pursue
maximum? And, as they all want to maximize ambiguity,
ideological purity (e.g., Adams et al. 2006; Lo, Proksch,
how can we predict varying levels of ambiguity among
and Slapin 2016). In contrast, moderate voters are risk-
parties with risk-acceptant constituencies? Or conversely,
acceptant, which incentivizes moderate parties to adopt
among parties with risk-averse supporters, how can we
ambiguous platforms. We do not claim that such explana-
predict varying levels of ambiguity among these parties?7
tions of ambiguity, based on the risk preferences of differ-
Our model avoids this problem.
ent types of voters, are incorrect. Both risk-averse and
Similar to previous works, risk-acceptant political
risk-accept voters do likely play a role in influencing the
players are responsible for ambiguity in our theoretical
ambiguity of parties. That explanation alone, however,
model. We differ from previous...
To continue reading
Request your trial