Backing up, not backing down: Mitigating audience costs through policy substitution
| Date | 01 July 2019 |
| Published date | 01 July 2019 |
| DOI | 10.1177/0022343319832641 |
| Author | Erik Lin-Greenberg |
| Subject Matter | Regular Articles |
Backing up, not backing down: Mitigating
audience costs through policy substitution
Erik Lin-Greenberg
Center for International Security and Cooperation, Stanford University
Abstract
Can a leader reduce the audience costs imposed for backing down completely on a threat by opting instead to ‘back
up’ to a less hawkish policy? Current research examines the political repercussions of making a threat and then taking
no action at all. Real world leaders, however, often ‘back up’ and implement policies that involve some action – for
instance, imposing sanctions after threatening military force, rather than backing down entirely. This article argues
that audience costs can be mitigated through policy substitution: backing up to less hawkish policies – that reduce
inconsistency between a leader’s words and deeds – may reduce audience costs. A series of original survey experiments
finds support for the argument and demonstrates that the population treats inconsistency as a continuum. The
findings have implications for domestic politics and crisis bargaining. Domestically, a leader who backs up faces lower
audience costs and is seen as more competent than one who backs down. Yet those on the receiving end of threats are
less likely to believe the future threats of a foreign leader who has previously backed up or backed down. Backing up
therefore degrades the credibility of crisis signals by making it difficult for rivals to distinguish between credible
threats and those that will be backed up.
Keywords
audience cost theory, crisis bargaining, policy substitution, public opinion about foreign policy
In 2012, President Obama declared a red line, threaten-
ing military action if the Assad regime used chemical
weapons. When Syrian forces launched a sarin attack
in 2013, the Obama administration instead invoked a
peaceful United Nations-mediated solution. Critics
claimed Obama had sullied the USA’s reputation, while
supporters believed he had achieved his policy objectives.
Can leaders reduce audience costs incurred for backing
down entirely from threats by implementing less hawk-
ish alternatives, like those taken by Obama in 2013?
Much research suggests that governments can credibly
signal their intentions during interstate crises by gener-
ating domestic audience costs. Since domestic constitu-
ents can punish their leaders for being inconsistent,
leaders have incentives to avoid bluffing and issue only
threats they intend to carry out, making threats credible
(Fearon, 1994; Schultz, 1998; Weeks, 2008). Empirical
studies on audience costs, however, have unrealistically
simplified the inconsistency between threats and actions,
typically treating whether a leader follows through as
dichotomous (Tomz, 2007; Levendusky & Horowitz,
2012; Kertzer & Brutger, 2016). In other words, leaders
either carry out their threat completely or back down by
taking no action at all. This simplification is problematic.
In practice, there can be significant variation in the
degree of inconsistency between a leader’s words and
deeds, something neglected in existing studies. Some
leaders follow through with their threats, while others
‘back up’ to a range of less escalatory actions. If leaders
face less punishment for backing up than for backing
down, they might renege on threats – albeit incremen-
tally – more than existing logics predict. Backing up can
subsequently degrade the credibility of crisis signaling by
making it difficult for rivals to discern between credible
threats and those that will be backed up.
This article introduces and tests an argument that
suggests leaders face lower audience costs when they
Corresponding author:
EL2723@columbia.edu
Journal of Peace Research
2019, Vol. 56(4) 559–574
ªThe Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0022343319832641
journals.sagepub.com/home/jpr
minimize the perceived inconsistency between their
words and deeds. They do this by ‘backing up’, which
I define as making a threat then subsequently taking a less
escalatory action – for either pre-meditated or post-hoc
reasons – instead of doing nothing at all. Doesbacking up
instead of backing down reduce audience costs? If so, why
does the public deal less harshly with leaders who back up
instead of backing down? Most importantly, what does
this mean for the credibility of signaling during crises?
To explore these questions, I synthesize the disparate
literatures on audience costs, consistency, and policy
substitution, and develop an explanation on the political
repercussions of partially reneging on threats. I test this
argument using original survey experiments that expose
respondents to a hypothetical invasion of an ally. I
manipulate whether the president threatens to deploy
ground forces to defend the ally and whether the presi-
dent follows through on his threat, backs up to airstrikes
or sanctions, or backs down altogether. I find that by
backing up rather than backing down, leaders suffer
lower audience costs. Analysis of qualitative data from
the survey suggests that much of this reduction can be
attributed to respondents’ beliefs that some action is
more consistent and desirable than reneging entirely.
The experiments also reveal that domestic and targeted
audiences see the future threats of a leader who backs up
as more credible than those of a leader who backs down.
This article makes three contributions to the study of
audience costs. First, introducing the concept of backing
up adds nuance to the inconsistency mechanism that
underlies audience cost theory. Existing studies that treat
threat fulfillment as binary overlook instances where
leaders lessen audience costs by taking some form of
action. Second, it confronts critiques that the public
cares little about inconsistency (Snyder & Borghard,
2011; Trachtenberg, 2012). The experiments demon-
strate that the public recognizes variation in the degree
of consistency between a president’s threats and actions,
which can affect audience costs and perceptions of a
leader’s reputation and competence. Third, it empirically
demonstrates that backing up along a continuum of con-
sistency degrades the credibility of a leader’s future
threats among a target population, potentially affecting
crisis bargaining.
Mitigating audience costs through policy
substitution
A large body of literature suggests domestic audiences
punish their leaders for making a foreign policy threat
and then reneging on it (Fearon, 1994; Schultz, 1998;
Tomz, 2007; Weeks, 2008). The political costs for bluff-
ing make leaders reluctant to issue idle threats (Fearon,
1994), helping to differentiate costly (and credible) sig-
nals from less credible cheap talk (Jervis, 1970; Fearon,
1994). Recent scholarship has identified several factors
that influence the consequences for reneging on threats.
Audience costs can be shaped by the salience of the issue
of stake (Clare, 2007), by the existence of an indepen-
dent media or whistleblowers (Potter & Baum, 2014),
and by the leader revealing new information about the
situation (Levendusky & Horowitz, 2012). Existing
research, however, has overlooked whether audience
costs are affected by the degree to which a leader follows
through with a threat.
Policy substitution
Before implementing a foreign policy, leaders generally
consider a range of possible options, rather than devel-
oping just one course of action. Research on policy sub-
stitution suggests that several strategies can be used to
address the same geopolitical issue (Most & Starr, 1984;
Palmer & Bhandari, 2000; Clark & Reed, 2005). For
instance, when deciding how to counter a rival, a leader
might choose between war (Gilpin, 1981), strengthening
alliances (Walt, 1990), levying sanctions, or bolstering
military power (Waltz, 1979). The ultimate decision is
informed in large part by the leader’s assessment of the
costs and benefits for each possible strategy, taking into
account a host of domestic and international factors
(Clark, Nordstrom & Reed, 2008).
Just as a leader considers multiple strategies before
enacting a foreign policy, a leader can also back up to
an alternate policy after threatening another strategy.
Existing audience cost research has touched upon policy
substitution, but does so by varying the initial threat or
demonstration of force rather than analyzing different
degrees of ‘backing up’ once a threat has been made
(Tomz, 2007). Critics might argue that backing up is
unlikely because leaders avoid making unambiguous
threats that commit them to specific actions (Snyder &
Diesing, 1978; Snyder & Borghard, 2011). Even ambig-
uous threats about ‘red lines’ and ‘grave consequences’,
however, can imply certain types of actions, from which
leaders can back up.
Backing up can be either premeditated or post-hoc.
When it is premeditated, a leader may issue a threat to
obtain concessions from a rival or to satisfy domestic
constituents, knowing that she will later back up to
another policy. In post-hoc cases, a leader may back up
after receiving new information about the situation
560 journal of PEACE RESEARCH 56(4)
To continue reading
Request your trial