Terrorism prevention with reelection concerns and valence competition
| Published date | 01 July 2019 |
| DOI | 10.1177/0951629819858666 |
| Author | Haritz Garro |
| Date | 01 July 2019 |
| Subject Matter | Articles |
Article
Journal of Theoretical Politics
2019, Vol.31(3) 330–369
ÓThe Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0951629819858666
journals.sagepub.com/home/jtp
Terrorism prevention with
reelection concerns and
valence competition
Haritz Garro
Northwestern University, Evanston, IL, USA
Abstract
In recent decades, fear of terrorism has increased among voters in liberal democracies. In
response, governments have adopted counterterrorist measures that curtail civil liberties, at
times prompting allegations that political expediency underlies such measures. I study politicians’
strategic design of counterterrorist policies in a model of valence competition under threat of
terrorism. The incumbent chooses a counterterrorist strategy that combines observable (state of
emergency) and unobservable (effort) actions. In equilibrium, emergencies and terrorist attacks
become endogenously informative about valence. The low-valence incumbent underprovides
effort relative to the high-valence incumbent, and at times declares a state of emergency in the
absence of a terrorist threat. Increasing voters’ information about the incumbent’s valence
improves politician selection, but at the expense of a higher incidence of unwarranted emergen-
cies and, under some conditions, terrorist attacks.
Keywords
Counterterrorism; valence competition; states of emergency; terrorism
1. Introduction
After 9/11, the fear of terrorist attacks significantly increased among the American
electorate. The U.S. government responded to the attacks by starting a war in
Afghanistan with the objective of dismantling al-Qaeda. Domestically, the Patriot
Act was passed, which included a battery of measures to facilitate law enforcement
agencies’ efforts to combat terrorism at the expense of curtailing privacy
Corresponding author:
Haritz Garro,Northwestern University, 2211 CampusDrive, Evanston, IL 60201, USA.
Email: haritzgarro@u.northwestern.edu
protections and civil liberties.
1
In March 2002, the Bush Administration
announced that the Department of Homeland Security would implement a color-
coded terror alert system. This system contained five levels of threat, each charac-
terized by a different color. Increases in the scale of threat were accompanied by
specific actions—such as increased police and security presence at high-profile tar-
get locations, border monitoring, and airport security restrictions—by federal
agencies as well as state and local governments.
Critics of the Bush Administration denounced the color-coded scale and argued
that the system was implemented to advance the political interests of the incum-
bent administration. On August 1, 2004, three days after John Kerry accepted the
Democratic presidential nomination in the National Convention, the alert level
was increased. This led Howard Dean to state in CNN that ‘I am concerned that
every time something happens that’s not good for President Bush, he plays this
trump card, which is terrorism’ (Nagourney and Halbfinger, 2004). Such specula-
tions of political manipulation resurfaced in 2009 when Tom Ridge, secretary of
Homeland Security during the Bush administration, published his memoir. Ridge
argued that when a tape from Bin Laden was revealed in October 2004, before the
general election, Attorney General John Ashcroft and Secretary of Defense
Donald H Rumsfeld pressed him for an increase in the alert level, even when there
was no technical basis to support such a decision. ‘Is this about security, or poli-
tics?,’ wrote Ridge (Ridge and Bloom, 2009).
2
In addition to the potential for polit-
ical manipulation, security experts also argued that the color system was ineffective
(Schneier, 2006), and that increases in the alert levels imposed sizable psychological
costs on the population (McDermott and Zimbardo, 2007). Willer (2004) analyzed
the changes to the color-coded alert levels during the February 2001 to May 2004
period, and found that terror warnings not only increased Bush’s approval ratings,
but also voters’ support of his management of the economy.
3
The intense political controversy that surrounded the color-coded alert system
epitomizes the suspicions of political manipulation that the implementation of
high-profile counterterrorist measures and their associated restrictions on civil lib-
erties customarily raise in liberal democracies. These suspicions have become
increasingly prevalent in recent decades as voters’ heightened fear of terrorist
attacks in the aftermath of 9/11 prompted extraordinary counterterrorist measures
in developed countries throughout the world. Although governments design coun-
terterrorist strategies to prevent or react to terrorist attacks, the design and execu-
tion of these strategies is not carried out in a vacuum. Instead, the process is
strongly influenced by electoral calculations. Voters are typically imperfectly
informed about the actual level of terrorist threat facing the country, and due to
the nature of the counterterrorist activity, cannot observe certain actions taken by
the government.
4
Such information asymmetries give rise to agency problems and
signaling motives that induce politicians to undertake observable counterterrorist
measures in the absence of terrorist threats, and to underprovide unobservable
actions when facing actual threats. In this paper, I present a model that aims to
capture the salient features of electoral competition under the threat of terrorism. I
study the distortions to the government’s counterterrorist strategy generated by
Garro 331
agency problems between voters and politicians, and identify factors that exacer-
bate these distortions.
An extensive body of literature has analyzed how the threat of terrorism affects
voting behavior and public opinion about counterterrorist policies, the economy,
or foreign policy (see, e.g., Davis and Silver, 2004; Hetherington and Suhay, 2011;
Malhotra and Popp, 2012; Merolla and Zechmeister, 2009). While the success of
the government’s counterterrorist strategy is a key metric of government perfor-
mance when voters perceive the country is facing a high level of terrorist threat, it
is not the only policy dimension that voters scrutinize and hold the government
accountable for. The state of the economy and the incumbent politician’s economic
valence are major drivers of elections (Alt et al., 2011; Lewis-Beck and Stegmaier,
2000; Margalit, 2011). Yet, in times of increased salience of terrorism—which may
occur due to an increase in the actual risk of terrorism but also merely due to
changes in voters’ perception—the fight against terrorism pervades all aspects of
the electoral battle. Although the counterterrorist capacity of a state is fixed in the
short to medium term,
5
and thus is unaffected by the identity of the incumbent
politician in power at a given time, I show that voters can learn about the incum-
bent’s valence by partially observing the government’s counterterrorist strategy
and its results. This happens because voters possess a blunt institutional tool—the
ability to reelect or remove the incumbent—to simultaneously ensure the selection
of high-valence politicians and discipline the incumbent’s provision of security
from terrorism. The equilibrium reelection strategy balances the rewards to avoid-
ing terrorist attacks and to possessing high valence. The end result is that incum-
bents face different incentives to exert costly counterterrorist effort depending on
their valence.
6
In this paper, I study how an uncertain threat of terrorism affects counterterror-
ist policy-making when politicians differ on their ability to manage the economy,
henceforth denoted as valence. In a two-period model, the incumbent politician—
knowing the level of terrorist threat facing the country—designs a counterterrorist
strategy that combines observable and unobservable actions. I model the observa-
ble action as a state of emergency: declaring an emergency imposes a loss of utility
on voters, as civil liberties, political rights, and freedom of movement are curtailed.
However, if the country is in fact facing a high terrorist threat, the state of emer-
gency is assumed to enhance the effectiveness of unobservable counterterrorist
effort. In turn, this increased effectiveness results in a lower future probability of
attack. Voters hold imperfect information about the level of terrorist threat facing
the country and about the incumbent’s valence. At the end of the first period but
before the election, they observe a noisy signal of the incumbent’s valence and
whether the terrorist group successfully perpetrated an attack. Based on this infor-
mation they either decide to reelect the incumbent or elect the challenger for the
second period.
Because voters observe whether the incumbent declared a state of emergency
and whether the country suffered a terrorist attack before the election, they can
condition their reelection strategy on these events, thus influencing the incumbent
politician’s counterterrorist strategy. However, in view of the second period, their
332 Journal of Theoretical Politics 31(3)
To continue reading
Request your trial