Content uploaded by Anna Getmansky
Author content
All content in this area was uploaded by Anna Getmansky on Aug 11, 2019
Content may be subject to copyright.
Anna Getmansky and Thomas Zeitzoff
Terrorism and voting: the effect of rocket
threat on voting in Israeli elections
Article (Accepted version)
(Refereed)
Original citation:
Getmansky, Anna and Zeitzoff, Thomas (2014) Terrorism and voting: the effect of rocket threat
on voting in Israeli elections. American Political Science Review, 108 (3). pp. 588-604. ISSN
0003-0554
DOI: 10.1017/S0003055414000288
© 2014 American Political Science Association
This version available at: http://eprints.lse.ac.uk/89172/
Available in LSE Research Online: July 2018
LSE has developed LSE Research Online so that users may access research output of the
School. Copyright © and Moral Rights for the papers on this site are retained by the individual
authors and/or other copyright owners. Users may download and/or print one copy of any
article(s) in LSE Research Online to facilitate their private study or for non-commercial research.
You may not engage in further distribution of the material or use it for any profit-making activities
or any commercial gain. You may freely distribute the URL (http://eprints.lse.ac.uk) of the LSE
Research Online website.
This document is the author’s final accepted version of the journal article. There may be
differences between this version and the published version. You are advised to consult the
publisher’s version if you wish to cite from it.
Terrorism and Voting: The Effect of Rocket Threat on
Voting in Israeli Elections
Forthcoming in the American Political Science Review
Anna Getmansky Thomas Zeitzoff
Abstract
How does the threat of becoming a victim of terrorism affect voting behavior? Localities
in southern Israel have been exposed to rocket attacks from the Gaza Strip since 2001.
Relying on variation across time and space in the range of rockets, we identify the
effect of this threat on voting in Israeli elections. We first show that the evolution of
the rockets’ range leads to exogenous variation in the threat of terrorism. We then
compare voting in national elections within and outside the rockets’ range. Our results
suggest that right-wing vote-share is 2 to 6 percentage points higher in localities that
are within the range– a substantively significant effect. Unlike previous studies that
explore the role of actual exposure to terrorism on political preferences and behavior,
we show that the mere threat of an attack affects voting.
How does the threat of becoming a victim of terrorism affect voting behavior? Are voters
more likely to support parties that favor extending greater concessions to terrorists in order
to avoid further violence, or are they more likely to vote for parties that oppose concessions,
and that favor more aggressive policy towards the perpetrators? These questions have direct
Anna Getmansky is a Lecturer, Lauder School of Government, Diplomacy, and Strategy, Interdisciplinary
Center Herzliya, Israel. While working on this paper, she was a postdoctoral fellow at the Center for
International Relations in the Department of Social and Decision Sciences at Carnegie Mellon University.
E-mail: agetmansky@idc.ac.il
Thomas Zeitzoff is an Assistant Professor, the Department of Justice, Law and Criminology, School of
Public Affairs, American University. E-mail: zeitzoff@gmail.com
We thank Neal Beck, Bruce Bueno de Mesquita, Baruch Fischoff, Esteban Klor, John Miller, Uzi Rubin,
Tolga Sinmazdemir, the reviewers and the editor of the Review for their helpful suggestions. We also benefited
from the comments of participants of 2012 Peace Science Society (International), and the Symposium on
Political Violence at the University of Pittsburgh. Finally, we thank the Israel Social Science Data Center
and the GIS Center at the Hebrew University of Jerusalem for providing us with election results data and
shapefiles, respectively.
1
implications for conflict resolution, as well as for understanding and assessing the effectiveness
of terrorism as a political tool. The existing literature that focuses primarily on the actual
exposure to terrorism has been mixed, with some studies suggesting that terrorism leads
voters to be more accommodating of terrorists’ demands, and others showing a hardening of
attitudes (a rightward shift) in the electorate.
Our study improves on the existing studies in several ways. We explore the effect of
a particular method of terrorism–being in the range of rocket fire. In the context of the
Israeli-Palestinian conflict, rocket launchers are confined to the Gaza Strip, and therefore
the effect of rockets depends on the maximum distance that these rockets can travel from
Gaza. We show that the rockets’ range is exogenous to the Israeli political circumstances,
and is determined by the availability of rocket technology and launching expertise in Gaza.
This allows us to assess the effect of rockets’ threat on voting. Specifically, we explore the
variation in the range of rockets from 2001 through 2009, and show that voters who reside in
the range are more likely to vote for right-wing parties. We estimate that the effect of being
in the rockets’ range is an increase of 2 to 6 percentage points in the right-wing vote-share,
and this implies additional 2 to 7 seats in the Israeli Knesset (parliament) out of 120 for the
right-wing bloc, if all voters in Israel were within the rockets’ range.1
We also unpack changes in different subsets of the right-wing bloc, as well as changes
in vote-shares of specific parties. We find that within the right-wing bloc, rocket range
primarily affects the nationalistic parties (rather than religious or Jewish-sectarian parties
1It is worth noting that rocket fire from Gaza is not the only instance in which rockets were used in
the context of an intrastate conflict. According to Rubin (2011), the IRA, the Tamil Tigers, and Afghan
insurgents possessed ballistic weapons. The IRA and the Tamils used mortars or rockets very sparsely,
and only against government or military targets. The Afghan mujahideens initially used rockets against
the Soviet forces, but after the Soviet withdrawal, rockets of various calibers were used by both sides in
the Afghan civil war. In the Israeli context, the PLO and later the Hizbollah both used rockets, including
against civilian population. What differentiates our case from these conflicts is not necessarily the type of
weapons, their use, and effect, but the fact that rockets’ launchers are geographically constrained to Gaza.
It makes the range of rockets crucial to determining the extent of this threat, and it also allows us to identify
the rockets’ effect on voting.
2
that are also considered to be right-wing). Furthermore, we show that incumbent parties are
not punished by voters for allowing the range of rockets to expand. Instead, right-wing parties
gain additional votes in localities that are within the range when the incumbent is from a
non-right-wing bloc, and the Likud party gains additional votes within the rockets’ range
also when it runs as the incumbent. This finding supports the explanation that right-wing
parties, and in particular nationalist parties (Likud), have a valence advantage over others
when it comes to dealing with security challenges. We present evidence that this finding is
not due to the ‘rally round the flag’ effect because other incumbents do not experience such
an increase in their vote-shares.
Terrorism, Violence, and Political Behavior
Previous Work
How does living under the threat of terrorism affect voting? This question is related to three
strands of extant literature: 1) the effect of violence on political behavior, 2) the strategic
nature of terrorist - government interactions, and 3) retrospective voting and issue ownership.
Existing studies explore various effects of violence and terrorism. While Pape (2003)
suggests that terrorism, and in particular suicide bombings, are effective against democratic
voters who pressure their governments to provide concessions to terrorists, Abrahms (2006,
2012) questions the overall effectiveness of terrorism as a political strategy. Many papers
assert that terrorism and counterterrorism lead to cycles of violence (Haushofer, Biletzki and
Kanwisher, 2010; Jaeger and Paserman, 2008). However, others provide evidence that some
forms of counterterrorism might be effective in reducing further insurgent violence (Lyall,
2009), or in curbing future terrorist attacks (Benmelech, Berrebi and Klor, 2010a).
Looking at the Israeli-Palestinian context, Berrebi and Klor (2008) use the variation in
Palestinian suicide terrorism across time and space to argue that attacks shift Israeli vot-
3
ers toward right-wing parties that are less supportive of granting political concessions to
Palestinians. They estimate that one suicide attack within three months before an election
increases right-wing parties’ vote-share in the targeted locality by 1.35 percentage points.
Moreover, suicide attacks increase the support for the right-wing bloc beyond those local-
ities in which they take place. Gould and Klor (2010) use a similar identification strategy
to make a more nuanced argument: while terrorism increases the vote-share of right-wing
parties, it also shifts the entire political spectrum to the left, thereby making voters and
parties more supportive of accommodating Palestinian political demands. Yet both Berrebi
and Klor (2008) and Gould and Klor (2010) only examine fatalities from suicide terrorism.
However, fatalities from suicide terrorism could influence electoral behavior through one of
two channels. (1) Casualties caused by terrorism or war can have a strong effect on voting.
Previous research shows that support for war, and the incumbents in office, are sensitive
to fatalities (Kibris, 2011) and the context in which they occur (Gelpi, Feaver and Reifler,
2006). Additionally, the political psychology literature suggests that exposure to violence
influences intergroup attitudes (Canetti-Nisim, Halperin, Sharvit and Hobfoll, 2009) and
political conservatism (Bonanno and Jost, 2006). (2) Terrorism can also strike fear into a
targeted populace (Huddy, Feldman and Cassese, 2007), and affect voting through the fear
of future attacks rather than past casualties. By only looking at suicide terrorism, it is un-
clear whether shifts in the electorate are due to casualty aversion, or the fear that terrorism
creates. Using rocket attacks from Gaza, which caused relatively few casualties but provoked
widespread fear (IICC, 2009), we are able to isolate the effect of fear over terrorism on voting
behavior.
Outside the Israeli-Palestinian conflict, Kibris (2011) finds that Turkish nationalistic
parties’ vote-shares increase following fatal attacks by Kurdish insurgents against police
targets. However, in the case of Spain, Bali (2007) and Montalvo (2011) show that the
2004 Madrid bombings mobilized opposition voters and shifted the electoral map to the
4
left against an unpopular foreign policy.2Several studies suggest that political responses
to terrorism might be mediated by psychological variables. For instance, Canetti-Nisim,
Halperin, Sharvit and Hobfoll (2009) find that exposure to terrorism leads individuals to
adopt more exclusionary attitudes towards minorities. Especially relevant to the current
study, Zeitzoff (2013) shows that residents living in localities with high exposure to rockets
from Gaza show more altruism to their ingroup than those in low exposure localities. Other
studies are less sanguine, finding that rocket exposure is associated with increased adolescent
violence and aggression (Henrich and Golan, 2013). Habituation to terror can also shape
voters’ behavior. Becker and Rubinstein (2011) show that reduction of services and goods
purchased in Israel during the Second Intifada occurred only among occasional consumers,
since regular users of such services are able to adjust their behavior. Numerous other studies
draw a connection between war-time, nationalism and patriotism (Baker and Oneal, 2001;
Skitka, 2005). Huddy, Feldman and Cassese (2007) show that perceptions of threat post-
9/11 were more likely to favor aggressive military actions. Finally, Abramson, Aldrich,
Rickershauser and Rohde (2007) connect George W. Bush’s 2004 presidential election win
to his ability to appear to be a strong leader, an especially important factor considering the
import US voters placed on combatting terrorism. This is also confirmed by Oates, Kaid
and Berry (2009), who find that fear was an important component of Bush’s campaign in
2004 that projected Bush as a safe and secure leader for troubled times.
Review of these studies suggests that exposure to terrorism might increase, decrease, or
have no effect on electoral support for right-wing parties. What complicates this question
even further is the strategic nature of terrorist - government interactions. Powell (2007)
shows that both terrorists and governments choose actions that maximize their utility, given
their expectations regarding the other side’s actions. Terrorists choose the targets and the
2A dissenting view is offered by Lago and Montero (2006), who find that that the bombings solidified
already negative feelings about the incumbent government, rather than shifting the electorate.
5
timing of their attacks strategically. For example they may try to derail peace talks (Kydd
and Walter, 2002), trigger government response that would facilitate terrorist recruitment
(Bueno de Mesquita and Dickson, 2007), or maximize government concessions (Lake, 2002;
Pape, 2003). Thus, they target those populations that are more likely to respond in the
desired manner, either by voting for right-wing parties (if terrorists’ goals are to “spoil”
talks or facilitate recruitment), or by voting for left-wing parties (if terrorists’ goals are to
extract concessions). Governments, who are motivated to protect against attacks as well as to
increase their vote-shares, are also strategic in their response to terrorism, and protect targets
to minimize the overall expected impact of attacks (Powell, 2007).3Thus, in equilibrium,
terrorist attacks might have no effect on voting, or they might strengthen public support for
the incumbent.
Finally, previous research on electoral behavior can illuminate how terrorism affects vot-
ing. Theories of retrospective voting suggests that voters evaluate candidates based on their
past performance and achievements in areas such as economic policy (Fiorina, 1978; Kiewiet
and Rivers, 1984). Alternatively, Petrocik (1996) and Wright (2012) argue that some parties
have an advantage in certain areas, and they tend to benefit when their issues become salient.
In the Israeli context, Schofield and Sened (2005) and Arian and Shamir (2008) argue that
right-wing political parties enjoy a valence advantage on security issues. In the context of
conflict and terrorism, there is evidence that incumbents lose electoral support following
attacks and casualties, as expected if the retrospective voting theory is correct (Bali, 2007;
Gassebner, Jong-A-Pin and Mierau, 2008; Gelpi, Feaver and Reifler, 2006; Karol and Miguel,
2007; Montalvo, 2011). However, Berrebi and Klor (2008) and Koch and Tkach (2012) find
that in Israel incumbents are not punished for suicide attacks, and the right-wing bloc in-
creases vote-shares regardless of the incumbent’s political affiliation, in accordance with the
3For instance, Bueno de Mesquita (2007) demonstrates how governments use allocation of counterterror-
ism funds to maximize their reelection chances.
6
valence theory.
Using Rocket Range to Understand Terrorism
In this paper we provide several contributions to understanding the connection between
terrorism and voting. First, the nature of rockets and the circumstances of their use allow us
to circumvent the problem of strategic targeting. Palestinians who fire rockets are confined
to Gaza, and the impact of their attacks depends on rocket technology that determines the
range and accuracy of their rockets.4Thus, at any given point in time, those within the
rockets range are threatened not because they are perceived to be attractive targets, but
because this is how far rockets can reach. Within the range, the targets of rocket attacks are
strategically chosen. However, due to limited accuracy and varying launching skills, everyone
within the range has some likelihood of being hit. Crucial to our identification strategy is
that selection into the range is not driven by strategic considerations, but by availability of
technology and the constraint to launch rockets from within Gaza. In fact, we show that
those residing within the range do not differ substantially from those living outside the range,
and we attribute the difference in voting behavior to being threatened by rockets. This is an
improvement over existing studies, both in the Israeli context as well as beyond, that so far
focus on the impact of those terrorist attacks whose targets are not determined by a random
choice.
Second, by examining the range rather than the actual targeting, we argue that experi-
encing an attack is not necessary to influence voting, and that voting could be also affected
by the mere threat of being a target. Focusing on threat (living in the range of fire) is
also closer to the view of terrorism as a strategy that seeks to instill fear in a target popu-
4The borders between Gaza and Israel, as well as Gaza and Egypt are tightly controlled and surrounded
by a double-wire fence with only 6 crossing points (OCHA, 2007). Therefore, Palestinians who seek to launch
rockets can only do it from within Gaza, and thus the range of rockets is a crucial factor that determines
who is threatened by these attacks and who is not.
7
lace, rather than to destroy the opponent militarily (Abramson, Aldrich, Rickershauser and
Rohde, 2007).
Rocket Fire From Gaza
Since 2001, areas of southern Israel have been under a continual threat of ballistic weapons
(mortars and rockets) launched by a variety of militant groups in the Gaza Strip.5Following
the Israeli disengagement from Gaza in 2005, and the ascension of Hamas, there was a
dramatic increase in rocket fire and range (IICC, 2009). Whereas initially rockets could
reach up to 4km into Israel, by the 2003 election the range extended to 10km, and continued
to double between elections. As shown in table 1, by 2009 14% of Israeli population was
within the range of rockets. Figure 1 depicts the map of areas within the range of rockets
during this period.
[Table 1 about here.]
[Figure 1 about here.]
The rockets employed by Palestinian groups are not usually aimed at military targets, but
are fired in the general direction of cities and towns (Fisher, 2006). Further emphasizing their
randomized nature, Hamas explicitly chose a production strategy to increase the number
of rockets at the expense of their accuracy (IICC, 2009). Most of the rockets fired into
Israel have been of the locally-manufactured Qassam, with a range of approximately 10
kilometers (IICC, 2009). Longer range, Grad and Grad-2 rockets (20km and 40km range,
respectively) have been fired less frequently. These rockets are smuggled into Gaza through
5For an extensive review of the ballistic threat from Gaza see IICC (2009) and Rubin (2011). One
difference between mortars and rockets is that the former have a shorter range. In this paper, we focus
on the range of rockets since all localities that are within the range of rockets are also within the range of
mortars, but not the other way around.
8
tunnels underneath the border with Egypt, or via the sea. During the 2012 Gaza Conflict,
Hamas used its longer range Fajr-5 rockets to strike Tel Aviv and Jerusalem for the first time
(Bronner, 2012).
According to the Israel Security Agency’s (ISA) annual reports, there were 4,735 mortar
and 4,561 rocket attacks from 2001 through 2009.6The direct physical result of rocket fire has
been far less lethal than that of suicide attacks: from 2001 through 2009, rockets have resulted
in 32 fatalities, all of whom were civilians.7Almost half of all fatalities occurred in the town of
Sderot that has been targeted by rockets more than any other locality. Despite their relatively
low lethality, rockets have had a profound psychological impact on residents within the
range. For example, Berger, Gelkopf and Heineberg (2012) estimate that 43.5% of students
in 7th and 8th grade in Sderot, within range since 2001, are suffering from post-traumatic
stress disorder (PTSD) – a rate that is much higher than the rate prevalent in the general
population. In addition, Henrich and Golan (2013) find that children who reside within the
range are more likely to commit acts of violence. PTSD, anxiety, and depression have also
been diagnosed among adults within the range (Berger and Gelkopf, 2007; Diamond, Lipsitz,
Fajerman and Rozenblat, 2010). Aside from their psychological impact, rockets have also
imposed large economic costs on residents and communities. Municipalities within the range
are facing fiscal challenges (Yagna, 2010), and at times of a higher likelihood of attacks (for
example, following assassinations of Palestinian military activists), schools and kindergartens
remain closed, and this imposes additional costs on families. The economic costs of rockets
also affect the national government since it has to allocate resources to compensate those
whose property is damaged by the rockets. In some cases, compensation is paid even if there
is no physical damage, usually to make up businesses for losses due to the uncertain security
6A single attack can involve firing of multiple mortars or rockets that are launched together and target
the same location.
7In comparison, suicide attacks during this period have resulted in more than 550 fatalities. Rockets
are responsible for 18 deaths, and 14 fatalities occurred due to mortars (ISA, http://www.shabak.gov.il/
publications/decade/decade33/Pages/gaza1.aspx).
9
situation.8
Initially, the public and the political leaders perceived rockets as a minor nuisance, partly
because of the unprecedented wave of suicide attacks in the early 2000s that targeted major
Israeli cities. Conversely, the rockets initially only affected peripheral communities in the
vicinity of Gaza, resulting in relatively little damage and few casualties. After 2003, suicide
attacks had tapered off. At the same time, the range of rockets, their accuracy and lethality,
as well as the number of attacks have increased.9Rubin (2011) suggests that the shift in
public attitudes towards rockets began with the first rocket fatality in June 2004, and by
late 2005 rockets were perceived as an important challenge that needs to be addressed.10
In response to rocket fire, the government has initiated a public campaign to inform res-
idents within the range about the threat.11 Localities within the range have been connected
to an early warning radar system that activates a siren whenever it detects a rocket launch
from Gaza. The siren is heard not only in a locality where a rocket ultimately falls, but also
in a cluster of nearby localities. Moreover, the Israeli Defense Forces hold special drills in
localities within the range, and distribute flyers to inform residents about how to behave in
a case of a rocket attack. These actions suggest that residents within the range are aware of
the threat also if their locality is not directly hit.
8For example, the Yad Mordechai kibbutz received more than 500,000 shekels (about $150,000) following
the Cast Lead operation in 2008-2009 because it paid its agricultural workers even though they could not
work due to the threat of rockets. The kibbutz did not incur physical damage (Hadar, 2013). Since 2012,
the government pays up to 280 shekels per day (about $80) to parents of children under the age of 14 who
miss work because of school closures.
9The ISA reports the following data on the annual number of rocket attacks: 17 (2002), 123 (2003), 276
(2004), 286 (2005), 1247 (2006), 938 (2007), 1270 (2008), and 404 (2009). Based on newspaper reports and
private communication with Uzi Rubin, most of the rockets launched from Gaza land in empty, uninhabited
places.
10For example, in 2003 several regional councils appealed to the Supreme Court demanding to receive
compensation due to the threat of rockets. In 2004, the government allocated special budget to assist
localities within 7km of Gaza in dealing with the economic consequences of living under the threat of
rockets.
11There are also several smartphone apps, most notably Tzeva Adom (translates as “code
red”), which alert residents to impending rockets. See http://www.jpost.com/On-the-Web/
Appaholic-When-smartphones-save-lives.
10
Within the range, authorities also provide shelters for pedestrians and fortify public and
residential buildings (Rubin, 2011).12 This is a further indication that residents within the
range are aware that they can be hit by rockets, even if their locality has not been directly
attacked in the past. The Home Front Command posts on its website the updated map of
ranges, and locality-specific information, such as the amount of time that residents have to
find shelter once they hear sirens.13
A major component of Israel’s response to rockets has been the development and the
deployment of the Iron Dome missile defense system that is designed to intercept rockets
fired from distances of 4km - 40km. The development of this system has received a high
level of public attention, and very generous funding that indicate that the rockets threat is
perceived as important.14 In addition to the defensive measures, Israel has been involved in
high-risk offensive operations in Gaza which it justified by referring to the threat of rockets.
Data and Variables of Interest
Range of Rockets
Our main explanatory variable is Rangei,t that indicates whether a locality is within the
range of rockets one day before elections.15 Being in the range depends on two factors: the
12The government decided in February 2008 to fortify all buildings within 4.5km from Gaza. This range
was chosen because the Iron Dome anti-rocket system can only engage rockets that are fired from at least
4km, and thus does not protect localities that are 4km from Gaza or less (Rubin, 2011). As a result, buildings
in these localities have been reinforced to minimize damage in case they are directly hit by a rocket.
13The map refers to rocket threat not only from Gaza, but also from the northern border: http://www.
oref.org.il/1096-en/Pakar.aspx.
14The initial estimated cost of the system was $300 million, but the actual cost is much higher, which
makes the Iron Dome one of the most expensive weapons systems produced in Israel (Rubin, 2011).These
high costs have sparked a public debate about whether it is worthwhile to pay such a high cost to save a few
people at the expense of an investment elsewhere (Globes, 2011).
15Locality is a municipal unit recognized by the Ministry of Interior. Localities are distinguished by
their population size and by whether they are urban or rural. There are two types of urban localities:
municipalities and local councils. The former are larger than the latter in terms of their population. Rural
municipalities are grouped in regional councils that include several small villages and communities. Our unit
of analysis is locality (either municipality, local council, or regional council) in an election year.
11
maximum distance that a rocket can travel from Gaza, and locality’s distance from Gaza.
The information on rockets range comes from the Israel Defense Forces and Home Front
Command websites. To determine whether a locality is within the range of rockets, we
used ArcGIS to calculate the shortest distance between a locality and Gaza perimeter. A
locality is considered to be within the range if one day before election that distance is less
than or equal to the rockets’ range. Some localities, mostly regional councils, are composed
of discontinuous units. We consider such localities to be within the rockets’ range if the
minimum distance between all units composing a locality and Gaza perimeter is less than
or equal to the rockets’ range.16 In the robustness checks section, we examine alternative
measures of distance.
Voting
Voting data are available from the Israel Social Sciences Data Center (ISDC) at The Hebrew
University of Jerusalem17. We classify Israeli parties into left-wing, centrist, and right-
wing blocs using the classification proposed by Arian and Shamir (2008), and similar to
Schofield and Sened (2005) (see table 2). These blocs differ with respect to their positions
on the Israeli-Palestinian issue: right-wing parties are less willing to provide concessions to
Palestinians than centrist and left-wing parties.
[Table 2 about here.]
Our main dependent variable is RightSharei,t that measures the right-wing bloc’s vote-
share in locality iin election t(1999, 2003, 2006, and 2009).18 In addition, we examine
16For example, if regional council Xconsists of three discontinuous units, a,b, and c, we measure the
shortest distance between Gaza perimeter and the perimeter of each of these units. Xis coded as within the
range if all the three distances are less than or equal to the rockets range one day before the election. If at
least one of these distances is greater than the rockets’ range, Xis coded as beyond the range, even if some
of its units are within the range.
17http://isdc.huji.ac.il/mainpage_e.html, accessed March 25, 2013.
18Israel is a parliamentary system with proportional representation. Despite this, in 1999 there was a
direct vote for the prime minister, in addition to the traditional vote for a party. Voting in 2003, 2006, and
12
vote-shares of nationalistic, religious ultra-orthodox parties, Russian immigrants’ parties, as
well as vote-shares of Likud, Labor, and Shas parties. We calculate vote-shares by dividing
the number of votes a bloc, a subset of a bloc, or a party received by the number of valid
votes in a locality in a given election. Vote-share variables are continuous, and range from 0
to 1.
We focus on vote-shares as our dependent variable for two reasons. First, they reflect
voting behavior, and are therefore directly relevant to understanding how terrorism threat
affects voters. Second, Israel has a highly proportional representation system, with vote-
shares closely corresponding to seat-shares in the parliament. Thus, they not only represent
changes in voting behavior, but also have direct implications for government composition
and policies.
Control Variables
We control for a broad set of variables that can affect voting. First, we include locality-
level time-varying demographic attributes available from the Central Bureau of Statistics:
population size (logged), median age, ratio of males to females, and the share of Jewish
population in a locality. We use a one year lagged value of these variables. In addition,
we control for the share of Jewish population with family origin in Asia, Africa, and in the
former Soviet Union because previous research shows that these groups of voters are more
likely to support right-wing parties (Berrebi and Klor, 2008). Locality-level data on residents
with origin in Asia and Africa are available from the 1995 and 2008 censuses, whereas similar
data on the share of residents with origin in the former Soviet Union was recorded only in
the 1995 census. We therefore use the 1995 values of these three variables in specifications
without locality fixed effects.19
2009 elections was for parties only.
19Using the 1995 data on residents with origin in Asia and Africa for 1999-2006 election and 2008 data for
2009 election does not change the effect of being in the range on voting.
13
To control for local economy, we include standardized locality mean wage adjusted for
inflation using 2006 as the base year.20 Education can also affect voting, and we control
for the share of high school graduates among residents who are 17-25 years old in each
locality-year.21
Migration could pose a problem if population composition within the range changes over
time. We control for population movement in and out of localities by including the net
migration divided by locality’s population.22
Previous studies demonstrate that other forms of terrorism, and in particular suicide
attacks, affect voting (Berrebi and Klor, 2008; Gould and Klor, 2010). We, therefore, control
for the logged number of local fatalities due to suicide attacks three months prior to an
election. We obtain information on local fatalities by combining data from Benmelech,
Berrebi and Klor (2010b) and Getmansky (2011).
In addition to the broad set of locality-level time-varying controls, we include locality fixed
effects that account for observable and unobservable time-invariant locality characteristics,
such as distance to border, distance to Gaza, or to the West Bank, distance to the center,
topography, and whether a locality is a regional capital. In specifications without fixed
effects, we explicitly control for some of these attributes.
Finally we include year dummies to control for secular trends that affect all localities in
a given year, such as the national economy, quality of electoral campaigns, leaders’ char-
acteristics, and events related to the Israeli-Palestinian conflict that influence Israel as a
whole.
20Due to limitations of locality-level time-varying data, we use the 1995 value of locality mean wage for
1999 elections, 2001 value for 2003 elections, 2003 value for 2006 elections, and 2006 value for 2009 elections.
We take the 2006 mean wage in locality ias the base line for that locality, and adjust the other three values
using the average consumer price index: 100
mean consumer price index ×mean wagei, when the mean consumer
price indices for 1995, 2001, 2003, and 2006 are 66%, 91%, 97%, and 100% respectively.
21High school graduation rates are only reported for 17-25 age group.
22Migration can be a potential problem if left-wing residents move out of the range and right-wing residents
move in, and we perform additional tests in the robustness checks section to rule out this possibility.
14
Empirical Strategy
Econometric Model
Our goal in this paper is to identify the causal effect of being in the range of rockets on
voting. We estimate the following linear model:
RightSharei,t =α×Rangei,t +β×(Xi,t−1)+γt+µi+εi,t,(1)
where RightSharei,t is the vote-share of right-wing parties in locality iin election t,Rangei,t
is a binary indicator of whether locality iis within rockets range one day before election t,
and Xi,t−1is a vector of lagged time-varying locality-level controls, as explained above. γt
is election year dummy that captures secular trends, µiis locality fixed effects, and εi,t is
locality-clustered robust error that captures unobserved determinants of voting in year t.23
The coefficient αidentifies the effect of being in the range on right-wing parties’ vote-
share. We expect αto be positive and statistically significant if localities within the range
have a higher right-wing vote-share than localities outside the range, conditional on Xi,t−1, µi,
and γt. A negative and statistically significant αwould imply that localities within the range
of rockets are less supportive of right-wing bloc, and more supporting of centrist or left-wing
parties. An αcoefficient that is statistically not significant would imply that being in the
range of rockets does not affect voting.
Is Rockets Range Exogenous to Voting?
We use the variation over time in the range of rockets to explore the effect of being in the range
on voting. Given that time invariant locality-level characteristics, such as distance to the
23We cluster the standard errors on locality because according to Wooldridge (2010), "when applying the
FE estimator, it is important to remember that nothing rules out serial correlation (page 310)." That said,
our results are robust to not clustering (these results are available in the online Appendix, table B.6).
15
Gaza Strip, remain constant, and controlling for time-varying locality attributes, we study
how voting is affected by being in the range of rockets. Our key identification assumption
is that the range of rockets is exogenous to voting. Essentially, we assume that localities
come into the range of rockets for reasons that are unrelated to their anticipated voting
behavior. To investigate this assumption, we examine whether observable variables that are
known to be correlated with voting also determine whether certain localities are within the
range of rockets. If localities within the range are similar to those out of the range, then any
differences in voting can be attributed to living in the range of rockets. Assessing similarity
between within- and out-of-the range localities is also important to minimize extrapolation
bias (King and Zeng, 2006).
Table 3 presents the summary statistics of the various variables we use to compare local-
ities within and outside the range.24
[Table 3 about here.]
Overall, the distributions of these variables is very similar across within- and out-of-the
range localities. For many variables (gender ratio, share of immigrants from the former
Soviet Union, the mean wage, presence of a regional capital, and turnout in 1999) localities
within the range have similar localities that are out of the range. Some variables, however,
have non-overlapping areas. Localities within the range are less populous than those outside
the range, and their median age is higher, implying fewer children than in localities outside
the range. Localities within the range have higher shares of residents with family origin
in Asia and in Africa, and a lower share of high school graduates. Furthermore, localities
within the range have higher right-wing vote-shares in 1999 election and they are on average
closer to the border than localities outside the range, but not by much.
It is important to note that even those variable that do not completely overlap have a
broad range of common support, implying that for many localities within the range there are
24Kernel density plots for each variable are visible in Appendix, figures A.1-A.5.
16
similar localities outside the range. Notwithstanding, in the robustness checks section, we
drop observations with non-overlapping values (outside of common support) to make sure
our results are not driven by outliers.
Results
Main Results
Table 4 presents the effect of being in the range on right-wing parties’ vote-shares. Column
1 reports the results of an OLS model with locality fixed effects. The estimated effect is a
4 percentage points increase in the right-wing vote-share within the range (95% confidence
interval is 3-5 percentage points increase). In column 2, we add time-varying controls, and
the estimated effect of being in the range of rockets is now a 6 percentage points increase
in the right-wing vote share (95% confidence interval now is 5-9 percentage points increase).
We add year dummies in column 3, which is our main specification as in equation (1). The
estimated effect of being in the range drops, but it is still positive and significant: being
in the range adds 2 percentage points to the right-wing vote share (95% confidence interval
is 0.4-3 percentage points increase). The fourth column presents the results of a lagged
dependent variable model without locality fixed effects and year dummies.25 In addition to
the time-varying control variables, in this model we also control for Jewish voters’ origin,
whether a locality is a regional capital, and distance to the closest border.26 The estimated
effect is an increase of 4 percentage points in right-wing vote-share in localities within the
range (95% confidence interval is 2-6 percentage points increase).
25The fixed effects model model and the lagged dependent variable model are alternative to each other.
Angrist and Pischke (2008) recommend reporting the results of both models because they can be seen as
bounding the causal effect of interest.
26In an online Appendix, we also report results of estimation using locality distance to the Gaza Strip
instead of to the closest border. The results remain substantively the same.
17
[Table 4 about here.]
Based on several alternative model specifications, our main results suggest that being
in the range of rockets increases locality right-wing vote-share by 2 to 6 percentage points
(95% confidence interval is 0.4-9 percentage points increase). Israel is a proportional repre-
sentation system with a single district. The electoral threshold in 1999 and 2003 elections
was 1.5% of the popular vote, and it was raised to 2% in 2006 and 2009 elections. The
Israeli parliament (Knesset) has 120 seat that are allocated to parties that pass the electoral
threshold proportionately to each party’s vote-share. Accordingly, a seat in the parliament
requires 0.8% of votes.27. Thus, an increase of 2 to 6 percentage points in the right-wing
parties’ vote-shares implies additional 2 to 7 seats for the right-wing bloc, if all voters were
within the rockets’ range.
Table 5 presents the effect of range on components of the right-wing bloc. Nationalistic
parties add 3 percentage points to their vote-share in localities that fall within the range
of rockets (95% confidence interval is 0.3-5 percentage points). Religious parties gain 1
percentage point, but this result is not statistically significant. Parties of immigrants from
the former Soviet Union lose 1 percentage point, but this result also falls short of statistical
significance.
[Table 5 about here.]
Looking at the effect of being in the range on individual parties, we find that Likud (the
largest right-wing party) gets additional 2 percentage points in localities that are within the
rockets’ range (95% confidence interval is 1-3 percentage points). This is similar to the effect
of being in the range on nationalistic and right-wing parties. Labor (the main left-wing
party) and Shas (the main religious party) experience a minor increase and decrease in their
27This means that the minimum number of seats a party can get is 2.
18
vote-shares in localities within the range, respectively, but these effects are not statistically
significant.28
Does the Effect of Rockets Depend on the Party in Power?
There are several possible explanations why localities within the range have higher right-wing
and nationalistic vote-shares. First, following the retrospective voting theory, support for
these parties might increase because voters punish non-right-wing parties for exposing them
to rockets. Indeed, the incumbent party in 2006 and 2009 elections was centrist (Kadima),
and this might explain the increase in right-wing vote-share.
Second, right-wing vote-share might also increase because these parties enjoy a valence
advantage on the security dimension. If the valence theory is correct, expansion of the
rockets’ range would increase the salience of security issues, and benefit right-wing parties
that are perceived to be more competent in matters of security.
To explore the conditional effect of rockets’ range on right-wing vote-share, we include a
dummy variable Rig htP Mtthat is equal to 1 if the incumbent is from the right-wing bloc,
and 0 if otherwise. We also include an interaction term: Rangei,t ×RightP Mtthat is equal
to 1 if locality iis within the rockets’ range one day before election tand the incumbent is
from the right-wing bloc, and it is equal to 0 otherwise. We estimate the following model:
RightSharei,t =αRangei,t +δ1RightP Mt+δ2Rangei,t ×Rig htP Mt(2)
+β(Xi,t−1)+γt+µi+εi,t,
The effect of being in the range under a right-wing incumbent is equal to α+δ2.29
28Due to space limitations the results of these tests are in an online Appendix, table A.1.
29This effect is equal to αunder a non-right-wing incumbent because the interaction term is then equal
to 0 (Brambor, Clark and Golder, 2006).
19
Column 1 in table 6 reports the effect of rockets’ range on right-wing vote-share. The
coefficients of the constitutive terms (Rangei,t,RightP Mt) and the interaction term are
statistically significant at 95% confidence level. Similarly to previous findings, right-wing
vote-share in localities within the range is 2 percentage points higher than beyond the range
when the incumbent is non-right-wing (95% confidence interval is 0.6, 3 percentage points).
The interaction term, however is negative: adding the two coefficients suggests that right-
wing parties lose 0.5 percentage points in localities within the range when the incumbent is
from the right-wing bloc. This result, however, is not statistically significant, implying that
when there is a right-wing incumbent, there is no difference in right-wing vote-share within
and out of the range.30 Overall, the marginal effect of being in the range, conditioning on the
incumbent’s affiliation, is a 1 percentage point increase in right-wing vote-share (this result
is significant only at 90% confidence level). In light of this, we conclude that right-wing
parties gain additional votes within the rockets’ range only when the incumbent is from a
non-right-wing bloc. When the incumbent is right-wing, these parties also do not achieve
substantial gains, as they do under non-right-wing incumbents. However, they are also not
punished by voters.
[Table 6 about here.]
Column 2 presents the effect of rockets’ range on Likud vote-shares, conditional on the
incumbent’s affiliation.31 Under a non-right-wing incumbent, there is a 2 percentage points
increase in Likud vote-share within the range (95% confidence interval is 0.5, 3 percentage
points). The overall marginal effect of Rangei,t on Likud vote-share, conditioning on the
incumbent, is a 2 percentage points increase (95% confidence interval is 0.8, 3.7 percent-
age points). Thus, Likud experiences a similar increase in its vote-share within the range,
30We use the margins command in Stata to make these calculations.
31Likud, as shown in table 2, is the main right-wing party, and the head of Likud serves as a prime minister
in right-wing governments.
20
regardless of who the incumbent is. Unlike the entire right-wing bloc, Likud’s vote-shares
increase also under a right-wing (Likud) incumbent. Even though the interaction term is not
significant, the overall effect (α+δ2) is positive and similar to the effect when the incumbent
is from a non-right-wing bloc. It is important to note that this model includes year dummies,
and therefore controls for secular trends that affect all Israeli localities, thereby isolating the
local effect of being in the range on voting.
This result is consistent with the assertion that Likud, as a right-wing party, has valence
when it comes to security problems. However, this might also be consistent with the ‘rally
round the flag’ effect, and a possible tendency of voters to support the incumbent party,
especially when confronted with security challenges. Therefore, in column 3 we also examine
the effect of rockets’ range on Kadima party’s vote-shares when Kadima was the incumbent.
Kadima was founded as a centrist party before 2006 election, and ran in 2006 and 2009
elections. The incumbent in these elections was also from Kadima.32 Since Kadima’s vote-
shares are available only for elections in which Kadima is an incumbent, there is no need
to include RightP Mtand the interaction term (they drop out because there is no variation
in RightP Mtduring these years). Results in column 3 suggest that expansion of rockets’
range does not affect Kadima’s vote-shares: The coefficient is negative, but substantively
small, and statistically not significant at acceptable significance levels. Thus, unlike Likud,
Kadima did not gain votes within the range when it ran as an incumbent.
These results lead us to the following conclusions: First, there is no evidence that in-
cumbents are punished within the rockets’ range: right-wing parties did not lose votes in
2003 election, and Kadima did not lose votes in 2006 and 2009 elections. Second, right-wing
parties gain votes within the range when the incumbent is non-right-wing. Moreover, Likud
gains additional votes even when they are run as an incumbent. This cannot be interpreted
32Ariel Sharon was the prime minister prior to 2006 election. He had left Likud, and formed Kadima in
2005, and ran in 2006 as the head of Kadima party. Ehud Olmert from Kadima was the prime minister
before 2009 election. Thus, we code Kadima as the incumbent before 2006 and 2009 election.
21
as ‘rally round the flag’ since Kadima’s vote-shares within the range do not increase when
Kadima is in power. Instead, we view these findings as partially consistent with the valence
argument: whereas Likud enjoys valence when it comes to security problems even when these
problems emerge during Likud’s time in office, right-wing bloc as a whole gains votes only
under a non-right-wing incumbent.
Robustness Checks
Using table 3 we identify areas of non-overlap between localities within and beyond the
rockets’ range. We reestimate our baseline specification (column 3 in table 4), and drop those
observations that are outside of common support. This reduces the number of localities to
132.
The results in table 7 are substantively similar to our previous findings. Column 1
reports the effect of being in the range on right-wing vote-share, controlling for time-varying
locality-level characteristics, as well as locality fixed-effects and year dummies. The result is
positive and statistically significant, and suggests that localities within the range experience
a 3 percentage points increase in the right-wing vote share. Column 2 similarly estimates the
effect of being in the range on nationalistic parties’ vote-share, and finds that it increases by
5 percentage points. These tests show that our results regarding the effect of being in the
range of rockets on voting do not depend on incomparable samples: in fact, when we omit
localities that are outside common support, the point estimate becomes larger.
[Table 7 about here.]
Since we study the effect of being in the rockets’ range, it is important to make sure that
our findings are robust to different methods of measuring distance from Gaza. Column 3 in
table 7 shows the results using the distance between Gaza perimeter and locality’s center
(rather than locality’s perimeter as before). In column 4, we subtract 5km from the range to
22
account for the possibility that rockets are not launched from Gaza perimeter, but from areas
within Gaza.33 The results in columns 3 and 4 in table 7 are very similar to our previous
results (compare to column 3 in table 4), suggesting that our findings do not depend on a
specific method of measuring distance or on assuming that rockets are launched from the
closest point within Gaza to Israel.
Are the Results Driven by the Intensity of Targeting?
One of the contributions of this paper is to show that voting behavior could change due
to the threat of being hit, and not necessarily following actual attacks. As we show in
the previous sections, rocket fire has been far less lethal and far less successful at hitting
inhabited places than suicide attacks. Nonetheless we detect a substantively-significant effect
on voting. Given that localities within the range are not substantially different from localities
outside the range, we believe that this effect is driven by the threat of future attacks rather
than by the intensity of past exposure to rockets.
Unfortunately, locality-level data on rocket hits are not publicly available due to the
concern that they could be used to improve rockets’ accuracy. Nonetheless, we can use our
data to demonstrate that this effect is not driven by direct targeting. For this purpose, we
drop the town of Sderot from our sample and repeat our tests. As mentioned above, Sderot
has been disproportionately targeted by rockets, and has absorbed half of the fatalities caused
by rocket attacks. We show that our results do not change if we exclude Sderot, and thus
are not driven by this exceptional case. The results of this test are available in the online
Appendix (see tables B.1). The results of this test confirm our main findings. What matters
to voters is not how many times they have been targeted, but whether they are in the range
33The Gaza Strip is between 6km and 12km wide, and therefore 5km constitutes the midpoints of its
widest part. In the online Appendix, in tables A.2 and A.3 we use two additional alternatives measures of
distance, and show that our results are robust to these measures as well.
23
or not in the range– which people are made objectively aware of by government officials.34
Is Turnout Affected by Rockets?
One potential explanation for the increase in the right-wing vote-share is that the composition
of the electorate changes. For example, it is possible that left-leaning voters become less likely
to vote when threatened by rockets, or right-leaning voters become more likely to vote. If
this is true, then the changes in right-wing vote-shares are not driven by changes in voters’
positions, but by changes in the composition of the electorate. To investigate this possibility,
in table 8, column 1 we report the effect of being in the rockets’ range on turnout, using
locality-level time-varying controls and fixed effects, as well as year dummies. These results
suggest that the effect of rockets range on turnout is substantively marginal and statistically
not significant. Localities that come into the range do not differ with respect to turnout
from localities that are outside the range.
This, however, does not rule out the possibility that coming into the range simultane-
ously affects both the right-leaning and the left-leaning voters, such that the overall turnout
remains the same, even though the composition of the electorate changes. To address this
possibility, we reestimate our main specification using a subset of localities with turnout
greater than 69% (the median turnout). In these localities, the majority of eligible voters
vote. Therefore, any difference in right-wing vote-shares must be due to voters changing
their positions, and not changes in the composition of the electorate (i.e. turnout effect).
Column 2 in table 8 is similar to the results we report in other specifications: being in the
range of rockets increases the right-wing vote-share by 2 percentage points.
[Table 8 about here.]
34The literature on probability judgement provides a possible explanation of why the effect of being in range
is what matters, and not how many times a locality has been targeted. Actual targeting and threat perception
are not linearly related (Fischhoff, 2011). Base-rate neglect, availability bias, the difficulty of distinguishing
small risks based on small samples, wishful thinking, or prospect theory’s probability weighting function can
make everyone within the range feel roughly equally threatened.
24
Are the Results Affected by Migration?
Another potential problem could be that the composition of residents changes when a local-
ity enters the rockets range. Left-leaning residents might choose to leave and move out of
the range, while right-leaning residents might choose to move into these localities to demon-
strate their patriotism. In our estimations so far, we control for the net migration (incoming
residents - outgoing residents). This, however, does not address the possibility that right-
leaning and left-leaning residents might be moving simultaneously in and out of localities,
respectively, such that the overall migration net is low despite changes in population com-
position.
To address this possibility, we use data on incoming and outgoing residents in each
locality-year.35 First, we reestimate our main specification using localities in which the
number of incoming and outgoing residents is below the median in each category (lowest
50%).36 Very few people move in and out of these localities, and any changes in vote-shares
are therefore likely to be not due to migration, but due to changes in preferences. Column 3
in table 8 presents the results of our main specification (comparable to column 3 of table 4).
These results suggest that among localities that experience low migration, those within the
range have 3 percentage points higher right-wing vote-share than those out of the range.
Second, we create a new variable that conservatively estimates the number of right-wing
vote-share in each locality, assuming that all incoming residents are right-leaning, and all
35This information is not available for regional councils, and for all localities prior to 2002. However, we
believe this data unavailability does not pose a serious limitation to our test. This is because migration
problem is more likely to affect urban communities than rural ones (regional councils). Furthermore, the
rockets problem started in April 2001, and as discussed in previous sections, initially was not perceived as
a serious threat. Thus limiting the test to 2002 and onwards should still be useful to detect the effect of
migration.
36The median number of incoming residents is 291, and the median number of outgoing resident are 256,
and we use all locality-years that are below these two thresholds. Focusing on localities with low migration
reduces the number of observations, especially within the range, from 71 to 23 locality-years. The lowest
thresholds that allow us to estimate the model are localities with fewer than 145 incoming residents, and
fewer than 130 outgoing residents (the lowest 35%). The results are substantively similar to the ones we
present here.
25
outgoing residents are left-leaning.37 This is a very strong assumption that allows us to
estimate the lowest possible right-wing vote-share in a locality given that migration could
pose a problem. We then reestimate our main specification using this conservative estimate
as the dependent variable. Column 2 in table 8 shows that the effect of range on this
conservative estimate of right-wing vote-share is still positive and statistically significant:
localities within the range experience a 3 percentage point increase in the right-wing vote-
share compared to localities that remain beyond the range.38
Summary and Implications
Summary
In this paper, we are asking how the threat of being targeted by terrorists affects voting
behavior. We answer this question by focusing on the effect of being in the range of rockets
fired from the Gaza Strip on voting in 2003-2009 Israeli parliamentary elections. We provide
evidence that localities in the range of rockets are very similar to localities that are not in
the range, and therefore the difference in voting is due to being in the range.
Our findings suggest that entering the rockets’ range substantially affects voting and
increases support for the right-wing bloc, and especially for nationalistic parties. We find that
right-wing vote share is 2 to 6 percentage points higher in localities that are within the range.
We attribute this finding to the valence advantage that right-wing parties in Israel have with
respect to security threats. Consistent with the valence explanation, we find that right-wing
parties are not punished by voters, and in some cases even gain additional votes in localities
that become part of the range during right-wing parties’ term in office. We demonstrate that
37First, we deduct the number of incoming residents in year t−1from the number of right-wing votes in
year t. Second, we add the number of outgoing residents in year t−1to the total number of voters who cast
a ballot in election year t. We then divide the former by the latter: rightwing v otesi,t−incoming residentsi,t−1
total votesi,t+outg oing residentsi,t−1
38The smaller number of observations is because data on incoming and outgoing residents are not available
for regional councils.
26
these results are driven by being in the range, and do not depend on actual targeting, and
on changes in the composition of the population and the electorate. Furthermore, we show
that our results are robust to various model specifications, and different methods of defining
and measuring the range.
Implications
Our finding that the threat of terrorism leads voters to support right-wing parties provides a
plausible mechanism for why some conflicts endure: voters who are threatened with violence
elect candidates who are less willing to make concessions. This further reduces the bargain-
ing range between conflict participants, and decreases the chances of reaching a negotiated
solution. Our results also provide empirical support for theories that posit that hard-line
groups may use violence and the reaction it engenders to foment conflict, and discredit more
dovish actors (Bueno de Mesquita and Dickson, 2007; Kydd and Walter, 2002).39
Our project is also related to recent studies on the efficacy of terrorism from the perspec-
tive of the terrorists (Abrahms, 2006, 2012). If the terrorists’ goal is to scare a population into
accepting the terrorists’ demands, our paper suggests that this might be counter-productive.
However, if the goal is to insure conflict between the two sides, and marginalize doves on both
sides, then it may be effective.40 By using the randomized threat of rockets from Gaza, our
study avoids issues of endogeneity – non random selection of targets and strategic interac-
tion between terrorists and targeted government – that limit the interpretations of previous
empirical studies.
In addition to finding that voters within the range harden their attitudes towards the
perpetrating group, we also detect an interesting incumbency effect. Rather than punishing
39For instance, many have argued that Hamas and other militant groups in Gaza have strategically used
rocket attacks to marginalize Fatah, and spoil any negotiations between Fatah and Israel. See http://www.
nytimes.com/2009/01/15/world/middleeast/15fatah.html?pagewanted=all&_r=0
40The question of which of these goals (or both) dominates the logic of terrorist violence, or the Israeli-
Palestinian conflict is beyond the scope of the present paper.
27
incumbents who hold office during increases in the threat of rockets, our findings show that
voters turn to the right-wing Likud party regardless of incumbency (and they do not punish
right-wing parties even if the range expanded during their term in office). This finding
parallels recent work that suggests voters choose parties who they feel “own” issues (such
as the case of right-wing parties and security in Israel) (Wright, 2012). This has serious
implications for bargaining and negotiations – violence leads to hard-line parties coming to
power, making settlement less likely and future violence more likely.
We argue, perhaps counterintuitively, that the nature of the threat from rockets’ range
provides a generalizable mechanism for how violence influences political behavior in other
conflicts. Previous studies find that exposure to violence affects voting and political pref-
erences (Berrebi and Klor, 2008; Gould and Klor, 2010), but that this effect dissipates if
violence increases beyond a certain threshold, as the targeted population accepts high levels
of violence as part of life (Jaeger, Klor, Miaari and Paserman, 2012). If violence has dimin-
ishing returns in its impact on political preferences, then most of the substantial effects will
manifest with violence that does not inflict large number of casualties, like in the case of
suicide attacks. Despite the number of rockets that have been launched into southern Israel,
there have been relatively few casualties. Furthermore, the localities that are within the
range in 2001-2009 have not been major targets of suicide bombings or other forms of ter-
rorist attacks. We argue that the focus on threat and low-levels of lethality is more relevant
to understanding the broader impact of terrorism (e.g. in the US and Europe) – where most
people are threatened by, rather than directly exposed to the violence. This focus on the
effect of rockets’ range differentiates our studies from purely retrospective studies of violence
(Bellows and Miguel, 2009; Kibris, 2011), as voters are weighing their past exposure and fear
from future exposure (Huddy, Feldman, Taber and Lahav, 2005). Given the stated goal of
terrorism to instill fear in a target populace (Abramson, Aldrich, Rickershauser and Rohde,
2007), this is a cleaner measure of the effect of terrorism.
28
Our research also has serious policy implications. Previous research on terrorism and
political behavior has mostly focused on suicide terrorism. This research is unable to adjudi-
cate on whether the main effect of terrorism on political behavior operates through casualty
aversion or through living under threat. Using data on the range of rockets from Gaza, we
show that merely being in range of terrorism, and not necessarily experiencing a high number
of casualties, is sufficient to have a strong influence on voting. This implies that governments
face a significant trade-off. Do they take every effort to warn residents and emphasize the
wider-level of threat, thereby fanning the effects of terrorism, or do they seek to reduce
perceptions about threat and provide mechanisms for coping? This question may depend
on the political affiliation of the leaders, with right-wing politicians showing a preference to
increase the warnings over terrorism and heighten the threat level, rather than dampen it
down (and left-wing politicians the opposite).41
Finally, our findings also add nuance to other studies of indiscriminate violence, such
as Lyall (2009) who finds that randomized artillery shelling in Chechnya reduces insurgent
attacks through intimidating the targeted population. Conversely, we find that randomized
threat of violence increases hardline attitudes. Taken together, these findings suggest that
the effect of violence (indiscriminate or otherwise) may be context dependent. We conjecture
that the targeted actors’ military strength moderates this relationship. Israel is a stronger
military actor compared to Hamas. Conversely, the Chechen rebels were weaker compared to
the Russian army. Exposure to violence thus may harden the populace of a targeted stronger
military actor, leading them to be more willing to fight. Yet, it may lead the populace of
the weaker actor to favor withdrawal, and concessions to the perpetrating actor. Given the
role of civilian victimization in several recent conflicts–Syrian Civil War, US drone strikes in
Pakistan, and post-Mubarak violence in Egypt–this hypothesis deserves further scrutiny.
41See Willer (2004) for evidence that government-issued terror warnings affect political support for the
incumbent.
29
References
Abrahms, Max. 2006. “Why terrorism does not work.” International Security 31(2):42–78.
Abrahms, Max. 2012. “The political effectiveness of terrorism revisited.” Comparative Polit-
ical Studies 45(3):366–393.
Abramson, Paul R., John H. Aldrich, Jill Rickershauser and David W. Rohde. 2007. “Fear
in the voting booth: The 2004 presidential election.” Political Behavior 29(2):197–220.
Angrist, Joshua D. and Jörn-Steffen Pischke. 2008. Mostly Harmless Econometrics. 1st ed.
Princeton University Press.
Arian, Asher and Michal Shamir. 2008. “A Decade Later, the World Had Changed, the
Cleavage Structure Remained Israel.” Party Politics 14(6):685–705.
Baker, William D. and John R. Oneal. 2001. “Patriotism or opinion leadership? The nature
and origins of the rally -’round-the-flag effect.” Journal of Conflict Resolution 45(5):661–
687.
Bali, Valentina A. 2007. “Terror and elections: Lessons from Spain.” Electoral Studies
26(3):669–687.
Becker, Gary S. and Yona Rubinstein. 2011. “Fear and the Response to Terrorism: An
Economic Analysis.” Centre for Economic Performance, London School of Economics,
Discussion Paper 1079.
Bellows, John and Edward Miguel. 2009. “War and local collective action in Sierra Leone.”
Journal of Public Economics 93(11):1144–1157.
Benmelech, Efraim, Claude Berrebi and Esteban F. Klor. 2010a. “Counter-Suicide Terror-
30
ism: Evidence from House Demolitions.” National Bureau of Economic Research (NBER)
Working Paper No. 16493.
Benmelech, Efraim, Claude Berrebi and Esteban F. Klor. 2010b. “The Economic Cost of
Harboring Terrorism.” Journal of Conflict Resolution 54(2):331–353.
Berger, Rony and Marc Gelkopf. 2007. The Impact of the Ongoing Traumatic Stress Con-
ditions on Sderot. Israel Trauma Centre for Victims of Terror and War (NATAL).
Berger, Rony, Marc Gelkopf and Yotam Heineberg. 2012. “A Teacher-Delivered Intervention
for Adolescents Exposed to Ongoing and Intense Traumatic War-Related Stress: A quasi-
Randomized Controlled Study.” The Journal of Adolescent Health 51(5):453–461.
Berrebi, Claude and Esteban F. Klor. 2008. “Are Voters Sensitive to Terrorism? Direct
Evidence from the Israeli Electorate.” American Political Science Review 102(3):279–301.
Bonanno, George A and John T Jost. 2006. “Conservative shift among high-exposure sur-
vivors of the September 11th terrorist attacks.” Basic and Applied Social Psychology
28(4):311–323.
Brambor, Thomas, William Clark and Matt Golder. 2006. “Understanding Interaction Mod-
els: Improving Empirical Analyses.” Political Analysis 14(1):63–82.
Bronner, Ethan. 2012. “With Longer Reach, Rockets Bolster Hamas Arsenal.” The New York
Times .
Bueno de Mesquita, Ethan. 2007. “Politics and the Suboptimal Provision of Counterterror.”
International Organization 61(1):9–36.
Bueno de Mesquita, Ethan and Eric S. Dickson. 2007. “The Propaganda of the Deed:
Terrorism, Counterterrorism, and Mobilization.” American Journal of Political Science
51(2):364–381.
31
Canetti-Nisim, Daphna, Eran Halperin, Keren Sharvit and Stevan E. Hobfoll. 2009. “A New
Stress-Based Model of Political Extremism Personal Exposure to Terrorism, Psychological
Distress, and Exclusionist Political Attitudes.” Journal of Conflict Resolution 53(3):363–
389.
Diamond, Gary M., Joshua D. Lipsitz, Zvi Fajerman and Omit Rozenblat. 2010. “Ongoing
Traumatic Stress Response (OTSR) in Sderot, Israel.” Professional Psychology: Research
and Practice 41(1):19–25.
Fiorina, Morris P. 1978. “Economic Retrospective Voting in American National Elections:
A Micro-Analysis.” American Journal of Political Science 22:426–443.
Fischhoff, Baruch. 2011. Risk Perception and Communication. In Risk: A Very Short In-
troduction, ed. Baruch Fischhoff and John Kadvany. New York, NY: Oxford University
Press pp. 940–954.
Fisher, Ian. 2006. “How Clumsy, Inaccurate Gaza Rockets Could Start a War.” The New
York Times .
Gassebner, Martin, Richard Jong-A-Pin and Jochen O. Mierau. 2008. “Terrorism and elec-
toral accountability: One strike, you’re out!” Economics Letters 100(1):126–129.
Gelpi, Christopher, Peter D. Feaver and Jason Reifler. 2006. “Success Matters: Casualty
Sensitivity and the War in Iraq.” International Security 30(3):7–46.
Getmansky, Anna. 2011. “Whose Life Is Worth More? Domestic Politics, Counterterrorism,
and Terrorists’ Selection of Targets.” Working Paper, New York University .
Globes. 2011. “Iron Dome Costs Billions, a Dead Person Costs 1.2 millions [Kipat Barzel
Olah Miliardim; Adam Met Oleh 1.2 Milion].” Globes .
URL: http://www.globes.co.il/news/article.aspx?did=1000637352
32
Gould, Eric D and Esteban F Klor. 2010. “Does Terrorism Work?” The Quarterly Journal
of Economics 125(4):1459–1510.
Hadar, Kaneh. 2013. “Reshut haMisim Tishalem Keh Hetzi Milion Shekel Leh Kibutz Yad
Mordechai [The Tax Authority To Pay Over Half Million Shekels to the Yad Mordechai
Kibbutz].” Calcalist, July 15.
URL: http: // www. calcalist. co. il/ local/ articles/ 1,7340,L-3607609,00.
html
Haushofer, Johannes, Anat Biletzki and Nancy Kanwisher. 2010. “Both sides retaliate
in the Israeli–Palestinian conflict.” Proceedings of the National Academy of Sciences
107(42):17927–17932.
Henrich, Christopher and Shahar Golan. 2013. “Effects of Exposure to Rocket Attacks on
Adolescent Distress and Violence: A 4-Year Longitudinal Study.” Journal of the American
Academy of Child and Adolescent Psychiatry 5(6):619–627.
Huddy, Leonie, Stanley Feldman, Charles Taber and Gallya Lahav. 2005. “Threat, anxiety,
and support of antiterrorism policies.” American Journal of Political Science 49(3):593–
608.
Huddy, Leonie, Stanley Feldman and Erin Cassese. 2007. On the Distinct Political Effects of
Anxiety and Anger. In The Political Dynamics of Feeling and Thinking, ed. Ann Crigler,
Michael MacKuen, George E. Marcus and W. Russell Neuman. University of Chicago
Press pp. 202–230.
IICC. 2009. Summary of Rocket Fire and Mortar Shelling in 2008. Gelilot, Israel: Israel Intel-
ligence Heritage and Commemoration Center. Available at http://www.terrorism-info.
org.il/data/pdf/PDF_19045_2.pdf.
33
Jaeger, David A. and Daniele M. Paserman. 2008. “The Cycle of Violence? An Empirical
Analysis of Fatalities in the Palestinian-Israeli Conflict.” The American Economic Review
98(4):1591–1604.
Jaeger, David A., Esteban F. Klor, Sami H. Miaari and M. Daniele Paserman. 2012. “The
Struggle for Palestinian Hearts and Minds: Violence and Public Opinion in the Second
Intifada.” Journal of Public Economics 96(3):54–368.
Karol, David and Edward Miguel. 2007. “The electoral cost of war: Iraq casualties and the
2004 US presidential election.” Journal of Politics 69(3):633–648.
Kibris, Arzu. 2011. “Funerals and Elections: The Effects of Terrorism on Voting Behavior
in Turkey.” Journal of Conflict Resolution 55(2):220–247.
Kiewiet, D. Roderick and Douglas Rivers. 1984. “A retrospective on retrospective voting.”
Political Behavior 6(4):369–393.
King, Gary and Langche Zeng. 2006. “The Dangers of Extreme Counterfactuals.” Political
Analysis 14(2):131–159.
Koch, Michael and Benjamin Tkach. 2012. “Deterring or Mobilizing? The Influence of
Government Partisanship and Force on the Frequency, Lethality and Suicide Attacks of
Terror Events.” Peace Economics, Peace Science and Public Policy 18.
Kydd, Andrew and Barbara F. Walter. 2002. “Sabotaging the Peace: The Politics of Ex-
tremist Violence.” International Organization 56(2):263–296.
Lago, Ignacio and José Ramón Montero. 2006. “The 2004 Election in Spain: Terrorism,
Accountability, and Voting.” Taiwan Journal of Democracy 2(1):13–35.
Lake, David. 2002. “Rational Extremism: Understanding Terrorism in the Twenty-first
Century.” Dialogue IO 1(1):15–29.
34
Lyall, Jason. 2009. “Does Indiscriminate Violence Incite Insurgent Attacks? Evidence from
Chechnya.” Journal of Conflict Resolution 53(3):331–362.
Montalvo, Jose G. 2011. “Voting after the bombings: A natural experiment on the effect of
terrorist attacks on democratic elections.” Review of Economics and Statistics 93(4):1146–
1154.
Oates, Sarah, Lynda Lee Kaid and Mike Berry. 2009. Terrorism, Elections, and Democracy:
Political Campaigns in the United States, Great Britain, and Russia. New York, NY:
Palgrave Macmillan.
OCHA. 2007. West Bank and Gaza Strip Closure Maps. UN Office for the Coordination of
Humanitarian Affairs. Available at http://www.ochaopt.org/documents/A3_Closure_
Booklet_December_2007.zip.
Pape, Robert A. 2003. “The Strategic Logic of Suicide Terrorism.” American Political Science
Review 97(3):343–361.
Petrocik, John R. 1996. “Issue Ownership in Presidential Elections, With a 1980 Case Study.”
American Journal of Political Science 40(3):825–850.
Powell, Robert. 2007. “Defending against Terrorist Attacks with Limited Resources.” Amer-
ican Political Science Review 101:527–541.
Rubin, Uzi. 2011. “The Missile Threat from Gaza: From Nuisance to Strategic Threat.”
The Begin - Sadat Center for Strategic Studies, Bar Ilan University, Mideast Security and
Policy Studies No.91 December. Available at http://www.biu.ac.il/SOC/besa/MSPS91.
pdf.
Schofield, Norman and Itai Sened. 2005. “Multiparty competition in Israel, 1988-96.” British
Journal of Political Science 35(4):635.
35
Skitka, Linda J. 2005. “Patriotism or Nationalism? Understanding Post-September 11, 2001,
Flag-Display Behavior.” Journal of Applied Social Psychology 35(10):1995–2011.
Willer, Robb. 2004. “The Effects of Government-Issued Terror Warnings on Presidential
Approval Ratings.” Current Research in Social Psychology 10(1):1–12.
Wooldridge, Jeffrey M. 2010. Econometric Analysis of Cross Section and Panel Data: Second
Edition. 2nd ed. Cambridge, MA: MIT Press.
Wright, John R. 2012. “Unemployment and the Democratic Electoral Advantage.” American
Political Science Review 106(4).
Yagna, Yanir. 2010. “Rocket-battered Sderot faces bankruptcy.”.
URL: http://www.haaretz.com/print-edition/news/rocket-battered-sderot-faces-
bankruptcy-1.265304
Zeitzoff, Thomas. 2013. “The Effect of Emotions And Exposure to Violence on Intragroup
Conflict: A Laboratory in the Field Experiment in Southern Israel.” Political Psychology
Forthcoming.
36
Figure 1: Rocket Ranges, 2003-2009 Elections
37
Table 1: Population In Range Of Rockets
Election Year 1999 2003 2006 2009
Number of people (% of all population) 0 (0%) 65,557 (1%) 210,082 (3%) 1,028,178 (14%)
Number of localities (% of all localities) 0 (0%) 10 (4%) 13 (5%) 48 (17%)
Range from Gaza (km) NA 10 20.4 43
38
Table 2: Parties in political blocs (vote-shares in parentheses)
blocs
Year Left Center Right
1999 Labor (20%), Meretz (8%), Shinui (5%), Nationalistic: Likud (14%), Mafdal (4%),
Am Echad (2%), Balad (2%), Chadash (3%), haMerkaz (5%) haIchud haLeumi (3%)
Raam (3%) Ultra-Orthodox: Yahadut haTorah (4%),
Shas (13%)
Russian: Israel Baaliya (5%), Israel Beteinu (3%)
2003 Labor (14%), Meretz (5%), Shinui (12%), Nationalistic: Likud (29%), Mafdal (4%),
Am Echad (3%), Chadash-Taal (3%), haIchud haLeumi (6%)
Balad (2%), Raam (2%) Ultra-Orthodox: Yahadut haTorah (4%),
Shas (8%)
Russian: Israel Baaliya (2%)
2006 Labor (15%), Meretz (4%), Chadash (3%), Kadima (22%) Nationalistic: Likud (9%),
Balad (2%), Raam-Taal (3%) haIchud haLeumi - Mafdal (7%)
Ultra-Orthodox: Yahadut haTorah (5%),
Shas (10%)
Russian: Israel Beteinu (9%)
2009 Labor (10%), Meretz (3%), Chadash (3%), Kadima (22%) Nationalistic: Likud (22%),
Balad (2%),Raam (3%) haBait haYehudi (3%),
haIchud haLeumi (3%)
Ultra-Orthodox: Yahadut haTorah (4%),
Shas (8%)
Russian: Israel Beteinu (12%)
39
Table 3: Summary statistics of pre-treatment variables
Variable Mean Min. Max. Mean Min. Max.
In Range Out of Range
(N=71, 50 localities) (N=1029, 237 localities)
Mean wage in 199942 1597.99 609 4341 1633.00 741.93 5723
Population in 1995 (log) 7.12 3.09 11.91 8.33 3.10 13.33
Median age in 1995 25.85 13.10 37.60 23.83 10.90 38.90
Male-to-female ratio in 1995 1.16 0.9 6.95 1.06 0.44 6.95
Jews share in 1995 0.85 0 1 0.64 0 1
High school graduates in 1998 0.18 0.08 0.76 0.35 0 0.84
Socio-economic cluster in 1995 3.57 1 10 4.51 1 10
Asia origin in 1995 0.13 0 0.46 0.09 0 0.64
Africa origin in 1995 0.23 0 0.55 0.12 0 0.55
Soviet Union origin in 1995 0.10 0 0.52 0.09 0 0.54
Turnout in 1999 0.78 0.63 0.87 0.77 0.10 0.98
Right-wing vote-share in 1999 0.46 0.06 1 0.39 0 1
Regional capital 0.06 0 1 0.09 0 1
Distance to closest border (km) 8.84 0 38.82 11.34 0 41.83
40
Table 4: The Effect of Being in the Rockets Range on Right-Wing Parties’ Vote-Shares
DV: Right-Wing Vote-Share
(1) (2) (3) (4)
Coef. Std. err. Coef. Std. err. Coef. Std. err. Coef. Std. err.
In range 0.04*** (0.01) 0.06*** (0.01) 0.02*** (0.01) 0.04*** (0.01)
Suicide terror fatalities 0.02* (0.01) 0.01 (0.01) 0.02*** (0.01)
Mean wage 0.10*** (0.01) -0.01 (0.01) -0.00 (0.01)
Population 0.02 (0.02) -0.05** (0.02) -0.00 (0.00)
Median age -0.00 (0.00) -0.01*** (0.00) 0.00 (0.00)
Male-to-female ratio -0.06 (0.06) -0.04 (0.07) 0.09 (0.07)
Net migration -0.08 (0.09) -0.03 (0.08) -0.01 (0.02)
Jews share 0.10 (0.07) 0.06 (0.04) 0.10*** (0.02)
High school graduates -0.08*** (0.02) -0.00 (0.04) -0.10** (0.04)
Origin from Asia in 1995 -0.04 (0.04)
Origin from Africa in 1995 0.03 (0.03)
Origin from the Soviet Union in 1995 0.05 (0.03)
Regional capital 0.02** (0.01)
Distance to border -0.00 (0.00)
Lagged DV 0.84*** (0.02)
Constant 0.40*** (0.00) 0.29 (0.24) 1.03*** (0.20) -0.04 (0.09)
Locality fe yes yes yes no
Year dummies no no yes no
R20.015 0.140 0.532 0.856
No. of obs (localities) 1000 (269) 790 (257) 790 (257) 705 (254)
* p<0.10, ** p<0.05, *** p<0.01
Each column estimates a linear panel regression of locality right-wing vote-share on whether a locality is within rockets range.
41
Table 5: The Effect of Being in the Rockets Range on Right-Wing Parties’ Vote-Shares
Nationalistic Religious Russian
(1) (2) (3)
Coef. Std. err. Coef. Std. err. Coef. Std. err.
In range 0.03** (0.01) 0.01 (0.01) -0.01 (0.02)
Suicide terror fatalities -0.01 (0.03) 0.00 (0.00) 0.02 (0.02)
Mean wage -0.04*** (0.01) -0.01 (0.01) -0.06*** (0.01)
Population -0.02 (0.02) 0.00 (0.02) -0.02 (0.04)
Median age -0.00 (0.00) -0.00** (0.00) 0.01 (0.00)
Male-to-female ratio -0.08 (0.09) 0.04 (0.04) -0.02 (0.09)
Net migration -0.02 (0.12) -0.10** (0.04) 0.09* (0.05)
Jews share 0.03 (0.05) 0.02 (0.02) 0.03 (0.06)
High school graduates 0.02 (0.04) -0.01 (0.02) 0.01 (0.04)
Constant 0.69*** (0.22) 0.10 (0.19) 0.12 (0.39)
Locality fe yes yes yes
Year dummies yes yes yes
R20.525 0.127 0.052
No. of obs (localities) 796 (262) 796 (262) 796 (262)
* p<0.10, ** p<0.05, *** p<0.01
Column 1 estimates a linear panel regression of locality’s nationalistic vote-share on whether
a locality is within rockets’ range. Column 2 estimates a linear panel regression of locality’s
religious vote-share on whether a locality is within the rockets’ range. Column 3 estimates
a linear panel regression of locality’s vote-share of parties associated with immigrants from
the former Soviet Union on whether a locality is within the rockets’ range.
42
Table 6: The Effect of Being in the Rockets Range on Vote-Shares Conditional on Incumbent
Right-Wing Likud Kadima
(1) (2) (3)
Coef. Std. err. Coef. Std. err. Coef. Std. err.
In range 0.02*** (0.01) 0.02*** (0.01) 0.00 (0.01)
RightPM 0.11*** (0.02) 0.11*** (0.02)
In range ×RightPM -0.03*** (0.01) 0.01 (0.02)
Suicide terror fatalities 0.01 (0.01) -0.01 (0.01) 0.02*** (0.01)
Mean wage -0.01 (0.01) 0.01 (0.01) -0.02 (0.02)
Population -0.05** (0.02) 0.02 (0.02) 0.06 (0.05)
Median age -0.01*** (0.00) -0.00** (0.00) 0.00 (0.00)
Male-to-female ratio -0.05 (0.07) -0.01 (0.03) -0.41*** (0.14)
Net migration -0.03 (0.08) -0.09 (0.09) 0.07*** (0.03)
Jews share 0.06 (0.04) 0.08** (0.04) 1.17 (0.86)
High school graduates 0.00 (0.04) -0.00 (0.04) 0.04 (0.06)
Constant 0.93*** (0.19) -0.04 (0.20) -0.76 (0.65)
Locality fe yes yes yes
Year dummies yes yes yes
R20.532 0.604 0.049
No. of obs (Localities) 796 (262) 796 (262) 495 (257)
* p<0.10, ** p<0.05, *** p<0.01
Each column estimate a linear panel regression of vote-shares on whether
a locality is within range, conditional on the incumbent.
43
Table 7: Robustness Tests
Dropping observations outside common support Alternative measures of distance from Gaza
Right-Wing Nationalistic Right-Wing
(1) (2) (3) (4)
Coef. Std. err. Coef. Std. err. Coef. Std. err. Coef. Std. err.
In range 0.03*** (0.01) 0.05** (0.02) 0.02*** (0.01) 0.02*** (0.01)
Suicide terror fatalities 0.03*** (0.01) 0.04*** (0.01) 0.02 (0.02) 0.02 (0.02)
Mean wage -0.03 (0.02) -0.10** (0.04) -0.02** (0.01) -0.02** (0.01)
Population -0.15*** (0.05) -0.15** (0.06) -0.45** (0.16) -0.45** (0.16)
Median age -0.01 (0.01) -0.00 (0.01) -0.01*** (0.00) -0.01*** (0.00)
Male-to-female ratio 0.44* (0.24) 0.61* (0.34) -0.05 (0.07) -0.05 (0.07)
Net migration 0.52 (0.34) 0.91** (0.46) -0.11 (0.10) -0.12 (0.10)
Jews share -0.40 (0.59) -1.39 (1.05) 0.06 (0.04) 0.06 (0.04)
High school graduates -0.07 (0.06) 0.02 (0.10) 0.01 (0.04) 0.01 (0.04)
Constant 1.89*** (0.67) 1.97** (0.92) 1.50*** (0.37) 1.51*** (0.37)
Locality fe yes yes yes no
Year dummies no no no no
R20.573 0.586 0.533 0.533
No. of obs (localities) 391 (132) 391 (132) 823 (275) 823 (275)
* p<0.10, ** p<0.05, *** p<0.01
Columns 1 and 2 report results of a linear regression of right-wing and nationalistic vote-shares, respectively on whether
a locality is within rockets range, after dropping observations outside common support. Columns 3 and 4 report results
of a linear model of right-wing vote-share on whether a locality is within the rockets range using alternative measures
of distance. In column 3, distance is measured from Gaza perimeter to locality’s center. In column 4, distance is measured
from Gaza perimeter to locality’s perimeter, and 5km are subtracted from rockets’ range to accommodate launching
from within Gaza. Robust standard errors clustered at locality level are in parentheses.
44
Table 8: Ruling Out Turnout and Migration Explanations
Turnout Migration
DV: Turnout DV: Right-Wing DV: Right-Wing
(1) (2) (3) (4)
Coef. Std. err. Coef. Std. err. Coef. Std. err. Coef. Std. err.
In range 0.00 (0.01) 0.02** (0.01) 0.03*** (0.01) 0.03*** (0.01)
Suicide terror fatalities 0.03** (0.01) 0.05*** (0.01) -0.00 (0.03) -0.00 (0.04)
Mean wage -0.01 (0.01) 0.01 (0.02) -0.01 (0.01) -0.01 (0.01)
Population -0.81*** (0.25) -0.47 (0.35) -0.98*** (0.33) -0.71** (0.32)
Median age -0.00*** (0.00) -0.00 (0.00) -0.00 (0.00) -0.00 (0.00)
Male-to-female ratio -0.18** (0.08) -0.13 (0.08) 0.32* (0.17) 0.40** (0.18)
Net migration -0.16 (0.12) -0.04 (0.11) 1.81 (2.25) -1.68 (2.37)
Jews share 0.11 (0.12) -0.10 (0.32) -0.60 (0.45) -0.53 (0.37)
High school graduates -0.02 (0.03) -0.06 (0.09) -0.03 (0.05) -0.04 (0.05)
1999.year 0.09*** (0.02) -0.03 (0.06)
Constant 2.72*** (0.54) 1.81** (0.91) 2.67*** (0.85) 2.01** (0.82)
R20.443 0.487 0.529 0.551
No. of obs (localities) 823 (275) 323 (156) 497 (173) 556 (191)
* p<0.10, ** p<0.05, *** p<0.01
Columns 1 report results of a linear regression of turnout on whether a locality is within rockets range. Columns 2 - 3 report
results of a linear regression of right-wing vote-share on whether a locality is within rockets range. Column 2 examines
only those localities with above median turnout (>69%), and column 3 those localities with below median migration.
Column 4 uses a sample in which we assume that all incoming residents are right-wing, and all outgoing resident are left-wing
Robust standard errors clustered at locality level are in parentheses. All regressions include fixed effects and year dummies.
45
