Emerging Trends in The Social and Behavioral Sciences

Regime Type and Terrorist Attacks

Media

Part of Regime Type and Terrorist Attacks

Title

Regime Type and Terrorist Attacks

extracted text

Regime Type and Terrorist Attacks
KARA KINGMA, BRYAN CRAMER, and ERICA CHENOWETH

Abstract
We review the current literature on why democracies experience terrorist attacks.
Noting that most of these studies were based on data that ended in 2004, we update
the data and analyze regime type and terrorist attacks through 2012. We identify a
key trend: 2009 appears to have been a watershed year, where terrorist attacks began
to occur more often in failed states and countries under military occupation than in
democratic ones. Most strikingly, we find that autocratic regimes have experienced
a modest increase in terrorist attacks, whereas democracies have experienced a generalized decrease. We then review the literature on terrorism in autocratic regimes,
which is mainly focused on explaining variation in terrorist attack patterns across
autocratic regimes with different capacities for coercion and co-optation and different sensitivities to audience costs. We conclude by highlighting some of the research
and policy implications on regime type and terrorism.

INTRODUCTION: DOES DEMOCRACY INCREASE TERRORISM?
Policymakers in the United States have focused on democracy promotion as
a means to reduce terrorism based on the belief that terrorism emerges from
autocracies. In this view, terrorists resort to violence because they are frustrated by the lack of political and economic freedom in such systems and
are unable to express their grievances peacefully. However, many studies
have shown that terrorism is more common in democracies than in autocratic regimes (Eubank & Weinberg, 1994, 1998, 2001; Eyerman, 1998), leading
many scholars to challenge this commonly-held policy view.
A number of explanations posit that the very nature of democratic systems encourages terrorism by decreasing its costs and risks. Democracies
are distinctly open and provide particular freedoms of which terrorists can
take advantage. More specifically, the civil rights guaranteed by democracies
allow terrorists to organize and act within them. In her 1981 work, Crenshaw argues that in democracies, “Terrorists view the context as permissive,
making terrorism a viable option. In a material sense, the means are placed

Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

at their disposal by the environment” (p. 383). In addition, the justice systems of democracies offer legal and civil protections for terrorists who are
apprehended (Schmid, 1992). Terrorists need not worry about overly harsh
penalties or extreme repression under these regimes.
As this mechanism would predict, Eubank and Weinberg (1998, 2001) find
that terrorism is most common in the freest political systems. Democracies
are vulnerable to terrorism, they argue, because they provide terrorists
opportunities and because terrorists can coerce their elected officials. These
officials are receptive to the voices of terrorists and to the larger public
negatively affected by terrorist violence (Eubank & Weinberg, 1994). It
is thus strategic for terrorists to act against democracies because elected
governments will concede to their demands in order to avoid additional
violence. Pape (2003) uses this mechanism to explain instances of suicide
terrorism in democracies, arguing that the tactic is often successful in forcing
governments to make territorial compromises.
Democracies also guarantee freedom of the press. Many scholars have
argued that terrorists seek publicity and thus are motivated to conduct
attacks in democracies where their grievance-motivated violence will reach
a wide audience (Hoffman, 2006). A freer press leads to more terrorism
because “[t]he modern news media, as the principal conduit of information
about such acts … plays a vital part in the terrorists’ calculus” (Hoffman,
2006, p. 174). Gadarian (2010) points to this explanation in her study on
the effect of emotional media coverage of terrorist events on the foreign
policy views of citizens. Under conditions of freedom of the press, the media
competes for viewers and readers by highlighting terrorist events.
However, some scholars question the link between press freedom and
terrorism as an explanation for terrorism in democracies. In their view, a free
press leads to more reports of terrorism but not necessarily more terrorism.
In other words, terrorist attacks are accurately reported in democracies
and underreported in authoritarian systems that generally lack a free press
(Sandler, 1995). This mechanism is difficult to prove, but a number of studies
have attempted to do so. For example, Drakos and Gofas (2006) conclude
in their 2006 work that underreporting is indeed a problem in countries
with less free presses. In a 2007 follow-up study, they find that countries
that undergo improvements in their Press Freedom status, as measured by
Freedom House, report greater terrorist activity in the following years, while
those whose Press Freedom status worsens report less terrorist activity.
Another set of explanations look more closely at the particularities of
democracies to understand why terrorists target some countries and not
others. Scholars of these mechanisms disagree with those who would
argue that more democracy leads to more terrorism. Rather, understanding
why terrorism occurs in democratic regimes requires looking beyond the

Regime Type and Terrorist Attacks

3

simple divide between democracies and autocracies. First, the institutional
environments of some democracies may limit potential terrorist groups from
voicing their opinions. Second, some democratic institutions may lead to a
general ineffectiveness of government, producing grievances that terrorists
seek to resolve through violence.
As elected systems of government, democracies offer the institutional
means for would-be terrorists to make their grievances heard peacefully.
However, some provide more or less access to the marginalized or extremist
groups to which terrorists belong. Young and Dugan (2011) find that
democratic systems with more veto players experience more terrorism.
Governments in systems with fewer veto players can more easily change
policies to meet terrorists’ demands, so terrorists are incentivized to act
through nonviolent political channels. Terrorism is a more effective strategy
and more likely to occur when policy change is unlikely in systems with
more veto players. Li (2005) similarly demonstrates that democratic participation reduces terrorism except in the presence of institutional constraints.
Political gridlock is more likely in democracies with institutional constraints,
motivating terrorists to use violence.
Aksoy (2012) looks at differences in the opportunities for political access
across democracies to explain incidences of terrorism. She concludes that
democracies that make it easier for smaller groups to participate (through
proportional representational electoral systems and higher median district
magnitudes) are less likely to experience terrorism. In less permissive electoral systems, the probability of terrorism grows when elections are less than
2.5 years away. Aksoy and Carter (2014) extend this argument by considering the relationship between electoral institutions and terrorist groups’ goals.
They demonstrate that democracies are only more likely to experience terrorism from anti-system groups, or those groups whose goals cannot be accommodated in a democratic system. Further, electoral permissiveness has a significant effect on the likelihood of attacks by within-system terrorist groups.
Democracies also differ in the level of competition that exists within them.
As systems of elected governments, democracies are sites of contestation
among various interests. Terrorists might feel it necessary to adopt violent
strategies and outbid others in order to achieve their goals. Chenoweth
(2010) finds that political systems with higher levels of political competition experience more terrorism. She writes “ … terrorism proliferates in
democracies … because of intergroup dynamics, when political organizations are motivated to escalate their activities due to political competition”
(p. 16). Moore, Bakker, and Hill (2011) offer support for this argument,
concluding that electoral contestation has a positive effect on terrorism.
Bloom (2004) and Sanchez-Cuenca and Aguilar (2009) demonstrate similar
mechanisms to explain terrorism in Palestine and Spain, respectively.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Finally, some explanations posit that terrorists use violence because of
grievances stemming from the policy choices of democratic governments.
Arguments for this mechanism look at general ideological viewpoints and
specific economic and social policies. For example, Koch and Cranmer (2007)
show that governments on the left are more susceptible to terrorism because
terrorists expect them to compromise. On the other hand, Danzell (2011)
finds that right-leaning governments tend to restrict political freedoms when
in power, leading opposition parties on the left to turn to violence. Burgoon
(2006) and Piazza (2009, 2011) link policies aimed at reducing economic
inequality to lower levels of terrorism. Piazza and Walsh (2010a, 2010b) find
that democracies that better respect physical integrity rights and protections
from political imprisonment and extrajudicial killings are less likely to be
targeted by terrorism.
In sum, various mechanisms have been proposed to explain why terrorism is more common in democratic systems of government. However, some
democracies are more vulnerable to terrorism than others. Studies that show
democratic regimes increase terrorism do not account for important differences among democracies. As mentioned, one proposed mechanism links
terrorism to the opportunities provided by democracy. Yet Abadie (2006)
demonstrates that democracies with mid-level measures of political freedom
are more vulnerable to terrorism than either democracies with advanced levels of political freedom or autocracies. Kurrild-Klitgaard, Justesen, and Klemmensen (2006) argue that democracies with higher levels of civil liberties
experience terrorism less often. Choi (2010) finds that the quality of rule of
law in a democracy is negatively related to the likelihood of terrorism.
Democratic systems may ultimately discourage terrorism by providing
would-be terrorists political access, but mechanisms of participation and
grievance resolution take time to establish and may not be evident in new
democracies. Eyerman (1998) and Piazza (2012) show that terrorism is more
likely in the youngest democracies. New democracies may fail to provide
representation to potential terrorists and in the provision of economic
and civil goods, leading to grievances. Chenoweth (2012) supports this
argument in her 2012 study, finding that partial democracies, or systems
where “ … elections are not totally free and fair or political participation is
not routine or transparent” (p. 357), experience more terrorist attacks than
advanced democracies or autocracies from 1970 until the mid-2000s, when
countries under foreign occupation experience the most.
The most convincing mechanisms see terrorism as a product of institutions
and policies that create grievances and limit terrorist groups’ access to
the political system. Various studies have shown that new or transitioning democracies tend to do both. Other research has demonstrated that
advanced democracies are likely to experience relatively few instances of

Regime Type and Terrorist Attacks

5

terrorism unless they act in other countries’ affairs (Braithwaite & Sobek,
2005; Savun & Phillips, 2009). Consequently, previous studies suggest that as
democracy spreads and/or current partial democracies fail to consolidate,
democracies are likely to continue to experience more terrorism than other
regime types.
SOME EMPIRICAL TRENDS
Much scholarship on the association between regime type and terrorism
focuses on the 1968–2004 period. START (2011) Relying on the Global
Terrorism Database (GTD), we present some summary statistics to evaluate
whether these extant propositions hold through 2012.
We aggregated the latest version of the GTD to a country-year format, and
then merged it with the POLITY IV database (Marshall, Jaggers, & Gurr,
2013). Data on total terrorist attacks and Polity score—a 21-point scale from
totalitarian government (−10) to full democracy (+10)—are complete from
1970 to 2012, with the exception of 1993.1
Vreeland (2008) suggests that researchers carefully examine the particular
types of regimes that fall in the −5 to +5 range on the Polity scale. Following Chenoweth (2013), we expand Goldstone et al.’s (2010) classification of
regime type to identify seven relevant categories.
•

•

•

Democracy (Polity score: +7 or higher): a system that combines free
and fair elections with routine and transparent political participation.
Examples include OECD countries, Costa Rica, Uruguay, and Mongolia
(Goldstone et al., 2010, p. 195).
Partial Democracy (PD) (Polity score: 0 through +6): a system in which
citizens choose the chief executive among competitive candidates, but
the elections are not totally free and fair, or political participation is not
routine or transparent. Examples include Albania or Venezuela in recent
years (ibid, p. 195). This category also includes factionalized democracies, which possess polarization among localized blocs pursuing their
interests at the national level. This category would include Venezuela in
the early 2000s (ibid, p. 195).
Partial Autocracy (PA) (Polity score: −6 through 0): states that repress
political participation yet purport to hold competitive elections for
national office. Examples include apartheid-era South Africa. PAs may
also include countries that allow competitive elections for parliament
but do not hold competitive elections for the chief executive. Jordan is
an example. PAs may include factionalism (ibid, p. 195).

1. Data are missing for 1993 for the GTD because of lost primary materials for that year.

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

•

•

•

•

Autocracy (Polity score: −7 through −10): these systems do not have effective competition for the chief executive, nor do they allow free political
participation. Examples include North Korea, China, Saudi Arabia, and
Libya under Qaddafi (ibid, p. 195).
Failed State (Polity score: −77): a country in which institutions have collapsed and there are no remaining functional sovereign political or governance institutions (Somalia 1992).
Under Occupation (Polity score: −66): a country that with no functional
sovereign political or governance institutions because an external country has removed them (Iraq 2004).
Transitioning (Polity score: −88): a country cannot be categorized as any
of the above regime types because its institutional structure is currently
in flux.

0

Total number of terrorist attacks (GTD)
1000 2000 3000 4000 5000

We then calculated the total number of terrorist attacks in each regime category from 1970 to 2012 (excluding 1993 due to missing data) according to
the GTD (LaFree & Dugan, 2007).
Figure 1 shows the distribution of terrorist attacks in the seven different
regime type categories. These absolute counts illustrate fairly clearly that,

1970

1980

1990
Year

2000

Democracies

Partial democracies

Autocracies
Failed states

Partial autocracies

2010

Occupied countries

Countries in transition

Figure 1 Number of terrorist attacks by regime type, 1970–2012.

200

400

600

800

7

0

Average number of terrorist attacks per regime type (GTD)

Regime Type and Terrorist Attacks

1970

1980

1990
Year

2000

Democracies

Partial democracies

Autocracies
Failed states

Partial autocracies

2010

Occupied countries

Countries in transition

Figure 2 Average number of terrorist attacks per regime type, 1970–2012.

consistent with previous literature, democracies tended to possess the highest number of terrorist attacks until the mid-2000s, when partial democracies,
autocracies, and countries under foreign occupation (Iraq and Afghanistan)
emerged as more common venues of terrorist attacks.
Of course, the global distribution of regime types has varied considerably
over time. As such, we normalized the absolute counts of terrorist attacks
by dividing the total number of attacks in each category by the total number
of countries in each regime type category to examine the average number of
terrorist attacks per regime category per year (Figure 2).
Figure 2 shows some striking differences. The average democracy experienced a stable number of terrorist attacks throughout the series, but it is
not the most explosive category. Instead, during the early 1980s, countries in
transition (e.g., El Salvador) and failed states (e.g., Lebanon) experienced the
highest number of terrorist attacks. And beginning in 2003, occupied countries (e.g., Iraq and Afghanistan) experienced the highest average number of
annual attacks per country.
Importantly, however, these trends are driven by small numbers of cases,
as there are very few countries in transition, failed states, or countries under
occupation in a given year. As such, the high number of average attack counts

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Table 1
Bivariate Association Between Polity Score and Terrorist Attacks
(Negative Binomial Regression)a
Model 1 (1970–2012) Model 2 (1970–2009) Model 3 (2010–2012)
DV: Number of
DV: Number of
DV: Number of Terrorist
Terrorist Attacks
terrorist attacks
Attacks
Polity2
Population (logged)
Constant
N
Wald chi2
Prob > chi2

0.117*** (0.019)
0.654*** (0.100)
−15.663*** (3.803)
6473
68.89
0.0000

0.082*** (0.022)
0.610*** (0.105)
−7.629*** (1.761)
5984
61.14
0.0000

0.049 (0.047)
1.096*** (0.232)
−15.66*** (3.803)
489
24.37
0.000

a Estimates produced using a zero-inflated event count did not significantly alter these results. Moreover,
a vuong test reveals that a zero-inflated negative binominal does not significantly improve the model over
a negative binomial regression in Model 3. For simplicity, the NBR is presented here; the ZINB commands
and results are available in the replication file, which can be accessed at www.ericachenoweth.com.
***p < 0.001.

may be driven by one or two cases in each category and should be taken with
a grain of salt.2
That said, these shifting distributions over time do suggest some possible temporal effects in the most common terrorist attack venue. Once
we exclude key outlying states—such as El Salvador, Lebanon, Iraq, and
Afghanistan—we may be seeing a shift in the types of regime that experience
the most terrorist attacks.
We therefore estimated a negative binomial regression on the association
between Polity score and terrorist attacks in general (controlling for logged
population from Banks, 2013), and then stratified the results by time period
(Table 1).
As with prior literature, we find that there is a strong, positive association
between the Polity score and the number of terrorist attacks (Model 1). We
find, however, that this positive association is conditional on whether the
data series takes place before 2010. Through 2009, there was a strong positive
association (Model 2), with both statistical and substantive findings nearly
identical to those produced in Model 1. However, from 2010 to 2012, the statistical significance of that association disappears, suggesting that in those
years, the effects of Polity score on terrorist attacks are not different from 0
(Model 3).
In Figure 3, we identify the predicted counts of terrorist attacks from Models 1 through 3, which we calculated using the prtab command in Stata.
2. On the basis of Figure 2, it is easier to understand policymakers’ preoccupations with weak and
failed states as the most vulnerable to terrorism.

Regime Type and Terrorist Attacks

9

60

50

40
Entire series (1970–2012)

30

1970–2009
2010–2012

20

10

0

−10 −8

−6

−4

−2

0

2

4

6

8

10

Figure 3 Predicted number of annual terrorist attacks by regime type, 1970–2012.

Several important findings are clear. First, overall terrorist events have
increased during 2010–2012 worldwide and in every regime type. On
average, however, controlling for population size, the predicted number of
annual terrorist attacks in full democracies declined to about 13 in the period
2010–2012 from about 18 in previous years. Second, although the time period
2010–2012 still shows a positive association between regime type and terrorist attacks, it is notable that countries with the lowest Polity scores—between
−10 and −5—have experienced a substantive increase in the number of
terrorist attacks they experienced during the pre-2010 period. Although it is
likely too early to speculate whether these trends will continue, this finding
suggests an important secular departure from previous trends.
WHICH NONDEMOCRACIES WILL SEE MORE TERRORISM?
SOME CLUES FROM THE LITERATURE
As Crenshaw notes, regimes that deny access to power and persecute dissenters create dissatisfaction (Crenshaw, 1981, p. 383). Previous research suggests that whether autocracies experience terrorism depends on their ability
to coerce and co-opt regime challenges (Wilson & Piazza, 2013, p. 942). Much
of the literature therefore concentrates on two types of coercive and co-optive
regime attributes—institutions and audience costs—which are highly variable across authoritarian regimes.
The different forms autocracies take matter when it comes to their ability to
coerce and co-opt opponents. For example, drawing on Geddes’ categories

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

of political dictatorship, Wilson and Piazza find that single-party autocracies
are better able to employ both coercion and co-optation to address political
opposition (Wilson & Piazza, 2013, p. 953). Single-party regimes can coerce
opponents through their control over the military, but they can also co-opt
opponents through their influence on institutions such as legislatures, political parties, and courts (Wilson & Piazza, 2013, p. 945). On the other hand,
democracies must try to co-opt terrorists because their ability to coerce terrorists is limited by guarantees of civil, political, and human rights (Wilson
& Piazza, 2013, p. 945). Military regimes are limited to coercion because they
are less capable of co-opting opponents and need to rely on coercive power to
maintain control (Wilson & Piazza, 2013, p. 945). Thus, those autocracies that
are able to complement repression with co-optive institutions and to deter
challenges while accommodating others are less prone to terrorism (Wilson
& Piazza, 2013, p. 952).
Others find that autocracies use formal institutions as mechanisms
through which to reduce the risk of terrorism. Aksoy, Carter, and Wright
(2012) explore how political party activity in dictatorships helps to mobilize
discontented opposition groups. Specifically, the ability of opposition groups
to participate in legislatures and voice their grievances may help to mitigate
the possibility these groups will pursue violence. Repression is a costly
and risky strategy for dictators, which is why they seek to purchase loyalty
to the regime by co-opting potential challengers. By co-opting potential
challengers, dictatorships bring their opponents into the political system
allowing them to gain some concessions from the regime (ibid, 2012). Once
opponents have been co-opted, through formalized roles in the legislature,
Aksoy et al. (2012) note they will be less likely to risk challenging the regime
for fear of losing their share of the regime’s spoils. However, the presence of
opposition political parties without a functional legislature makes regimes
more likely targets of terrorism, further supporting Wilson and Piazza’s
hypothesis on the fundamental importance of co-optation as a tool of
authoritarian management of potential terrorist groups (ibid, 2012). In fact,
Aksoy et al.’s (2012, p. 818) data reveal that dictatorships with opposition
parties and no legislatures experience a nearly 60% higher probability of
emergence of violent groups. As long as autocracies can use institutions such
as legislatures and elections to successfully co-opt potential challengers,
Aksoy et al. argue, they need not resort to repression.
Dictators also have to be concerned with the potential audience costs in
the event of terrorist attacks. Audience costs are the domestic punishment
that leaders face for backing down from public threats (Fearon, 1994, pp.
577–592). Autocratic leaders face accountability threats not from the public,
but rather from the domestic, regime elites on whom they rely for support. It
is these elites that can act like voting publics in democracies by coordinating

Regime Type and Terrorist Attacks

11

to punish an authoritarian leader (Weeks, 2008, p. 36). If dictators are unable
to respond to the challenge of terrorist actions, regime elites may hold the
leader accountable by replacing him. Aksoy, Carter, and Wright (2013, pp.
4–5) find strong evidence that increasing political violence is associated with
a higher risk of a reshuffling coup (where an autocratic leader is replaced but
not the entire regime). C. R. Conrad, J. Conrad, and Young (2014) suggest
that these authoritarian audience costs may increase incentives to use terrorism against authoritarians. Because terrorists want their actions to have as
much of an impact as possible, there are greater incentives to launch terrorist
actions against states with high audience costs.
What, then, are the key takeaways from this literature in further understanding terrorism in autocracies? The first trend would be to look more
closely at state behavior itself (Chenoweth, 2013, p. 375). Wilson and Piazza’s
article is a step in this direction, along with Aksoy et al.’s suggestions to
attempt to understand the interaction of opposition groups and dictators
(2012, p. 823) and how terrorism influences incumbent dictators (2013, p. 6).
The growing research on audience costs is a trend that also opens promising research into future understanding of trends between terrorism and
autocracy. As Weeks notes, the literature on accountability remains narrow
and much remains to be learned about domestic politics in nondemocracies
(Weeks, 2008, pp. 59–60). Autocracies are not as monolithic as they were
once understood to be, and perhaps widespread acceptance of this reality
has made them more likely targets of terrorist attacks.
THE FUTURE OF REGIME TYPE AND TERRORIST ATTACKS
While we are not certain whether the autocracy-terrorism trend will
continue, we can draw several conclusions from recent trends. First, the
emergence of semi-autocratic systems and “hybrid regimes” is likely to
complicate previous assumptions about the reasons why democracies were
so susceptible to terrorism (Levitsky & Way, 2011). Indeed, many of the
mechanisms, which thought to make democracies more vulnerable—such as
political competition, mobilization, and transition—are now quite common
to many autocracies.
Second, we do see variation in the types of autocracies and transitioning
countries that experience terrorism. As such, a promising line of research
is attempting to explain this variation. Future research that focuses on the
behavioral aspects of autocracy is likely to yield new insights into the relationship between these regimes and terrorist groups.
Finally, it is clear that countries under military occupation—including Iraq
and Afghanistan—have accounted for much of the terrorist violence in recent
years. As such, an important development to note is how the United States

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

and NATO withdrawal from Afghanistan—and the Afghan government’s
ability to consolidate power—will affect terrorism there in the future. However, recent trends do not paint an optimistic picture. If these trends continue,
then neither a slide toward autocracy nor a consolidation of democracy is
likely to reduce terrorist violence in Afghanistan. Indeed, if recent trends persist or accelerate, the regime type of the host country may have little or no
effect on prospects for future violence.
REFERENCES
Abadie, A. (2006). Poverty, political freedom, and the roots of terrorism. American
Economic Review, 96(2), 50–56.
Aksoy, D. (2012). Elections and the timing of terrorist attacks in democracies. Working paper, Department of Politics, Princeton University.
Aksoy, D., & Carter, D. B. (2014). Electoral institutions and the emergence of terrorist
groups. British Journal of Political Science, 44(1), 181–204.
Aksoy, D., Carter, D. B., & Wright, J. (2012). Terrorism in dictatorships. Journal of
Politics, 74(3), 810–826.
Aksoy, D., Carter, D. B., & Wright, J. (2013). Terrorism and the fate of dictators. Working paper, Princeton University.
Banks, A. (2013). Cross-national time series data archive. Jerusalem, Israel: Databanks
International.
Bloom, M. (2004). Palestinian suicide bombing: Public support, market share, and
outbidding. Political Science Quarterly, 119(1), 61–88.
Braithwaite, A., & Sobek, D. A. (2005). Victim of success: American dominance and
international terrorism. Conflict Management and Peace Science, 22(2), 135–148.
Burgoon, B. (2006). On welfare and terror: Social welfare policies and politicaleconomic roots of terrorism. Journal of Conflict Resolution, 50(2), 176–203.
Chenoweth, E. (2010). Democratic competition and terrorist activity. The Journal of
Politics, 72(1), 16–30.
Chenoweth, E. (2012). Is terrorism still a democratic phenomenon? The Journal of
International Relations, 32(8), 85–100.
Chenoweth, E. (2013). Terrorism and democracy. Annual Review of Political Science,
16, 85–100.
Choi, S. W. (2010). Fighting terrorism through the rule of law? Journal of Conflict Resolution, 54(6), 940–966.
Conrad, C. R., Conrad, J., & Young, J. K. (2014). Tyrants and terrorism: Why some
autocrats are terrorized while others are not. International Studies Quarterly, 58(3),
539–549.
Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399.
Danzell, O. E. (2011). Political parties: When do they turn to terror? Journal of Conflict
Resolution, 55(1), 85–105.
Drakos, K., & Gofas, A. (2006). The devil you know but are afraid to face: Underreporting bias and its distorting effects on the study of terrorism. Journal of Conflict
Resolution, 50(5), 714–735.

Regime Type and Terrorist Attacks

13

Drakos, K., & Gofas, G. (2007). Evidence for the existence of under-reporting bias
in observed terrorist activity: The message in press freedom status transitions.
Democracy and Security, 3, 139–155.
Eubank, W. L., & Weinberg, L. (1994). Does democracy encourage terrorism? Terrorism and Political Violence, 6(4), 417–463.
Eubank, W. L., & Weinberg, L. (1998). Terrorism and democracy: What recent events
disclose. Terrorism and Political Violence, 10(1), 108–118.
Eubank, W. L., & Weinberg, L. (2001). Terrorism and democracy: Perpetrators and
victims. Terrorism and Political Violence, 13(1), 155–164.
Eyerman, J. (1998). Terrorism and democratic states: Soft targets or accessible systems. International Interactions, 24(2), 151–170.
Fearon, J. D. (1994). Domestic political audiences and the escalation of international
disputes. The American Political Science Review, 88(3), 577–592.
Gadarian, S. K. (2010). The politics of threat: How terrorism shapes foreign policy
attitudes. The Journal of Politics, 72(2), 469–483.
Goldstone, J. A., Bates, R. H., Epstein, D. L., Gurr, T. R., Lustik, M. B., Marshall,
M. G., … , Woodward, M. (2010). A global model for forecasting political instability. American Journal of Political Science, 54(1),190–208.
Hoffman, B. (2006). Inside terrorism (2nd ed.). New York, NY: Columbia University
Press.
Koch, M. T., & Cramner, S. (2007). Testing the ‘Dick Cheney’ hypothesis: Do governments of the left attract more terrorism than governments of the right? Conflict
Management and Peace Science, 4, 311–326.
Kurrild-Klitgaard, P., Justesen, M. K., & Klemmensen, R. (2006). The political economy of freedom, democracy, and transnational terrorism. Public Choice, 128(1),
289–315.
LaFree, G., & Dugan, L. K. (2007). Introducing the Global Terrorism Database. Terrorism and Political Violence, 19, 181–204.
Levitsky, S., & Way, L. A. (2011). Competitive authoritarianism: Hybrid regimes after the
Cold War. New York, NY: Cambridge University Press.
Li, Q. (2005). Does democracy promote or reduce transnational terrorist incidents?
Journal of Conflict Resolution, 2, 278–297.
Marshall, M. G., Jaggers, K., & Gurr, T. R. (2013). POLITY IV Project: Political regime
characteristics and transitions, 1800–2012 Dataset User’s Manual. Retrieved from
http://www.systemicpeace.org/polity/polity4.htm
Moore, W.H, Bakker, R., & Hill, D. W. (2011). How Much Terror? Dissidents, governments, institutions, and the cross-national study of terror attacks. Working paper,
Department of Political Science, Florida State University.
National Consortium for the Study of Terrorism and Responses to Terrorism
(START). (2011). The Global Terrorism Database. College Park, MD. Retrieved from
http://www.start.umd.edu/gtd/
Pape, R. A. (2003). The strategic logic of suicide terrorism. American Political Science
Review, 97(3), 343–361.
Piazza, J. A. (2009). Economic development, unresolved political conflict, and terrorism in India. Studies in Conflict & Terrorism, 32, 406–419.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Piazza, J. A. (2011). Poverty, minority economic discrimination and domestic terrorism. Journal of Peace Research, 48(3), 339–353.
Piazza, J. A. (2012). Regime age and terrorism: Are new democracies prone to terrorism? Presented at Annual Meeting of the International Studies Association, 53rd,
San Diego, 1–4 April.
Piazza, J. A., & Walsh, J. L. (2010a). Why respecting physical integrity rights reduces
terrorism. Comparative Political Studies, 43(5), 551–577.
Piazza, J. A., & Walsh, J. L. (2010b). Physical integrity rights and terrorism. PS: Political Science and Politics, 43(4), 411–414.
Sanchez-Cuenca, I., & Aguilar, P. (2009). Terrorist violence and popular mobilization:
The case of the Spanish transition to democracy. Politics & Society, 37(3), 428–453.
Sandler, T. (1995). On the relationship between democracy and terrorism. Terrorism
and Political Violence, 7(4), 1–9.
Savun, B., & Phillips, B. (2009). Democracy, foreign policy, and terrorism. The Journal
of Conflict Resolution, 53(6), 878–904.
Schmid, A. (1992). Terrorism and democracy. Terrorism and Political Violence, 4(4),
14–25.
Vreeland, J. R. (2008). The effect of political regime on civil war: Unpacking anocracy.
Journal of Conflict Resolution, 52(3), 401–425.
Weeks, J. L. (2008). Autocratic audience costs: Regime type and signaling resolve.
International Organization, 62(1), 35–64.
Wilson, M. C., & Piazza, J. A. (2013). Autocracies and terrorism: Conditioning effects
of authoritarian regime type on terrorist attacks. American Journal of Political Science, 57(4), 941–955.
Young, J. K., & Dugan, L. K. (2011). Veto players and terror. Journal of Peace Research,
48(1), 19–33.

FURTHER READING
Askoy, D., & Carter, D. B. (2014). Electoral institutions and the emergence of terrorist
groups. British Journal of Political Science, 44(1), 181–204.
Aksoy, D., Carter, D. B., & Wright, J. (2012). Terrorism in dictatorships. Journal of
Politics, 74(3), 810–826.
Chenoweth, E. (2013). Terrorism and democracy. Annual Review of Political Science,
16, 85–100.
Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399.
Wilson, M. C., & Piazza, J. A. (2013). Autocracies and terrorism: Conditioning effects
of authoritarian regime type on terrorist attacks. American Journal of Political Science, 57(4), 941–955.

KARA KINGMA SHORT BIOGRAPHY
Kara Kingma is a PhD student in International Studies at the Josef Korbel
School of International Studies. She focuses on the fields of Comparative Politics and International Political Economy.

Regime Type and Terrorist Attacks

15

BRYAN CRAMER SHORT BIOGRAPHY
Bryan Cramer is a Project Assistant at the National Democratic Institute and
a graduate of the MA program at the Josef Korbel School of International
Studies.
ERICA CHENOWETH SHORT BIOGRAPHY
Erica Chenoweth is an Associate Professor at the Josef Korbel School of International Studies, University of Denver, and an Associate Senior Researcher
at the Peace Research Institute Oslo (PRIO).
RELATED ESSAYS
Varieties of Capitalism (Political Science), Peter A. Hall
Emerging Trends in Social Network Analysis of Terrorism and Counterterrorism (Sociology), David Knoke
Why Do States Sign Alliances? (Political Science), Brett Ashley Leeds
Participatory Governance (Political Science), Stephanie L. McNulty and Brian
Wampler
Intervention and Regime Change (Political Science), John M. Owen IV and
Roger G. Herbert Jr.
War and Social Movements (Political Science), Sidney Tarrow
Trends in the Analysis of Interstate Rivalries (Political Science), William R.
Thompson



Regime Type and Terrorist Attacks
KARA KINGMA, BRYAN CRAMER, and ERICA CHENOWETH

Abstract
We review the current literature on why democracies experience terrorist attacks.
Noting that most of these studies were based on data that ended in 2004, we update
the data and analyze regime type and terrorist attacks through 2012. We identify a
key trend: 2009 appears to have been a watershed year, where terrorist attacks began
to occur more often in failed states and countries under military occupation than in
democratic ones. Most strikingly, we find that autocratic regimes have experienced
a modest increase in terrorist attacks, whereas democracies have experienced a generalized decrease. We then review the literature on terrorism in autocratic regimes,
which is mainly focused on explaining variation in terrorist attack patterns across
autocratic regimes with different capacities for coercion and co-optation and different sensitivities to audience costs. We conclude by highlighting some of the research
and policy implications on regime type and terrorism.

INTRODUCTION: DOES DEMOCRACY INCREASE TERRORISM?
Policymakers in the United States have focused on democracy promotion as
a means to reduce terrorism based on the belief that terrorism emerges from
autocracies. In this view, terrorists resort to violence because they are frustrated by the lack of political and economic freedom in such systems and
are unable to express their grievances peacefully. However, many studies
have shown that terrorism is more common in democracies than in autocratic regimes (Eubank & Weinberg, 1994, 1998, 2001; Eyerman, 1998), leading
many scholars to challenge this commonly-held policy view.
A number of explanations posit that the very nature of democratic systems encourages terrorism by decreasing its costs and risks. Democracies
are distinctly open and provide particular freedoms of which terrorists can
take advantage. More specifically, the civil rights guaranteed by democracies
allow terrorists to organize and act within them. In her 1981 work, Crenshaw argues that in democracies, “Terrorists view the context as permissive,
making terrorism a viable option. In a material sense, the means are placed

Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

at their disposal by the environment” (p. 383). In addition, the justice systems of democracies offer legal and civil protections for terrorists who are
apprehended (Schmid, 1992). Terrorists need not worry about overly harsh
penalties or extreme repression under these regimes.
As this mechanism would predict, Eubank and Weinberg (1998, 2001) find
that terrorism is most common in the freest political systems. Democracies
are vulnerable to terrorism, they argue, because they provide terrorists
opportunities and because terrorists can coerce their elected officials. These
officials are receptive to the voices of terrorists and to the larger public
negatively affected by terrorist violence (Eubank & Weinberg, 1994). It
is thus strategic for terrorists to act against democracies because elected
governments will concede to their demands in order to avoid additional
violence. Pape (2003) uses this mechanism to explain instances of suicide
terrorism in democracies, arguing that the tactic is often successful in forcing
governments to make territorial compromises.
Democracies also guarantee freedom of the press. Many scholars have
argued that terrorists seek publicity and thus are motivated to conduct
attacks in democracies where their grievance-motivated violence will reach
a wide audience (Hoffman, 2006). A freer press leads to more terrorism
because “[t]he modern news media, as the principal conduit of information
about such acts … plays a vital part in the terrorists’ calculus” (Hoffman,
2006, p. 174). Gadarian (2010) points to this explanation in her study on
the effect of emotional media coverage of terrorist events on the foreign
policy views of citizens. Under conditions of freedom of the press, the media
competes for viewers and readers by highlighting terrorist events.
However, some scholars question the link between press freedom and
terrorism as an explanation for terrorism in democracies. In their view, a free
press leads to more reports of terrorism but not necessarily more terrorism.
In other words, terrorist attacks are accurately reported in democracies
and underreported in authoritarian systems that generally lack a free press
(Sandler, 1995). This mechanism is difficult to prove, but a number of studies
have attempted to do so. For example, Drakos and Gofas (2006) conclude
in their 2006 work that underreporting is indeed a problem in countries
with less free presses. In a 2007 follow-up study, they find that countries
that undergo improvements in their Press Freedom status, as measured by
Freedom House, report greater terrorist activity in the following years, while
those whose Press Freedom status worsens report less terrorist activity.
Another set of explanations look more closely at the particularities of
democracies to understand why terrorists target some countries and not
others. Scholars of these mechanisms disagree with those who would
argue that more democracy leads to more terrorism. Rather, understanding
why terrorism occurs in democratic regimes requires looking beyond the

Regime Type and Terrorist Attacks

3

simple divide between democracies and autocracies. First, the institutional
environments of some democracies may limit potential terrorist groups from
voicing their opinions. Second, some democratic institutions may lead to a
general ineffectiveness of government, producing grievances that terrorists
seek to resolve through violence.
As elected systems of government, democracies offer the institutional
means for would-be terrorists to make their grievances heard peacefully.
However, some provide more or less access to the marginalized or extremist
groups to which terrorists belong. Young and Dugan (2011) find that
democratic systems with more veto players experience more terrorism.
Governments in systems with fewer veto players can more easily change
policies to meet terrorists’ demands, so terrorists are incentivized to act
through nonviolent political channels. Terrorism is a more effective strategy
and more likely to occur when policy change is unlikely in systems with
more veto players. Li (2005) similarly demonstrates that democratic participation reduces terrorism except in the presence of institutional constraints.
Political gridlock is more likely in democracies with institutional constraints,
motivating terrorists to use violence.
Aksoy (2012) looks at differences in the opportunities for political access
across democracies to explain incidences of terrorism. She concludes that
democracies that make it easier for smaller groups to participate (through
proportional representational electoral systems and higher median district
magnitudes) are less likely to experience terrorism. In less permissive electoral systems, the probability of terrorism grows when elections are less than
2.5 years away. Aksoy and Carter (2014) extend this argument by considering the relationship between electoral institutions and terrorist groups’ goals.
They demonstrate that democracies are only more likely to experience terrorism from anti-system groups, or those groups whose goals cannot be accommodated in a democratic system. Further, electoral permissiveness has a significant effect on the likelihood of attacks by within-system terrorist groups.
Democracies also differ in the level of competition that exists within them.
As systems of elected governments, democracies are sites of contestation
among various interests. Terrorists might feel it necessary to adopt violent
strategies and outbid others in order to achieve their goals. Chenoweth
(2010) finds that political systems with higher levels of political competition experience more terrorism. She writes “ … terrorism proliferates in
democracies … because of intergroup dynamics, when political organizations are motivated to escalate their activities due to political competition”
(p. 16). Moore, Bakker, and Hill (2011) offer support for this argument,
concluding that electoral contestation has a positive effect on terrorism.
Bloom (2004) and Sanchez-Cuenca and Aguilar (2009) demonstrate similar
mechanisms to explain terrorism in Palestine and Spain, respectively.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Finally, some explanations posit that terrorists use violence because of
grievances stemming from the policy choices of democratic governments.
Arguments for this mechanism look at general ideological viewpoints and
specific economic and social policies. For example, Koch and Cranmer (2007)
show that governments on the left are more susceptible to terrorism because
terrorists expect them to compromise. On the other hand, Danzell (2011)
finds that right-leaning governments tend to restrict political freedoms when
in power, leading opposition parties on the left to turn to violence. Burgoon
(2006) and Piazza (2009, 2011) link policies aimed at reducing economic
inequality to lower levels of terrorism. Piazza and Walsh (2010a, 2010b) find
that democracies that better respect physical integrity rights and protections
from political imprisonment and extrajudicial killings are less likely to be
targeted by terrorism.
In sum, various mechanisms have been proposed to explain why terrorism is more common in democratic systems of government. However, some
democracies are more vulnerable to terrorism than others. Studies that show
democratic regimes increase terrorism do not account for important differences among democracies. As mentioned, one proposed mechanism links
terrorism to the opportunities provided by democracy. Yet Abadie (2006)
demonstrates that democracies with mid-level measures of political freedom
are more vulnerable to terrorism than either democracies with advanced levels of political freedom or autocracies. Kurrild-Klitgaard, Justesen, and Klemmensen (2006) argue that democracies with higher levels of civil liberties
experience terrorism less often. Choi (2010) finds that the quality of rule of
law in a democracy is negatively related to the likelihood of terrorism.
Democratic systems may ultimately discourage terrorism by providing
would-be terrorists political access, but mechanisms of participation and
grievance resolution take time to establish and may not be evident in new
democracies. Eyerman (1998) and Piazza (2012) show that terrorism is more
likely in the youngest democracies. New democracies may fail to provide
representation to potential terrorists and in the provision of economic
and civil goods, leading to grievances. Chenoweth (2012) supports this
argument in her 2012 study, finding that partial democracies, or systems
where “ … elections are not totally free and fair or political participation is
not routine or transparent” (p. 357), experience more terrorist attacks than
advanced democracies or autocracies from 1970 until the mid-2000s, when
countries under foreign occupation experience the most.
The most convincing mechanisms see terrorism as a product of institutions
and policies that create grievances and limit terrorist groups’ access to
the political system. Various studies have shown that new or transitioning democracies tend to do both. Other research has demonstrated that
advanced democracies are likely to experience relatively few instances of

Regime Type and Terrorist Attacks

5

terrorism unless they act in other countries’ affairs (Braithwaite & Sobek,
2005; Savun & Phillips, 2009). Consequently, previous studies suggest that as
democracy spreads and/or current partial democracies fail to consolidate,
democracies are likely to continue to experience more terrorism than other
regime types.
SOME EMPIRICAL TRENDS
Much scholarship on the association between regime type and terrorism
focuses on the 1968–2004 period. START (2011) Relying on the Global
Terrorism Database (GTD), we present some summary statistics to evaluate
whether these extant propositions hold through 2012.
We aggregated the latest version of the GTD to a country-year format, and
then merged it with the POLITY IV database (Marshall, Jaggers, & Gurr,
2013). Data on total terrorist attacks and Polity score—a 21-point scale from
totalitarian government (−10) to full democracy (+10)—are complete from
1970 to 2012, with the exception of 1993.1
Vreeland (2008) suggests that researchers carefully examine the particular
types of regimes that fall in the −5 to +5 range on the Polity scale. Following Chenoweth (2013), we expand Goldstone et al.’s (2010) classification of
regime type to identify seven relevant categories.
•

•

•

Democracy (Polity score: +7 or higher): a system that combines free
and fair elections with routine and transparent political participation.
Examples include OECD countries, Costa Rica, Uruguay, and Mongolia
(Goldstone et al., 2010, p. 195).
Partial Democracy (PD) (Polity score: 0 through +6): a system in which
citizens choose the chief executive among competitive candidates, but
the elections are not totally free and fair, or political participation is not
routine or transparent. Examples include Albania or Venezuela in recent
years (ibid, p. 195). This category also includes factionalized democracies, which possess polarization among localized blocs pursuing their
interests at the national level. This category would include Venezuela in
the early 2000s (ibid, p. 195).
Partial Autocracy (PA) (Polity score: −6 through 0): states that repress
political participation yet purport to hold competitive elections for
national office. Examples include apartheid-era South Africa. PAs may
also include countries that allow competitive elections for parliament
but do not hold competitive elections for the chief executive. Jordan is
an example. PAs may include factionalism (ibid, p. 195).

1. Data are missing for 1993 for the GTD because of lost primary materials for that year.

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

•

•

•

•

Autocracy (Polity score: −7 through −10): these systems do not have effective competition for the chief executive, nor do they allow free political
participation. Examples include North Korea, China, Saudi Arabia, and
Libya under Qaddafi (ibid, p. 195).
Failed State (Polity score: −77): a country in which institutions have collapsed and there are no remaining functional sovereign political or governance institutions (Somalia 1992).
Under Occupation (Polity score: −66): a country that with no functional
sovereign political or governance institutions because an external country has removed them (Iraq 2004).
Transitioning (Polity score: −88): a country cannot be categorized as any
of the above regime types because its institutional structure is currently
in flux.

0

Total number of terrorist attacks (GTD)
1000 2000 3000 4000 5000

We then calculated the total number of terrorist attacks in each regime category from 1970 to 2012 (excluding 1993 due to missing data) according to
the GTD (LaFree & Dugan, 2007).
Figure 1 shows the distribution of terrorist attacks in the seven different
regime type categories. These absolute counts illustrate fairly clearly that,

1970

1980

1990
Year

2000

Democracies

Partial democracies

Autocracies
Failed states

Partial autocracies

2010

Occupied countries

Countries in transition

Figure 1 Number of terrorist attacks by regime type, 1970–2012.

200

400

600

800

7

0

Average number of terrorist attacks per regime type (GTD)

Regime Type and Terrorist Attacks

1970

1980

1990
Year

2000

Democracies

Partial democracies

Autocracies
Failed states

Partial autocracies

2010

Occupied countries

Countries in transition

Figure 2 Average number of terrorist attacks per regime type, 1970–2012.

consistent with previous literature, democracies tended to possess the highest number of terrorist attacks until the mid-2000s, when partial democracies,
autocracies, and countries under foreign occupation (Iraq and Afghanistan)
emerged as more common venues of terrorist attacks.
Of course, the global distribution of regime types has varied considerably
over time. As such, we normalized the absolute counts of terrorist attacks
by dividing the total number of attacks in each category by the total number
of countries in each regime type category to examine the average number of
terrorist attacks per regime category per year (Figure 2).
Figure 2 shows some striking differences. The average democracy experienced a stable number of terrorist attacks throughout the series, but it is
not the most explosive category. Instead, during the early 1980s, countries in
transition (e.g., El Salvador) and failed states (e.g., Lebanon) experienced the
highest number of terrorist attacks. And beginning in 2003, occupied countries (e.g., Iraq and Afghanistan) experienced the highest average number of
annual attacks per country.
Importantly, however, these trends are driven by small numbers of cases,
as there are very few countries in transition, failed states, or countries under
occupation in a given year. As such, the high number of average attack counts

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Table 1
Bivariate Association Between Polity Score and Terrorist Attacks
(Negative Binomial Regression)a
Model 1 (1970–2012) Model 2 (1970–2009) Model 3 (2010–2012)
DV: Number of
DV: Number of
DV: Number of Terrorist
Terrorist Attacks
terrorist attacks
Attacks
Polity2
Population (logged)
Constant
N
Wald chi2
Prob > chi2

0.117*** (0.019)
0.654*** (0.100)
−15.663*** (3.803)
6473
68.89
0.0000

0.082*** (0.022)
0.610*** (0.105)
−7.629*** (1.761)
5984
61.14
0.0000

0.049 (0.047)
1.096*** (0.232)
−15.66*** (3.803)
489
24.37
0.000

a Estimates produced using a zero-inflated event count did not significantly alter these results. Moreover,
a vuong test reveals that a zero-inflated negative binominal does not significantly improve the model over
a negative binomial regression in Model 3. For simplicity, the NBR is presented here; the ZINB commands
and results are available in the replication file, which can be accessed at www.ericachenoweth.com.
***p < 0.001.

may be driven by one or two cases in each category and should be taken with
a grain of salt.2
That said, these shifting distributions over time do suggest some possible temporal effects in the most common terrorist attack venue. Once
we exclude key outlying states—such as El Salvador, Lebanon, Iraq, and
Afghanistan—we may be seeing a shift in the types of regime that experience
the most terrorist attacks.
We therefore estimated a negative binomial regression on the association
between Polity score and terrorist attacks in general (controlling for logged
population from Banks, 2013), and then stratified the results by time period
(Table 1).
As with prior literature, we find that there is a strong, positive association
between the Polity score and the number of terrorist attacks (Model 1). We
find, however, that this positive association is conditional on whether the
data series takes place before 2010. Through 2009, there was a strong positive
association (Model 2), with both statistical and substantive findings nearly
identical to those produced in Model 1. However, from 2010 to 2012, the statistical significance of that association disappears, suggesting that in those
years, the effects of Polity score on terrorist attacks are not different from 0
(Model 3).
In Figure 3, we identify the predicted counts of terrorist attacks from Models 1 through 3, which we calculated using the prtab command in Stata.
2. On the basis of Figure 2, it is easier to understand policymakers’ preoccupations with weak and
failed states as the most vulnerable to terrorism.

Regime Type and Terrorist Attacks

9

60

50

40
Entire series (1970–2012)

30

1970–2009
2010–2012

20

10

0

−10 −8

−6

−4

−2

0

2

4

6

8

10

Figure 3 Predicted number of annual terrorist attacks by regime type, 1970–2012.

Several important findings are clear. First, overall terrorist events have
increased during 2010–2012 worldwide and in every regime type. On
average, however, controlling for population size, the predicted number of
annual terrorist attacks in full democracies declined to about 13 in the period
2010–2012 from about 18 in previous years. Second, although the time period
2010–2012 still shows a positive association between regime type and terrorist attacks, it is notable that countries with the lowest Polity scores—between
−10 and −5—have experienced a substantive increase in the number of
terrorist attacks they experienced during the pre-2010 period. Although it is
likely too early to speculate whether these trends will continue, this finding
suggests an important secular departure from previous trends.
WHICH NONDEMOCRACIES WILL SEE MORE TERRORISM?
SOME CLUES FROM THE LITERATURE
As Crenshaw notes, regimes that deny access to power and persecute dissenters create dissatisfaction (Crenshaw, 1981, p. 383). Previous research suggests that whether autocracies experience terrorism depends on their ability
to coerce and co-opt regime challenges (Wilson & Piazza, 2013, p. 942). Much
of the literature therefore concentrates on two types of coercive and co-optive
regime attributes—institutions and audience costs—which are highly variable across authoritarian regimes.
The different forms autocracies take matter when it comes to their ability to
coerce and co-opt opponents. For example, drawing on Geddes’ categories

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

of political dictatorship, Wilson and Piazza find that single-party autocracies
are better able to employ both coercion and co-optation to address political
opposition (Wilson & Piazza, 2013, p. 953). Single-party regimes can coerce
opponents through their control over the military, but they can also co-opt
opponents through their influence on institutions such as legislatures, political parties, and courts (Wilson & Piazza, 2013, p. 945). On the other hand,
democracies must try to co-opt terrorists because their ability to coerce terrorists is limited by guarantees of civil, political, and human rights (Wilson
& Piazza, 2013, p. 945). Military regimes are limited to coercion because they
are less capable of co-opting opponents and need to rely on coercive power to
maintain control (Wilson & Piazza, 2013, p. 945). Thus, those autocracies that
are able to complement repression with co-optive institutions and to deter
challenges while accommodating others are less prone to terrorism (Wilson
& Piazza, 2013, p. 952).
Others find that autocracies use formal institutions as mechanisms
through which to reduce the risk of terrorism. Aksoy, Carter, and Wright
(2012) explore how political party activity in dictatorships helps to mobilize
discontented opposition groups. Specifically, the ability of opposition groups
to participate in legislatures and voice their grievances may help to mitigate
the possibility these groups will pursue violence. Repression is a costly
and risky strategy for dictators, which is why they seek to purchase loyalty
to the regime by co-opting potential challengers. By co-opting potential
challengers, dictatorships bring their opponents into the political system
allowing them to gain some concessions from the regime (ibid, 2012). Once
opponents have been co-opted, through formalized roles in the legislature,
Aksoy et al. (2012) note they will be less likely to risk challenging the regime
for fear of losing their share of the regime’s spoils. However, the presence of
opposition political parties without a functional legislature makes regimes
more likely targets of terrorism, further supporting Wilson and Piazza’s
hypothesis on the fundamental importance of co-optation as a tool of
authoritarian management of potential terrorist groups (ibid, 2012). In fact,
Aksoy et al.’s (2012, p. 818) data reveal that dictatorships with opposition
parties and no legislatures experience a nearly 60% higher probability of
emergence of violent groups. As long as autocracies can use institutions such
as legislatures and elections to successfully co-opt potential challengers,
Aksoy et al. argue, they need not resort to repression.
Dictators also have to be concerned with the potential audience costs in
the event of terrorist attacks. Audience costs are the domestic punishment
that leaders face for backing down from public threats (Fearon, 1994, pp.
577–592). Autocratic leaders face accountability threats not from the public,
but rather from the domestic, regime elites on whom they rely for support. It
is these elites that can act like voting publics in democracies by coordinating

Regime Type and Terrorist Attacks

11

to punish an authoritarian leader (Weeks, 2008, p. 36). If dictators are unable
to respond to the challenge of terrorist actions, regime elites may hold the
leader accountable by replacing him. Aksoy, Carter, and Wright (2013, pp.
4–5) find strong evidence that increasing political violence is associated with
a higher risk of a reshuffling coup (where an autocratic leader is replaced but
not the entire regime). C. R. Conrad, J. Conrad, and Young (2014) suggest
that these authoritarian audience costs may increase incentives to use terrorism against authoritarians. Because terrorists want their actions to have as
much of an impact as possible, there are greater incentives to launch terrorist
actions against states with high audience costs.
What, then, are the key takeaways from this literature in further understanding terrorism in autocracies? The first trend would be to look more
closely at state behavior itself (Chenoweth, 2013, p. 375). Wilson and Piazza’s
article is a step in this direction, along with Aksoy et al.’s suggestions to
attempt to understand the interaction of opposition groups and dictators
(2012, p. 823) and how terrorism influences incumbent dictators (2013, p. 6).
The growing research on audience costs is a trend that also opens promising research into future understanding of trends between terrorism and
autocracy. As Weeks notes, the literature on accountability remains narrow
and much remains to be learned about domestic politics in nondemocracies
(Weeks, 2008, pp. 59–60). Autocracies are not as monolithic as they were
once understood to be, and perhaps widespread acceptance of this reality
has made them more likely targets of terrorist attacks.
THE FUTURE OF REGIME TYPE AND TERRORIST ATTACKS
While we are not certain whether the autocracy-terrorism trend will
continue, we can draw several conclusions from recent trends. First, the
emergence of semi-autocratic systems and “hybrid regimes” is likely to
complicate previous assumptions about the reasons why democracies were
so susceptible to terrorism (Levitsky & Way, 2011). Indeed, many of the
mechanisms, which thought to make democracies more vulnerable—such as
political competition, mobilization, and transition—are now quite common
to many autocracies.
Second, we do see variation in the types of autocracies and transitioning
countries that experience terrorism. As such, a promising line of research
is attempting to explain this variation. Future research that focuses on the
behavioral aspects of autocracy is likely to yield new insights into the relationship between these regimes and terrorist groups.
Finally, it is clear that countries under military occupation—including Iraq
and Afghanistan—have accounted for much of the terrorist violence in recent
years. As such, an important development to note is how the United States

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

and NATO withdrawal from Afghanistan—and the Afghan government’s
ability to consolidate power—will affect terrorism there in the future. However, recent trends do not paint an optimistic picture. If these trends continue,
then neither a slide toward autocracy nor a consolidation of democracy is
likely to reduce terrorist violence in Afghanistan. Indeed, if recent trends persist or accelerate, the regime type of the host country may have little or no
effect on prospects for future violence.
REFERENCES
Abadie, A. (2006). Poverty, political freedom, and the roots of terrorism. American
Economic Review, 96(2), 50–56.
Aksoy, D. (2012). Elections and the timing of terrorist attacks in democracies. Working paper, Department of Politics, Princeton University.
Aksoy, D., & Carter, D. B. (2014). Electoral institutions and the emergence of terrorist
groups. British Journal of Political Science, 44(1), 181–204.
Aksoy, D., Carter, D. B., & Wright, J. (2012). Terrorism in dictatorships. Journal of
Politics, 74(3), 810–826.
Aksoy, D., Carter, D. B., & Wright, J. (2013). Terrorism and the fate of dictators. Working paper, Princeton University.
Banks, A. (2013). Cross-national time series data archive. Jerusalem, Israel: Databanks
International.
Bloom, M. (2004). Palestinian suicide bombing: Public support, market share, and
outbidding. Political Science Quarterly, 119(1), 61–88.
Braithwaite, A., & Sobek, D. A. (2005). Victim of success: American dominance and
international terrorism. Conflict Management and Peace Science, 22(2), 135–148.
Burgoon, B. (2006). On welfare and terror: Social welfare policies and politicaleconomic roots of terrorism. Journal of Conflict Resolution, 50(2), 176–203.
Chenoweth, E. (2010). Democratic competition and terrorist activity. The Journal of
Politics, 72(1), 16–30.
Chenoweth, E. (2012). Is terrorism still a democratic phenomenon? The Journal of
International Relations, 32(8), 85–100.
Chenoweth, E. (2013). Terrorism and democracy. Annual Review of Political Science,
16, 85–100.
Choi, S. W. (2010). Fighting terrorism through the rule of law? Journal of Conflict Resolution, 54(6), 940–966.
Conrad, C. R., Conrad, J., & Young, J. K. (2014). Tyrants and terrorism: Why some
autocrats are terrorized while others are not. International Studies Quarterly, 58(3),
539–549.
Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399.
Danzell, O. E. (2011). Political parties: When do they turn to terror? Journal of Conflict
Resolution, 55(1), 85–105.
Drakos, K., & Gofas, A. (2006). The devil you know but are afraid to face: Underreporting bias and its distorting effects on the study of terrorism. Journal of Conflict
Resolution, 50(5), 714–735.

Regime Type and Terrorist Attacks

13

Drakos, K., & Gofas, G. (2007). Evidence for the existence of under-reporting bias
in observed terrorist activity: The message in press freedom status transitions.
Democracy and Security, 3, 139–155.
Eubank, W. L., & Weinberg, L. (1994). Does democracy encourage terrorism? Terrorism and Political Violence, 6(4), 417–463.
Eubank, W. L., & Weinberg, L. (1998). Terrorism and democracy: What recent events
disclose. Terrorism and Political Violence, 10(1), 108–118.
Eubank, W. L., & Weinberg, L. (2001). Terrorism and democracy: Perpetrators and
victims. Terrorism and Political Violence, 13(1), 155–164.
Eyerman, J. (1998). Terrorism and democratic states: Soft targets or accessible systems. International Interactions, 24(2), 151–170.
Fearon, J. D. (1994). Domestic political audiences and the escalation of international
disputes. The American Political Science Review, 88(3), 577–592.
Gadarian, S. K. (2010). The politics of threat: How terrorism shapes foreign policy
attitudes. The Journal of Politics, 72(2), 469–483.
Goldstone, J. A., Bates, R. H., Epstein, D. L., Gurr, T. R., Lustik, M. B., Marshall,
M. G., … , Woodward, M. (2010). A global model for forecasting political instability. American Journal of Political Science, 54(1),190–208.
Hoffman, B. (2006). Inside terrorism (2nd ed.). New York, NY: Columbia University
Press.
Koch, M. T., & Cramner, S. (2007). Testing the ‘Dick Cheney’ hypothesis: Do governments of the left attract more terrorism than governments of the right? Conflict
Management and Peace Science, 4, 311–326.
Kurrild-Klitgaard, P., Justesen, M. K., & Klemmensen, R. (2006). The political economy of freedom, democracy, and transnational terrorism. Public Choice, 128(1),
289–315.
LaFree, G., & Dugan, L. K. (2007). Introducing the Global Terrorism Database. Terrorism and Political Violence, 19, 181–204.
Levitsky, S., & Way, L. A. (2011). Competitive authoritarianism: Hybrid regimes after the
Cold War. New York, NY: Cambridge University Press.
Li, Q. (2005). Does democracy promote or reduce transnational terrorist incidents?
Journal of Conflict Resolution, 2, 278–297.
Marshall, M. G., Jaggers, K., & Gurr, T. R. (2013). POLITY IV Project: Political regime
characteristics and transitions, 1800–2012 Dataset User’s Manual. Retrieved from
http://www.systemicpeace.org/polity/polity4.htm
Moore, W.H, Bakker, R., & Hill, D. W. (2011). How Much Terror? Dissidents, governments, institutions, and the cross-national study of terror attacks. Working paper,
Department of Political Science, Florida State University.
National Consortium for the Study of Terrorism and Responses to Terrorism
(START). (2011). The Global Terrorism Database. College Park, MD. Retrieved from
http://www.start.umd.edu/gtd/
Pape, R. A. (2003). The strategic logic of suicide terrorism. American Political Science
Review, 97(3), 343–361.
Piazza, J. A. (2009). Economic development, unresolved political conflict, and terrorism in India. Studies in Conflict & Terrorism, 32, 406–419.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Piazza, J. A. (2011). Poverty, minority economic discrimination and domestic terrorism. Journal of Peace Research, 48(3), 339–353.
Piazza, J. A. (2012). Regime age and terrorism: Are new democracies prone to terrorism? Presented at Annual Meeting of the International Studies Association, 53rd,
San Diego, 1–4 April.
Piazza, J. A., & Walsh, J. L. (2010a). Why respecting physical integrity rights reduces
terrorism. Comparative Political Studies, 43(5), 551–577.
Piazza, J. A., & Walsh, J. L. (2010b). Physical integrity rights and terrorism. PS: Political Science and Politics, 43(4), 411–414.
Sanchez-Cuenca, I., & Aguilar, P. (2009). Terrorist violence and popular mobilization:
The case of the Spanish transition to democracy. Politics & Society, 37(3), 428–453.
Sandler, T. (1995). On the relationship between democracy and terrorism. Terrorism
and Political Violence, 7(4), 1–9.
Savun, B., & Phillips, B. (2009). Democracy, foreign policy, and terrorism. The Journal
of Conflict Resolution, 53(6), 878–904.
Schmid, A. (1992). Terrorism and democracy. Terrorism and Political Violence, 4(4),
14–25.
Vreeland, J. R. (2008). The effect of political regime on civil war: Unpacking anocracy.
Journal of Conflict Resolution, 52(3), 401–425.
Weeks, J. L. (2008). Autocratic audience costs: Regime type and signaling resolve.
International Organization, 62(1), 35–64.
Wilson, M. C., & Piazza, J. A. (2013). Autocracies and terrorism: Conditioning effects
of authoritarian regime type on terrorist attacks. American Journal of Political Science, 57(4), 941–955.
Young, J. K., & Dugan, L. K. (2011). Veto players and terror. Journal of Peace Research,
48(1), 19–33.

FURTHER READING
Askoy, D., & Carter, D. B. (2014). Electoral institutions and the emergence of terrorist
groups. British Journal of Political Science, 44(1), 181–204.
Aksoy, D., Carter, D. B., & Wright, J. (2012). Terrorism in dictatorships. Journal of
Politics, 74(3), 810–826.
Chenoweth, E. (2013). Terrorism and democracy. Annual Review of Political Science,
16, 85–100.
Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399.
Wilson, M. C., & Piazza, J. A. (2013). Autocracies and terrorism: Conditioning effects
of authoritarian regime type on terrorist attacks. American Journal of Political Science, 57(4), 941–955.

KARA KINGMA SHORT BIOGRAPHY
Kara Kingma is a PhD student in International Studies at the Josef Korbel
School of International Studies. She focuses on the fields of Comparative Politics and International Political Economy.

Regime Type and Terrorist Attacks

15

BRYAN CRAMER SHORT BIOGRAPHY
Bryan Cramer is a Project Assistant at the National Democratic Institute and
a graduate of the MA program at the Josef Korbel School of International
Studies.
ERICA CHENOWETH SHORT BIOGRAPHY
Erica Chenoweth is an Associate Professor at the Josef Korbel School of International Studies, University of Denver, and an Associate Senior Researcher
at the Peace Research Institute Oslo (PRIO).
RELATED ESSAYS
Varieties of Capitalism (Political Science), Peter A. Hall
Emerging Trends in Social Network Analysis of Terrorism and Counterterrorism (Sociology), David Knoke
Why Do States Sign Alliances? (Political Science), Brett Ashley Leeds
Participatory Governance (Political Science), Stephanie L. McNulty and Brian
Wampler
Intervention and Regime Change (Political Science), John M. Owen IV and
Roger G. Herbert Jr.
War and Social Movements (Political Science), Sidney Tarrow
Trends in the Analysis of Interstate Rivalries (Political Science), William R.
Thompson