Income and Terrorism: Insights From Subnational Data

Abstract

This paper first introduces a theoretical formalization connecting a polity’s income level to terrorism. Our framework can accommodate different underlying assumptions about individual- and society-level grievances, yielding competing hypotheses. We then construct a panel database to study terrorism for 1527 subnational regions in 75 countries between 1970 and 2014. Results consistently imply an inverted U-shape that remains robust to incorporating a comprehensive set of region-level covariates, region- and time-fixed effects, as well as estimating an array of alternative specifications. The threat of terrorism systematically rises as low-income polities become richer, peaking at GDP/capita levels of ≈ US$12,800 (in constant 2005 PPP US$), but then falls consistently above that level. This pattern emerges for domestic and transnational terrorism alike. While peaks differ by perpetrator ideology, the inverted U shape also prevails across ideology-specific subsamples. In sum, alleviating poverty may first exacerbate terrorism, contrary to much of the proposed recipes advocated since 9/11.

Keywords

subnational income subnational terrorism domestic terrorism transnational terrorism

“We won’t win the war against terror without addressing the problem of poverty.” (Wolfensohn, then-President of the World Bank, 2002).

Introduction

In the aftermath of the 9/11 attacks, US President George W. Bush, US Secretary of State John Kerry, British Prime Minister Tony Blair, along with other prominent politicians, policymakers, and commentators explicitly linked terrorism to poverty (Bush 2002; Easterly 2016; Krueger 2007; Sterman 2015). However, cross-country research has produced ambiguous and sometimes contradictory evidence for a potentially negative relationship between income and terrorism (Abadie 2006; Azam and Thelen 2008; Walsh and Piazza 2010).¹ Yet, despite the inconclusive evidence, the hypothesis linking poverty to terrorism remains one of the most common myths related to terrorism (Gaibulloev and Sandler 2022).

This paper makes two contributions. First, we introduce a basic theoretical framework to provide a structured way of thinking about potential links between a polity’s income level and the degree of terrorism it experiences. Our assumptions are informed by the existing descriptive evidence that has motivated prior studies. In particular, we aim to understand at which combination of GDP/capita and grievances (broadly defined) at the personal and societal level an individual turns to terrorism. Our framework is able to accommodate different perspectives of the existing literature and produces competing hypotheses about how income levels inform a polity’s degree of terrorism. The main propositions suggest an inverted U-shape under certain assumptions (see also Freytag et al., 2011; De la Calle and Sánchez-Cuenca, 2012), while the poverty-highlighting ‘Bush Hypothesis’ emerges under other assumptions.² Finally, our theoretical framework suggests society-level grievances moderate the peak of the income-terrorism link.

Second, in order to test these hypotheses, we construct a novel database of terrorism at the subnational level. We suggest that our understanding of the income-terrorism nexus sharpens substantially once we zoom in to the subnational level, i.e., studying Balochistan, California, Catalonia, and Île-de-France instead of Pakistan, the United States, Spain, and France. Two basic stylized facts motivate this refocus. First, terror attacks often cluster regionally within a country, rather than being spread out uniformly. For example, in the United Kingdom from 1970 to 2014, we identify striking differences between Northern Ireland (1544 attacks) and the North (four attacks). Similarly, while the Chilean O’Higgins region was completely spared of terror attacks over that entire time period, the metropolitan region of Santiago suffered 1612 attacks. And second, income levels across regions within a country often differ more than incomes across countries. For example, the average income of Moscow exceeds the average income of Sicily, even though Italy is on average approximately three times richer than Russia. Such substantial within-country heterogeneities – both for the degree of terrorism and for income levels – are lost when studying country-level aggregates.

Our empirical approach matches subnational (regional) data on GDP/capita (from Gennaioli et al. 2014) with subnational data on terror attacks (from START 2017b) for 1527 regions in 75 countries between 1970 and 2014. These sample countries are statistically representative of the global relationship between income and terrorism. Our unit of analysis constitutes the second-largest administrative unit in the respective nation, i.e., a federal state, county, or province, depending on the country. Our main specifications hold constant potential confounders associated with (i) population size, (ii) regions hosting a country’s capital city, (iii) oil production, as well as (iv) region- and period-fixed effects. Region-fixed effects prove particularly powerful in accounting for unobservable time-invariant differences across regions, such as geographical attributes that often correlate with terrorist activity (e.g., mountainous terrain or ruggedness) and unique histories of ethnic and religious conflict or colonization experiences. These fixed effects also reasonably control for those environmental and societal aspects that only change slowly over time within a region, such as fractionalization and polarization along ethnic or religious dimensions.

The empirical results lend firm support to our theoretical proposition suggesting an inverted U-shaped pattern between income levels and terrorism. This finding is consistent with the cross-country evidence documented by Freytag et al. (2011), De la Calle and Sánchez-Cuenca, 2012, and Enders and Hoover (2012) who posit low-income polities lack the resources and potential gains to invite terrorism, while high-income polities can afford effective counterterrorism measures and offer substantial opportunity costs to prospective terrorists. Our findings suggest that, as incomes in poor regions increase, terrorism becomes substantially more likely until an estimated peak of approximately US$12,800 (in constant 2005 PPP US$). In our sample, 63 percent of all observations would fall below that threshold. Above that, higher income levels would be associated with a decline in terrorism. Importantly, we find this inverted U-shaped pattern for domestic and transnational terrorism alike.

Finally, we consider potential heterogeneity along perpetrator ideology. Illustrating the generality of our findings, the inverted U-shape independently emerges for all identifiable ideologies with (i) Islamist, (ii) left-wing, (iii) right-wing, (iv) separatist, and (v) other religious groups. The consistency with which this pattern emerges across regions around the world for over 45 years suggests a systematic inverted U-shape link between income and terrorism that transcends time, ideology, and space. Interestingly, religious terrorism peaks at income levels that are lower than those for left-or right-wing terrorism – a relationship that was proposed by Enders et al. (2016) but, to our knowledge, has remained untested since.

Overall, our study contributes to a wider understanding of terrorism determinants, while particularly informing the debate on the link between income and terrorism (e.g., see Gaibulloev and Sandler 2022). Beyond terrorism, this paper also informs the literature on how GDP/capita can affect non-economic variables, as well as the benefits and costs associated with that development process (e.g., see Bloom and Canning 2000; Gürlük 2009; Friedman 2010). Our results confirm prior findings in suggesting raising income levels in poor polities might be accompanied by elevated chances of experiencing terrorism. Finally, this paper introduces a panel dataset connecting income and terrorism at the subnational level, which we hope encourages future research into terrorism-related dynamics at a disaggregated level.

Related Literature

The political and scholarly debate that followed 9/11 inextricably linked poverty to terrorism (Haggar 2021; Odede 2015; Pilgrim 2015). The underlying hypothesis is grounded in existing work on civil conflict (Abadie 2006), civil war (Collier and Hoeffler 2004; Miguel et al. 2004), and political coups (Alesina et al. 1996). As another form of political violence, terrorism has been suggested to follow a similar logic: Poverty is accompanied by grievances that can motivate terrorism (Piazza 2007).

Nevertheless, two decades after 9/11, the corresponding empirical evidence remains inconclusive. Cross-country studies have produced negative, positive, and null results – an artefact we illustrate in Table 1. Similarly, individual-level studies have failed to establish a systematic correlation between poverty and terrorism (Benmelech et al. 2012; Berrebi 2007; Hassan 2002; Krueger and Malečková 2003; Sageman 2004).

Table 1.

Overview of the Quantitative Literature Linking GDP/Capita to Terrorism (Based on Gosling 2017).

Statistically Insignificant	Statistically Significant Negative	Statistically Significant Positive	Non-Linear, Inverted U-Shape
Abadie (2006)	Azam and Delacroix (2006)	Blomberg and Hess (2008b,a).^a	Freytag et al. (2011)
Basuchoudhary and Shughart (2010)	Azam and Thelen (2008)	Blomberg and Rosendorff (2006).^b	De la Calle and Sánchez-Cuenca, 2012
Berman and Laitin (2008)	Blomberg and Hess (2008b).^a	Burgoon (2006).^c	Enders and Hoover (2012).^d
Campos and Gassebner (2013)	Braithwaite and Li (2007)	Eyerman (1998)	Enders et al. (2016).^e
Crenshaw et al. (2007)	Bravo and Dias (2006).^c	Koch and Cranmer (2007).^c
Dreher and Fischer (2010, 2011)	Li (2005)	Kurrild-Klitgaard et al. (2006)
Gassebner and Luechinger (2011)	Li and Schaub (2004)	Neumayer and Plümper (2009).^f
Goldstein (2005).^c	Testas (2004)	Piazza (2007, 2008a, 2011).^c
Krueger and Laitin (2008)		Plümper and Neumayer (2010)
Krueger and Malečková (2003)		Walsh and Piazza (2010)
Piazza (2006, 2008b).^c		Tavares (2004)
Sambanis (2008)

^aBlomberg and Hess (2008b) find a negative (positive) association with ‘low (lower) income’ countries.

^bBlomberg and Rosendorff (2006) find a positive (negative) association between income of the host (source) country and the terrorism in the host country.

^cGDP/capita constitutes one component of a composite indicator, such as the Human Development Index or the Government Capability Index.

^dEnders and Hoover (2012) further delineate between nonlinearities among low- and high-income country-year observations.

^eEnders et al. (2016) employ nonlinear smooth transition regressions.

^fNeumayer and Plümper (2009) find a positive association between terrorism and the distance of GDP/capita between the target and the source country.

Theoretically, the inconclusive link between income and terrorism may be owed to an incomplete functional form that conceals nonlinearities (Enders and Hoover 2012; Enders et al. 2016). While low-income polities do not offer sufficient human and monetary resources to support terrorism, high-income societies may be able to employ effective counterterrorism strategies (Enders et al. 2016; Lai 2007). From a sociological perspective, Maslow’s (1943) hierarchy of needs implies political and societal prospects only gain relevance once basic physiological needs are met. Thus, ideological and political considerations may not constitute primary objectives in impoverished societies, i.e., political violence in the form of terrorism could play less of a role. At the other end of the income spectrum, economic grievances are less likely to arise in richer countries where governments can usually leverage substantial funds to address concerns of their citizenry (Lai 2007).

Consequently, ceteris paribus, terrorism, whether domestic or transnational, may peak at medium incomes. A handful of cross-country studies support this perspective (De la Calle and Sánchez-Cuenca 2012; Freytag et al. 2011; Lai 2007). Further, Enders et al. (2016) suggest the peak of terrorism may have changed over time, owing to the shift from left-wing ideologies, which were concentrated in relatively wealthy countries, to religious fundamentalists that predominantly live in the developing world. What has remained elusive, aside from consistent empirical evidence, is a basic theoretical framework to formalize the income-terrorism relationship.

Theoretical Framework

In the following, we formulate a simple framework relating GDP/capita to terrorism. Our framework should be viewed as one potential theoretical construct to explore how, and under which assumptions, terrorism can respond to changes in income.

Basic Assumptions and Utilities

Suppose polity j ∈ {1, …, N} is composed of n individuals. Each individual i ∈ {1, …, n} holds views $π_{i} \in R$ , such that a positive and large π_i indicates stronger anti-establishment views. For example, a large π_i may be indicative of i’s family members having been persecuted, individual land or assets being seized, or any other form of grievances that are specific to i’s experience with the regime. Alternatively, an individual’s views could be shaped by their own political or religious beliefs and to what extent they align or not with the current regime (e.g., the desire to implement one’s preferred interpretation of religious tenets).

In addition to i’s individual views, there exist polity-level perspectives pertaining to the regime $p_{j} \in R_{+}$ , which reflect society-level grievances and discontent with the status quo. This may include, but is not limited to, persecutions of the entire society along ethnic or religious dimensions and, more generally, grievances against the regime that are shared by the entire populace.³ As another example, one may think about relatively homogeneous ethnic or religious minorities that seek political and economic independence of their region from a national state, i.e., separatist groups.

Every individual i chooses whether to engage in terrorism or join the formal labor market. For simplicity, assume working in the labor market yields income y_j (with y_j ≥ 0), i.e., GDP/capita of polity j, and the individual draws utility

U_{w o r k} = y_{j} .

(1)

The agent compares U_work to the utility from engaging in terrorism, which is described by

U_{t e r r o r} = π_{i} + p_{j} f (y_{j}) .

(2)

U_terror is characterized by two components. First, utility increases with i’s personal views towards the status quo – terrorism renders more satisfaction if individual grievances are larger.⁴ Second, polity-level views, p_j, are moderated by a general function f(y_j) that relates income to terrorist activities. We will describe f(y_j) in more detail; for now, equation (2) only posits a simple connection between the utility drawn from terrorism and society-level grievances, combined with GDP/capita.

Terrorism in Society

Individual i engages in terrorism if U_terror ≥ U_work, i.e., if

π_{i} + p_{j} f (y_{j}) \geq y_{j} .

(3)

The share of terrorists in polity j then becomes a function of income with

T (y_{j}) = \frac{\sum I_{i} (y_{j})}{n} where I_{i} = {\begin{array}{c} 1 & if π_{i} + p_{j} f (y_{j}) \geq y_{j} \\ 0 & otherwise . \end{array}

(4)

Throughout our theoretical illustration, we posit that a higher share of terrorists in society translates to a higher degree of terrorism. Naturally, this ignores degrees of terrorists (e.g., some individuals may become influential leaders), but it simplifies the analysis and allows for basic deductions pertaining to the role of income.⁵

Equation (4) clearly allows for two basic deductions. First, terrorism is more likely if individuals hold more pronounced anti-regime views, π_i. Second, polity-level grievances or discontent with the government, p_j, can, under certain assumptions about f(y_j), affect both the degree of terrorism and the role of income in explaining terrorism.

Linking Income and Terrorism

For the role of income, equations (3) and (4) highlight the importance of f(y_j). While the exact characteristics of f(y_j) remain an empirical question, this structure allows us to track how prior observations in the literature translate to the income-terrorism nexus.

Scenario I: The inverted U shape

For the first scenario, assume returns from terrorism increase as the polity becomes richer, i.e., f ′(y_j) > 0. Importantly, this does not necessarily imply the degree of terrorism increases with income since both the left- and right-hand side of equation (3) rise with y_j. This assumption of f ′(y_j) > 0 is consistent with the idea that higher incomes translate to (i) a higher prize to be seized by terrorist activities (e.g., see Grossman 1991) and/or (ii) greater surplus available for terrorist activities (Enders and Hoover 2012). Note, too, that f ′(y_j) > 0 is implicitly the assumption made by the contributions in the literature suggesting a positive relationship between terrorism and income (see, e.g., Kurrild-Klitgaard et al. 2006; Piazza, 2006).

Further, assume the returns to engaging in terrorism increase at a decreasing rate with income such that f″(y_j) < 0. This assumption is consistent with the arguments proposed by Lai (2007), Enders and Hoover (2012), and Enders et al. (2016), whereby richer countries dedicate disproportionately greater resources to counter-terrorism measures. Finally, for technical reasons, we impose the Inada conditions (Inada 1963; Uzawa 1963) on f(y_j), i.e., f(0) = 0, $\lim_{y_{j} \to \infty} f^{'} (y_{j}) = 0$ , and $\lim_{y_{j} \to 0} f^{'} (y_{j}) = \infty$ .

In this scenario, the left-hand side of equation (3) first rises sharply at low income levels but later flattens out as income levels increase. However, the right-hand side rises linearly with income. Under reasonable assumptions about the distribution of π_i among the populace, these dynamics then generate the inverted U-shaped relationship that has been described and identified for country-level data by Enders and Hoover (2012).⁶

Consider the simple example of f(y_j) = y^α (with 0 < α < 1). Figure 1 illustrates three individual cases: low (π_low), moderate (π_moderate), and high levels (π_high) of individual-level grievances with π_low < π_moderate < π_high.

Figure 1.

Relationship between polity j’s income and the benefits and costs of terrorism.

Now consider the three marked income levels of y_low, y_middle, and y_high. The individual with π_high engages in terrorism at all three income levels – this is the hardcore extremist. In contrast, the individual with π_low never engages in terrorism. However, the individual with π_moderate engages in terrorism for intermediate levels of income but not when their polity is poor or rich. These insights lead us to formulate H_A regarding a non-linear relationship between income and terrorism:

H_A:

As income levels rise in a polity, the degree of terrorism first increases and then decreases after passing a certain threshold level of income, everything else equal.

Scenario II: The ‘bush hypothesis’

Our framework also lends itself to illustrating what we label the ‘Bush Hypothesis’ – the idea that terrorism is inherently a poverty-related issue. To yield that result, we only need to impose f′(y_j) ≤ 1/p_j on f(y_j), i.e., rising income yields lower marginal utility for terrorism than for working in the labor market.⁷ Intuitively, this could be the case if security and counterterrorism operations improve steadily with income, even in poor polities.

From equation (3), it is then straightforward to observe higher incomes are associated with less terrorism, everything else equal, as the left-hand side decreases and the right-hand side increases with y_j. In this context, the stronger assumption of f′(y_j) ≤ 0 also emerges frequently in the associated literature (e.g., see Li and Schaub 2004; Li 2005).

This allows us to formulate an alternative hypothesis with:

H_B:

As income levels rise in a polity, the degree of terrorism consistently decreases, everything else equal.

Grievances and the income-terrorism link

Neither H_A nor H_B pay particular attention to society-wide grievances, p_j. Nevertheless, allowing p_j to change can yield useful insights about the role of polity-level characteristics. Equation (3) shows the link between income and terrorism becomes more pronounced with higher levels of anti-establishment attitudes (i.e., a larger p_j).

We illustrate this in Figure 2, where we return to the inverted U-shaped pattern derived in Scenario I: The inverted U shape. Consider three levels of polity-level grievances, such that p₁ < p₂ < p₃. The benefits from terrorism for a person with individual views π_i will be different for the three levels of p_j. Consequently, the lower p_j, the lower the income at which the net utility from terrorism reaches its maximum.⁸

Figure 2.

Change in the benefits and costs of terrorism in response to changes in polity-level grievances, p_j.

As a consequence, we derive our third and final hypothesis related to polity-level grievances:

H_C:

The relationship between income and terrorism exhibits heterogeneity along the level of societal grievances. If the income-terrorism nexus follows an inverted U shape, then terrorism peaks at a higher income level if anti-regime attitudes are more pronounced.

Our theoretical framework allows for additional scenarios (e.g., considering f ′(y_j) > 0) and extensions, but we restrict ourselves to the most prominent scenarios here.

Data

Subnational Income Levels

With these theoretical hypotheses in mind, we now turn to our empirical exploration. We derive data on region-level income from Gennaioli et al. (2014) who record real GDP/capita (in constant 2005 PPP US$) in 5-year intervals for subnational units in a global sample.⁹ As comprehensive data on terrorism start in 1970, we consider observations from 1970 to 2010, producing a maximum of nine observations per region and an average of six observations per region. Table 2 documents summary statistics of all variables in our main analysis, while Table A1 summarizes the variables used in additional analyses. Table A2 shows full data coverage for each country and year.

Table 2.

Summary Statistics for main variables at the subnational (regional) level for 1527 regions (n = 8383 for all variables). Variables in Panel A come from Gennaioli et al. (2014), while variables in Panel B come from START (2017b).

Variable	Mean	Min.	Description
Variable	(Std. Dev.)	(Max.)	Description
Panel A: Independent variables
Ln(GDP/capita)_i,t	8.85 (1.17)	5.24 (12.02)	GDP/capita in 2005 PPP US$ (we apply the natural logarithm)
Ln(Population size)_i,t	13.77 (1.41)	9.21 (19.09)	Population (we apply the natural logarithm)
Capital_i	0.05 (0.22)	0 (1)	=1 if hosts country’s capital
Oil_i,t	10.26 (20.56)	0 (89.38)	Cumulative oil and gas production (per capita; we apply the natural logarithm)
Panel B: Dependent variables
Terror attacks_i,t,…,t+4	7.40 (46.50)	0 (1479)	# of terror attacks in t, .., t + 4
Domestic attacks_i,t,…,t+4	6.27 (43.15)	0 (1461)	# of non-transnational terror attacks in t, .., t + 4
Transnational attacks_i,t,…,t+4	1.13 (10.90)	0 (705)	# of transnational terror attacks in t, .., t + 4
Islamist attacks_i,t,…,t+4	0.46 (13.73)	0 (1136)	# of terror attacks by Islamist groups in t, .., t + 4
Leftist attacks_i,t,…,t+4	2.73 (22.74)	0 (987)	# of terror attacks by leftist groups in t, .., t + 4
Rightist attacks_i,t,…,t+4	0.13 (1.21)	0 (44)	# of terror attacks by right-wing groups in t, .., t + 4
Separatist attacks_i,t,…,t+4	1.59 (19.29)	0 (1230)	# of terror attacks by separatist groups in t, .., t + 4
Religious non-Islamist attacks_i,t,…,t+4	0.40 (8.74)	0 (674)	# of terror attacks by religious, non-Islamist groups in t, .., t + 4

Consistent with the literature, we calculate the natural logarithm of GDP/capita (e.g., see Freytag et al. 2011, Enders and Hoover 2012, Enders et al. 2016; Krieger and Meierrieks, 2019). Alternatively, using GDP/capita levels (sans logarithm) produces consistent results (see Table A5). To allow for nonlinearities, we follow Enders and Hoover (2012) to incorporate a squared term of that variable.

Figure 3 visualizes the global coverage of our sample. African regions remain under-represented and notable omissions including Iraq and Afghanistan – two of the countries most affected by terrorism. As such selection issues may threaten the generalizability of our findings, we carefully compare global country-level results for all years with those from studying our sample countries and years. These estimations produce consistent coefficients, suggesting our results are unlikely to suffer from systematic selection issues (see Table A3).

Figure 3.

Regional sample coverage (source: Gennaioli et al., 2014).

Subnational Terrorism

For data on terrorism, we access the well-known Global Terrorism Database (GTD). Studying information on the location of each terror attack allows us to assign each attack to a within-country region. Online Appendix B explains this procedure in detail. We then aggregate attacks over 5-year intervals (following Gaibulloev et al. 2017) and merge that variable with Gennaioli et al.’s (2014) data. For example, GDP/capita for Catalonia in 1970 is matched with terror attacks in Catalonia between 1970 and 1974.

Our main dependent variable measures the number of terror attacks, which constitutes the most commonly employed measure in the literature. Additional estimations distinguish between domestic and transnational attacks.¹⁰ Figure 4 plots GDP/capita against the number of terror attacks. Panel A considers all terrorist attacks, while Panels B and C distinguish between domestic and transnational terrorism. Although these graphs do not incorporate potentially confounding factors yet, they do imply a nonlinear relationship between regional income and terrorism in the form of an inverted U-shape.

Figure 4.

Subnational GDP/capita and terror attacks, displayed by kernel-weighted local polynomial smoothing along with 95 percent confidence intervals. Panel A: GDP/capita and terror attacks, Panel B: GDP/capita and domestic terror attacks, Panel C: GDP/capita and transnational terror attacks.

Potential Confounders

Our estimations include a list of region-level covariates that may independently be associated with terrorism and income levels. Following the literature, we incorporate population size, oil production (to control for resource-curse-related dynamics; see Tavares 2004; Sambanis 2008), and a binary indicator for hosting the nation’s capital (because of a potential concentration of cultural, political, and religious targets).¹¹ As the data on educational attainment feature a number of missing values in our sample period, we do not include that variable in our main regressions. Nevertheless, incorporating educational attainment produces consistent results for that limited sample (see Table A5). Finally, accounting for lagged terror attacks also leaves our main conclusions unchanged (see Table A5).

A major advantage of our subnational data structure comes from combining within-country variation in terrorism and income with the panel dimension of repeated information for each region. Our data allow us to account for region-fixed effects to hold constant time-invariant region-specific particularities. This accounts for prevalent correlates of terrorism, such as geography and terrain, unique historical features pertaining to civil conflict, colonization, and other aspects, as well as long-term cultural, economic, and political artefacts. Year-fixed effects absorb all time-specific global developments that may independently correlate with terrorism.

Empirical Strategy

Main Specification

Our main empirical strategy employs a negative binomial regression model in line with the associated literature because the dependent variable constitutes a non-negative count variable and exhibits overdispersion (Gaibulloev et al. 2017; Piazza 2013; Walsh and Piazza 2010; Young and Dugan 2011; Young and Findley 2011). For region i and year t, we estimate:

A t t a c k s_{i, (t, \dots, t + 4)} = β_{0} + β_{1} Ln {(G D P / c a p i t a)}_{i, t} + β_{2} Ln {(G D P / c a p i t a)}_{i, t}^{2} + X_{i, t} β_{3} + λ_{i} + γ_{t} + δ_{i, t}

(5)

where β₁ and β₂ represent our main coefficients of interest. Note that observations do not overlap, as for each region we employ an observation for, say, 1970–1974, another for 1975–1979, and so on. We begin with a linear form assuming β₂ = 0, which would allow for Hypothesis H_B to emerge. Then, we relax this assumption to allow for nonlinearity in accordance with Hypothesis H_A, Figure 4 empirically, and Enders and Hoover (2012). X _i,t constitutes the matrix of control variables introduced in Potential Confounders; λ_i and γ_t capture region- and year-fixed effects; and δ_i,t represents the usual error term.

Potential Sources of Endogeneity

Endogeneity pertaining to reverse causality and omitted variables remains a threat to identifying causal relationship in the associated literature. First, reverse causality implies regions (or countries) may become poorer because of terrorism. Aggregating the dependent variable over years t to t + 4, while measuring independent variables in year t alleviates such concerns. Alternatively, predicting terrorism in t + 1 until t + 4, thereby not leaving any overlap between the dependent and independent variables, produces consistent results (see Table A5). Acknowledging potential path dependency, we also account for terror attacks in the preceding 5 years and derive consistent results (see Table A5). Further, estimating the reverse relationship, i.e., predicting income levels with terror attacks, while controlling for region-fixed effects, produces a null relationship (p-value of 0.734; results available upon request).¹² In sum, reverse causality is unlikely to pose a systematic threat to the interpretation of our results.

Second, theoretically, omitted variables could still influence both regional income levels and terrorism. We control for a list of notable confounders in our main specifications, and additional robustness tests incorporate educational attainment levels, yielding consistent results (see Table A5). As discussed, region-fixed effects account for any statistical variation in terrorism owed to time-invariant cultural, ethnic, language, or religious heterogeneity at the regional level. For example, cultural heritage, religious denominations or language may differ geographically within a country.

Similarly, geographical characteristics within a country often vary, and any potential association between poverty and terrorism may differ along such dimensions. For instance, Colombia’s more hospitable regions happen to be wealthier (e.g., Cundinamarca with Bogotá or Antioquía with Medellín) than the difficult-to-access rainforest. Region-fixed effects capture such heterogeneity. Further, if a region differs systematically from the country average in, say, the de facto implementation of law and order, region-fixed effects account for these differences. Finally, region-fixed effects also implicitly account for country-fixed effects, i.e., any country-level heterogeneity relevant for terrorism is accounted for, such as historical events or colonial ties. Overall, while we cannot fully exclude the possibility of omitted variables affecting our estimates, our rich list of additional regressors should be able to substantially resolve such concerns.

Nevertheless, it is important to note which factors our analyses are unable to account for. In particular, unobservable aspects that inform terrorism and do change within a region over time can influence our derived coefficients associated with income levels. For example, changes in regional governance, regional ethnic polities, or within-region inequality are only incorporated to the extent that they are correlated with our observables of population size, oil production, hosting the country’s capital, educational attainment, and lagged terror attacks.

Regional Income and Terrorism

Main Results

Table 3 reports our main regression results. Column (1) considers a univariate regression that only employs a linear term of GDP/capita to predict terror attacks. The respective coefficient is negative and statistically significant at the 1 percent level (p-value of 0.000). Conclusions from this specification would support H_B, lending credence to the Bush Hypothesis.

Table 3.

Main Results, Predicting Terror Attacks for Region I in Years T,…,T + 4 in a Negative Binomial Regression Framework.

	(1)	(2)	(3)	(4)	(5)	(6)
					Domestic Terrorism	International Terrorism
Ln(GDP/capita)_i,t	−0.351^∗∗∗ (0.094)	3.131^∗∗ (1.294)	4.690^∗∗∗ (1.287)	3.517^∗∗∗ (0.379)	3.677^∗∗∗ (0.416)	5.472^∗∗∗
${Ln (G D P / c a p i t a)}_{i, t}^{2}$		−0.197^∗∗∗ (0.072)	−0.282^∗∗∗ (0.072)	−0.186^∗∗∗ (0.021)	−0.202^∗∗∗ (0.024)	−0.298^∗∗∗ (0.036)
Control variables^a and time-period-fixed effects			✓	✓	✓	✓
Region-fixed effects				✓	✓	✓
GDP/capita at maximum		2826	4,087	12,763	8969	9713
N ^b	8383	8383	8383	5351	5055	3357
# of regions^b	1527	1527	1526	863	802	517
# of time periods	9	9	9	9	9	9

Notes: Standard errors clustered at the regional level are displayed in parentheses for columns (1)–(3) while columns (4)–(6) report standard errors based on the observed information matrix, using the option vce(oim) in STATA.

^aControl variables include the logarithm of population size, a binary indicator for the location of the capital city, and the natural logarithm of oil produced.

^bThe decline in the number of observations in columns (4)–(6) stems from the introduction of region-fixed effects, where regions with no terror attacks are dropped automatically.

^∗p < 0.10, ^∗∗p < 0.05, ^∗∗∗p < 0.01.

However, upon allowing for nonlinearity in column (2), that conclusion changes, suggesting an inverted U-shape in line with H_A: GDP/capita becomes a positive predictor, while its squared term emerges as a negative predictor (p-values of 0.016 and 0.006). The fourth row from the bottom reports the GDP/capita level at which the income-terrorism relationship is suggested to peak, corresponding to US$2826.

Columns (3) and (4) first add the covariates introduced in equation (5) and time-period-fixed effects, before also accounting for region-fixed effects. The inverted U-shape persists, while the suggested peak rises to US$12,763. This value roughly corresponds to regions such as Quintana Roo (Mexico) in 1980 or Kaliningrad (Russia) in 2010. It is important to highlight that the specification in column (4) exploits within-region variation only, i.e., we only compare the same region to itself at different income levels. Thus, the derived coefficients do not rely on any cross-regional differences, not even within the same country. A corollary of that statistical artefact is that a low-income region is suggested to experience rising likelihoods of terrorism as its own GDP/capita levels increase; but as soon as GDP/capita levels surpass the peak, terrorism diminishes, everything else equal.

Columns (5) and (6) delineate between domestic and transnational terrorism, acknowledging the often-proposed distinction between these types of terrorism and their underlying dynamics (Enders and Hoover 2012; Enders et al. 2016). Results are fully consistent: In both cases, we derive statistical significance at the one percent level for both coefficients of interest, as well as the signs suggested by H_A and our benchmark specification from column (4). Domestic terrorism peaks at lower levels of GDP/capita than transnational terrorism, but the corresponding difference remains small (a conclusion that also emerges from Figure 5). In terms of our theoretical framework, these differences could suggest that polity-level grievances for transnational terrorism are higher than those for domestic terrorism.

Figure 5.

Visualizing regression results from columns (4)–(6) of Table 3. GDP/capita and terrorism, GDP/capita and domestic terror attacks, GDP/capita and transnational terror attacks.

Figure 5 visualizes the suggested relationships from columns (4)–(6). The inverted U-shapes are comparable for domestic and transnational terrorism, which implies a universal nonlinearity of the relationship between income and terrorism, supporting H_A. Interestingly, the slope for transnational terrorism rises more sharply and peaks at a higher income level than the slope for domestic terrorism. Within our theoretical framework, this finding would be consistent with society-level grievances, p_j, being higher for transnational terrorism, everything else equal. One explanation for that hypothesis may relate to the fact that domestic grievances (i) can also be expressed in non-terrorist ways (e.g., voting, protests, or political mobilization) and (ii) are easier to address for the domestic government. However, these options are often unavailable for cross-border grievances.¹³ In our sample, transnational terrorism is indeed more lethal, reaching a maximum of 395 deaths per attack, while domestic terrorism reaches its maximum at 76 deaths per attack. While attack lethality does not necessarily proxy for underlying grievances, these descriptive statistics would be consistent with polity-level grievances, p_j, being more substantial in transnational settings.

Robustness Checks

We conduct a series of alternative specifications to test the validity of these results. In particular, we implement alternative estimation techniques and measures of terrorism by (i) calculating bootstrapped standard errors, (ii) applying Poisson and Ordinary Least Square (OLS) methods, (iii) considering alternative measures of terrorism with attacks per year, terror per capita, a binary indicator for experiencing any terrorism, and deaths from terrorism. Across all these specifications, the inverted U shape prevails with remarkable consistency (see Table A4), providing further support for H_A.

Table A5 documents regression results from (i) considering levels of GDP/capita (i.e., not applying the natural logarithm), (ii) controlling for years of educational attainment at the regional level, (iii) controlling for terror attacks in the past 5 years, (iv) using an alternative time frame for our outcome variable (from t + 1 to t + 4), and (v) considering annual GDP/capita data as reported in Gennaioli et al. (2014) without adjusting observations to conform with our 5-year panel structure. Again, results remain consistent.

Terror Group Ideologies

Finally, we explore the link between income and terrorism for various ideological motivations. Following the common distinctions in the associated literature, we delineate between Islamist, left-wing, right-wing, ethnic/separatist, and religious non-Islamist groups (e.g., see Kis-Katos et al., 2014). The results in Table 4 provide further support for a universal nonlinearity, as the inverted U-shaped pattern emerges across all five group ideologies.¹⁴ These results also prevail when delineating between domestic and transnational terrorism (Tables A6 and A7).

Table 4.

Distinguishing by Group Ideology, Predicting the Number of Terror Attacks for Subnational Region i in Years t,…,t + 4 in a Negative Binomial Regression Framework.

	(1)	(2)	(3)	(4)	(5)
Group Identity:	Islamist	Left-Wing	Right-Wing	Ethnic/Separatist	Religious Non-Islamist
Ln(GDP/capita)_i,t	7.075^∗∗∗ (1.378)	4.829^∗∗∗ (0.661)	10.399^∗∗∗ (1.760)	5.519^∗∗∗ (0.782)	10.039^∗∗∗ (1.687)
${Ln (G D P / c a p i t a)}_{i, t}^{2}$	−0.408^∗∗∗ (0.081)	−0.269^∗∗∗ (0.037)	−0.569^∗∗∗ (0.098)	−0.308^∗∗∗ (0.044)	−0.584^∗∗∗ (0.095)
Control variables^a	✓	✓	✓	✓	✓
Time-period- and region-fixed effects	✓	✓	✓	✓	✓
GDP/capita at maximum	5827	7910	9302	7781	5405
N	841	3060	1456	2036	806
# of regions	145	441	191	306	107
# of time periods	9	9	9	9	9

Notes: Standard errors based on the observed information matrix, using the option vce(oim) in STATA, are displayed in parentheses.

^aControl variables include the logarithm of population size, a binary indicator for the location of the capital city, and the natural logarithm of oil produced.

^∗p < 0.10, ^∗∗p < 0.05, ^∗∗∗p < 0.01.

Notably, the corresponding peaks differ in terms of magnitude, although moderately. This finding supports the theoretical proposition that peaks in terrorism can differ by perpetrator ideology (e.g., Enders et al. 2016): The peak of terrorism associated with Islamist and other religious ideologies occur at income levels that are lower than those for left-wing or right-wing ideologies. In light of our theoretical framework, this finding is consistent with the notion that polity-level grievances may be stronger in regions hosting left- or right-wing terrorism than in regions hosting Islamist terrorists.

This result is in line with evidence from the Profiles of Individual Radicalization in the United States database (PIRUS; START 2017b). Profiles of Individual Radicalization in the United State categorizes radical beliefs in an ascending order of intensity, while providing information on the ideological views of radical individuals.¹⁵ A careful inspection of the data shows that for the individuals for whom such information is available 64 percent of Islamists hold strong radical views, compared to 76 percent of the far-rightists and 87 percent of the far-leftists. Although these data may well suffer from selection issues, they do provide a meaningful background to our findings in Table 4.

Conclusion

This paper first introduces a basic theoretical framework to formalize potential relationships between income levels and terrorism. Taking a parsimonious approach, the framework is able to generate the main competing hypotheses of the associated literature. In addition, the framework yields a hypothesis pertaining to the role of society-level grievances in the income-terrorism nexus.

Second, we construct a region-level database, mapping terror attacks to income levels and other covariates at the subnational level. This allows us to study panel data for 1527 subnational entities from 1970 to 2014, exploring the validity of our theoretical hypotheses. Most importantly, all results firmly support an inverted U shape in linking income levels to terrorism. This result prevails when accounting for a comprehensive set of covariates, as well as region- and year-fixed effects; when delineating between domestic and transnational terrorism; and even when distinguishing between terror group ideology. Contrary to the post-9/11 claims of poverty being a monotonically positive predictor of terrorism, these results suggest poverty alleviation could imply more terrorism for polities currently positioned to the left of the income peaks we derive. Finally, we derive insightful heterogeneity for the peak level of income along group ideologies and underlying grievances.

Naturally, we advise some caution in a potentially causal interpretation of these findings since, similar to most of the cross-country literature, our analysis is not able to fully resolve all empirical challenges. For example, unobservable factors that change within a subnational region over time may still be able to bias the coefficients we derive. Nevertheless, the subnational data structure allows us to substantially alleviate these endogeneity concerns. Our most complete specifications exploit within-region variation only, i.e., any time-invariant differences across regions (even within the same country) are held constant. Carefully structuring corresponding time sequencing by using contemporaneous GDP/capita levels to predict subsequent terrorism further alleviates concerns about reverse causality. Results also remain consistent when accounting for lagged terror levels.

In sum, the fact that the inverted U-shape emerges in virtually all settings provides what we believe to be the strongest empirical evidence to date for a systematic, universal link between income levels and terrorism. In terms of concrete policy takeaways, the most explicit conclusion from our analyses cautions societies to anticipate a rise in terrorism (rather than a decline) as their polity is lifted out of poverty – contradicting Bush’s post-9/11 conjecture. Thus, initial economic growth would need to be accompanied by careful observation of terrorism-related dynamics and perhaps appropriate counter-terrorism strategies. In turn, once a society is on its path from middle-to high-income status, surpassing the income-terrorism peak, economic growth contributes to reducing terrorism, on average. Overall, we hope these insights can inform regional, national, and international policymakers, as well as inspire further research into a topic that has informed substantial political and societal decisions since 9/11.

Supplemental Material

Supplemental Material - Income and Terrorism: Insights From Subnational Data

Supplemental Material for Income and Terrorism: Insights From Subnational Datas by Michael Jetter, Rafat Mahmood, and David Stadelmann in Journal of Conflict Resolution

Supplemental Material

Supplemental Material - Income and Terrorism: Insights From Subnational Data

Supplemental Material for Income and Terrorism: Insights From Subnational Datas by Michael Jetter, Rafat Mahmood, and David Stadelmann in Journal of Conflict Resolution

Footnotes

Acknowledgements

We are very grateful to the editor, Paul Huth, and the anonymous referees for their thorough engagement with our study. Their comments substantially improved our paper. We are also grateful for comments from Tomohiko Inui, Tim Krieger, Daniel Meierrieks, and the seminar participants at New York University Abu Dhabi, which helped us improve our paper.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Michael Jetter

Rafat Mahmood

Supplemental Material

Supplemental material for this article is available online

Notes

References

Abadie

2006. “Poverty, Political Freedom, and the Roots of Terrorism.” American Economic Review 96 (2): 50‐56.

Alesina

Özler

Roubini

Swagel

. 1996. “Political Instability and Economic Growth.” Journal of Economic Growth 1 (2): 189‐211.

Azam

J.-P.

Delacroix

. 2006. “Aid and the Delegated Fight Against Terrorism.” Review of Development Economics 10 (2): 330‐344.

Azam

J.-P.

Thelen

. 2008. “The Roles of Foreign Aid and Education in the War on Terror.” Public Choice 135 (3–4): 375‐397.

Basuchoudhary

Shughart

W. F.

. 2010. “On Ethnic Conflict and the Origins of Transnational Terrorism.” Defence and Peace Economics 21 (1): 65‐87.

Benmelech

Berrebi

Klor

E. F.

. 2012. “Economic Conditions and the Quality of Suicide Terrorism.” The Journal of Politics 74 (1): 113‐128.

Berman

Laitin

D. D.

. 2008. “Religion, Terrorism and Public Goods: Testing The Club Model.” Journal of Public Economics 92 (10–11): 1942‐1967.

Berrebi

2007. “Evidence About the Link Between Education, Poverty and Terrorism Among Palestinians.” Peace Economics, Peace Science, and Public Policy 13 (1): 1‐38.

Blomberg

S. B.

Hess

G. D.

. 2008a. “The Lexus and the Olive Branch: Globalization, Democratization and Terrorism.” In Terrorism, Economic Development, and Political Openness. Cambridge: Cambridge University Press.

10.

Blomberg

S. B.

Hess

G. D.

. 2008b. “From (no) Butter to Guns? Understanding the Economic Role in Transnational Terrorism.” In Terrorism, Economic Development, and Political Openness. Cambridge: Cambridge University Press.

11.

Blomberg

S. B.

Rosendorff

B. P.

. 2006. A Gravity Model of Globalization, Democracy and Transnational Terrorism. USC CLEO Research Paper. (C06-6).

12.

Bloom

D. E.

Canning

. 2000. “The Health and Wealth of Nations.” Science 287 (5456): 1207‐1209.

13.

Braithwaite

. 2007. “Transnational Terrorism Hot Spots: Identification and Impact Evaluation.” Conflict Management and Peace Science 24 (4): 281‐296.

14.

Bravo

A. B. S.

Dias

C. M. M.

. 2006. “An Empirical Analysis of Terrorism: Deprivation, Islamism and Geopolitical Factors.” Defence and Peace Economics 17 (4): 329‐341.

15.

Brückner

Ciccone

Tesei

. 2012. “Oil Price Shocks, Income, and Democracy.” Review of Economics and Statistics 94 (2): 389‐399.

16.

Burgoon

2006. “On Welfare and Terror: Social Welfare Policies and Political-Economic Roots of Terrorism.” Journal of Conflict Resolution 50 (2): 176‐203.

17.

Bush

G. W.

2002. United States of America: Remarks by Mr. George W. Bush President at the International Conference on Financing for Development. March 22, 2002.

18.

Campos

N. F.

Gassebner

. 2013. “International Terrorism, Domestic Political Instability, and the Escalation Effect.” Economics & Politics 25 (1): 27‐47.

19.

Collier

Hoeffler

. 2004. “Greed and Grievance in Civil War.” Oxford Economic Papers 56 (4): 563‐595.

20.

Crenshaw

Robison

Jenkins

J. C.

(2007). The “Roots” of Transnational Terrorism: A Replication and Extension of Burgoon. In Annual Meetings of the American Sociological Association, New York City, NY, August 2007.

21.

De la Calle

Sánchez-Cuenca

. 2012. “Rebels Without a Territory: An Analysis of Nonterritorial Conflicts in the World, 1970–1997.” Journal of Conflict Resolution 56 (4): 580‐603.

22.

Dreher

Fischer

J. A.

. 2010. “Government Decentralization as a Disincentive for Transnational Terror? An Empirical Analysis.” International Economic Review 51 (4): 981‐1002.

23.

Dreher

Fischer

J. A.

. 2011. “Does Government Decentralization Reduce Domestic Terror? An Empirical Test.” Economics Letters 111 (3): 223‐225.

24.

Easterly

2016. The War on Terror vs. the War on Poverty. New York: The New York Review of Books.

25.

Enders

Hoover

G. A.

. 2012. “The Nonlinear Relationship Between Terrorism and Poverty.” American Economic Review 102 (3): 267‐72.

26.

Enders

Hoover

G. A.

Sandler

. 2016. “The Changing Nonlinear Relationship Between Income and Terrorism.” Journal of Conflict Resolution 60 (2): 195‐225.

27.

Enders

Sandler

Gaibulloev

. 2011. “Domestic Versus Transnational Terrorism: Data, Decomposition, and Dynamics.” Journal of Peace Research 48 (3): 319‐337.

28.

Eyerman

1998. “Terrorism and Democratic States: Soft Targets or Accessible Systems.” International Interactions 24 (2): 151‐170.

29.

Freytag

Krüger

J. J.

Meierrieks

Schneider

. 2011. “The Origins of Terrorism: Cross-Country Estimates of Socio-Economic Determinants of Terrorism.” European Journal of Political Economy 27: S5‐S16.

30.

Friedman

B. M.

2010. The Moral Consequences of Economic Growth. New York: Vintage.

31.

Gaibulloev

Piazza

J. A.

Sandler

. 2017. “Regime Types and Terrorism.” International Organization 71: 1‐32.

32.

Gaibulloev

Sandler

. 2022. “Common Myths of Terrorism.” Journal of Economic Surveys 37: 271‐301.

33.

Gassebner

Luechinger

. 2011. “Lock, Stock, and Barrel: A Comprehensive Assessment of the Determinants of Terror.” Public Choice 149 (3–4): 235‐261.

34.

Gennaioli

La Porta

De Silanes

F. L.

Shleifer

. 2014. “Growth in Regions.” Journal of Economic Growth 19 (3): 259‐309.

35.

Goldstein

K. B.

2005. “Unemployment, Inequality and Terrorism: Another Look at the Relationship Between Economics and Terrorism.” Undergraduate Economic Review 1 (1): 6.

36.

Gosling

T. L.

(2017). “States of Terror: Regional Income and Terrorism.” Unpublished manuscript.

37.

Grossman

H. I.

1991. “A General Equilibrium Model of Insurrections.” The American Economic Review 81: 912‐921.

38.

Gürlük

2009. “Economic Growth, Industrial Pollution and Human Development in the Mediterranean Region.” Ecological Economics 68 (8–9): 2327‐2335.

39.

Haggar

K. E.

2021. President Al-Sisi Fights Terrorism by Eliminating Informal Settlements. Daily News Egypt.

40.

Hassan

2002. “An Arsenal of Believers.” Le Débat 120 (3): 134‐143.

41.

Inada

K.-I.

1963. “On a Two-Sector Model of Economic Growth: Comments and a Generalization.” The Review of Economic Studies 30 (2): 119‐127.

42.

Jetter

Stadelmann

. 2019. “Terror Per Capita.” Southern Economic Journal 86 (1): 286‐304.

43.

Kis-Katos

Liebert

Schulze

G. G.

. 2014. “On the Heterogeneity of Terror.” European Economic Review 68: 116‐136.

44.

Koch

M. T.

Cranmer

. 2007. “Testing the “Dick Cheney” Hypothesis: Do Governments of the Left Attract More Terrorism Than Governments of the Right?” Conflict Management and Peace Science 24 (4): 311‐326.

45.

Krieger

Meierrieks

. 2019. “Income Inequality, Redistribution and Domestic Terrorism.” World Development 116: 125‐136.

46.

Krueger

2007. “What Makes a Terrorist? It’s Not Poverty and Lack of Education, According to Economic Research by Princeton’s Alan Krueger Look Elsewhere.” The American (Washington, DC) 1 (7): 16‐22.

47.

Krueger

A. B.

Laitin

D. D.

. 2008. “Kto Kogo?: A Cross-Country Study of the Origins and Targets of Terrorism.” In Terrorism, Economic Development, and Political Openness, 148‐173. Cambridge: Cambridge University Press.

48.

Krueger

A. B.

Malečková

. 2003. “Education, Poverty and Terrorism: Is There a Causal Connection?” Journal of Economic Perspectives 17 (4): 119‐144.

49.

Kurrild-Klitgaard

Justesen

M. K.

Klemmensen

. 2006. “The Political Economy of Freedom, Democracy and Transnational Terrorism.” Public Choice 128 (1–2): 289‐315.

50.

Lai

2007. “Draining the Swamp”: An Empirical Examination of the Production of International Terrorism, 1968 - 1998.” Conflict Management and Peace Science 24 (4): 297‐310.

51.

2005. “Does Democracy Promote or Reduce Transnational Terrorist Incidents?” Journal of Conflict Resolution 49 (2): 278‐297.

52.

Schaub

. 2004. “Economic Globalization and Transnational Terrorism: A Pooled Time-Series Analysis.” Journal of Conflict Resolution 48 (2): 230‐258.

53.

Maslow

A. H.

1943. “A Theory of Human Motivation.” Psychological Review 50 (4): 370‐396.

54.

Miguel

Satyanath

Sergenti

. 2004. “Economic Shocks and Civil Conflict: An Instrumental Variables Approach.” Journal of Political Economy 112 (4): 725‐753.

55.

Neumayer

Plümper

. 2009. “International Terrorism and the Clash of Civilizations.” British Journal of Political Science 39 (4): 711‐734.

56.

Odede

2015. If You Really Want to Fight Terrorism, Start by Fighting Child Poverty. The Guardian. Published on August 21, 2015.

57.

Piazza

J. A.

2006. “Rooted in Poverty?: Terrorism, Poor Economic Development, and Social Cleavages.” Terrorism and Political Violence 18 (1): 159‐177.

58.

Piazza

J. A.

2007. “Draining the Swamp: Democracy Promotion, State Failure, and Terrorism in 19 Middle Eastern Countries.” Studies in Conflict & Terrorism 30 (6): 521‐539.

59.

Piazza

J. A.

2008a. “Incubators of Terror: Do Failed and Failing States Promote Transnational Terrorism?” International Studies Quarterly 52 (3): 469‐488.

60.

Piazza

J. A.

2008b. “Do Democracy and Free Markets Protect Us From Terrorism?” International Politics 45 (1): 72‐91.

61.

Piazza

J. A.

2011. “Poverty, Minority Economic Discrimination, and Domestic Terrorism.” Journal of Peace Research 48 (3): 339‐353.

62.

Piazza

J. A.

2013. “Regime Age and Terrorism: Are New Democracies Prone to Terrorism?” International Interactions 39 (2): 246‐263.

63.

Pilgrim

2015. Poverty and Injustice Feeding Terrorism, French Minister Says. France 24 News. Published on February 10, 2015.

64.

Plümper

Neumayer

. 2010. “The Friend of My Enemy is My Enemy: International Alliances and International Terrorism.” European Journal of Political Research 49 (1): 75‐96.

65.

Sageman

2004. Understanding Terror Networks. Philadelphia: University of Pennsylvania Press.

66.

Sambanis

2008. “Terrorism and Civil War.” In Terrorism, Economic Development, and Political Openness, 174‐206. Cambridge: Cambridge University Press.

67.

START . 2017a. Profiles of Individual Radicalization in the United States (PIRUS) [Data file].

68.

START . 2017b. Global Terrorism Database. National Consortium for the Study of Terrorism and Responses to Terrorism (START). Retrieved from http://www.start.umd.edu/gtd.

69.

Sterman

2015. Don’t Dismiss Poverty’s Role in Terrorism Yet. Time 4.

70.

Tavares

2004. “The Open Society Assesses its Enemies: Shocks, Disasters and Terrorist Attacks.” Journal of Monetary Economics 51 (5): 1039‐1070.

71.

Testas

2004. “Determinants of Terrorism in the Muslim World: An Empirical Cross-Sectional Analysis.” Terrorism and Political Violence 16 (2): 253‐273.

72.

Uzawa

1963. “On a Two-Sector Model of Economic Growth II.” The Review of Economic Studies 30 (2): 105‐118.

73.

Villiger

2013. “Political Violence: Switzerland, a Special Case?” Terrorism and Political Violence 25 (5): 672‐687.

74.

Walsh

J. I.

Piazza

J. A.

. 2010. “Why Respecting Physical Integrity Rights Reduces Terrorism.” Comparative Political Studies 43 (5): 551‐577.

75.

Wolfensohn

J. D.

2002. “Fight Terrorism by Ending Poverty.” New Perspectives Quarterly 19 (2): 42‐44.

76.

Young

J. K.

Dugan

. 2011. “Veto Players and Terror.” Journal of Peace Research 48 (1): 19‐33.

77.

Young

J. K.

Findley

M. G.

. 2011. “Promise and Pitfalls of Terrorism Research.” International Studies Review 13 (3): 411‐431.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

1.63 MB

0.00 MB

0.29 MB