Economic Sanctions, Poverty, and International Terrorism: An Empirical Analysis

Abstract

This study examines the impact of economic sanctions on international terrorism. It is argued that sanctions intensify economic hardships on the poor within countries and this increases their level of grievance and makes them more likely to support or engage in international terrorism. Further, economic sanctions are conceptualized as creating an opportunity for rogue leaders to manipulate aggrieved poor people to terrorize foreign entities who are demonized as engaging in a foreign encroachment on the sanctioned nation's sovereignty. A cross-sectional, time-series data analysis of 152 countries for the past three decades provides evidence that ceteris paribus, economic sanctions are positively associated with international terrorism. This finding suggests that, although the main purpose of economic sanctions is to coerce rogue countries to conform to international norms and laws, they can unintentionally produce a negative ramification and become a cause of international terrorism.

Keywords:

• economic sanctions
• international terrorism
• poverty

Acknowledgments

We thank Samuel Bassett, John Van Benthuysen, Han Dorussen, Josh Pakter, and Dursun Peksen for their invaluable contribution to the research. The data and replication files are available at http://dvn.iq.harvard.edu/dvn/dv/internationalinteractions, but all questions regarding replication should be directed to the first author.

Notes

¹For an excellent recent survey on poverty and terrorism, see Krieger and Meierrieks (2011) and Piazza (2011).

² Gurr (1970:37) observes that “the frustration-aggression mechanism is in this sense analogous to the law of gravity: men who are frustrated have an innate disposition to do violence to its source in proportion to the intensity of their frustration” (emphasis added).

³It is interesting to ask why aggrieved poor people resort to terrorism rather than some other tactic, such as peaceful protest, participation in elections, creating new political movements, etc. The answer can be found in Kruglanski and Fishman's (2006:194) tool perspective that “views terrorism as a means to an end, a tactic of warfare that anyone could use… Its major concern is the conditions under which an individual or a group would opt for a given course of action versus its possible alternatives, given these actors’ objectives.” This study agrees that terrorism is the most logical “tool” for aggrieved poor people to use. When faced with the unbearably poor financial conditions as a result of the hostile atmosphere created by economic sanctions, aggrieved poor people turn to terrorist tactics to avenge because they have nothing to lose.

⁴ Hudson's (2002) survey research shows that international terrorist groups recruit members heavily from poor areas.

⁵Although it is plausible that the poor may identify domestic sources as the cause of their plight and focus their anger accordingly, this conceptualization falls out of the purview of this project.

⁶Terrorism is referred to as the premeditated use, or threat of use, of extra-normal violence by non-state actors that is employed to garner a political, economic, religious, or ideological objective through the intimidation of a large audience (Enders and Sandler 2006).

⁷The GTD defines terrorism as an intentional act of violence, or threat of violence, by a non-state actor in order to attain a political, economic, religious, or social goal (LaFree and Dugan 2007). For more detailed information on the GTD, see http://www.start.umd.edu/gtd/.

⁸The most frequently used data on human right violations (that is, the Cingranelli-Richards (CIRI) index of physical integrity rights) is compiled since 1981. If this variable is included as an additional control variable, the study period would be limited to the much shorter period, 1981–2004, instead of 1968–2004. Consequently, the omission of thirteen years would make this research unable to capture a broader spectrum of terrorist activity. In addition, as the role of human right abuses is well-documented in other studies (for example, Piazza and Walsh 2010; Walsh and Piazza 2010), replicating the same results would provide little added values to the current study.

⁹It may be argued that for economic sanctions to have an adverse effect on the sanctioned state and then for the poor to organize a terrorist event, it is likely that more than a year will lapse. When 2- to 5-year lags are included in the model specification, the results show that the first- and second-year lags are significant. This finding is consistent with Mueller and Mueller's (1999:49) assessment that upon the imposition of economic sanctions, “in a matter of months or years whole economies can be devastated, as happened in Haiti in 1991 and Serbia in 1992.” The implication of this finding is that poor people tend to be outraged by their economic ordeal in the first two years of sanctions, as their terrorist action is influenced by short-term memory rather than long-term memory.

¹⁰This study also creates three more sanction variables by identifying economic sanctions imposed by the United States, unilateral sanctions, and multilateral sanctions, separately. As shown in Appendix A, the empirical results for these sanctions variables are quite similar to those in Model 1, namely, irrespective of the type of sanctions, their impact on terrorism is counterproductive.

¹¹The Vuong (1989) test may be used to make a statistical choice between negative binomial regression and zero-inflated negative binomial regression (see Greene 2003; Hilbe 2007; Long and Freese 2006). However, because this study estimates robust standard errors, the Vuong test cannot be applied. When Vuong tests are performed without the robust option, zero-inflated negative binomial regression is preferred to negative binomial regression.

¹²Although this study has focused on the causal link between sanctions and international terrorism, it can also be speculated that the imposition of sanctions may instigate domestic terrorist incidents. This possibility is tested in Appendix B. Model 1 evaluates the sanction effect on domestic terrorism and Model 2 considers both domestic and international terrorism. The overall results support the speculation that when the poor are frustrated by economic sanctions, they are also likely to terrorize domestic targets.

¹³After the occurrence of the September 11 terrorist attacks, some sanctions researchers began to extend their research inquiry into the area of terrorism and reported conflicting findings. For example, de Jonge Oudraat (2004) suggests that economic sanctions may be an effective weapon against international terrorism, while Einisman (2000) and Fandl (2004) present their legal perspectives on why economic sanctions are ineffective for fighting terrorism, and O'Sullivan's (2003) case studies on Iran, Iraq, Libya, and Sudan show that the record on economic sanctions is mixed. Unfortunately, these studies fall short of offering systematic evidence to support the sanctions-terrorism connection because their findings are mixed and because only a small number of case studies or historical accounts are introduced as examples.

¹⁴Building simultaneous equations models assumes that political and economic events occur on an annual basis. However, this assumption may confuse cause and effect in the data analysis. An alternative approach may be a dyadic set up with monthly data on the date that sanctions are imposed and the dates of terrorist attacks. As this approach would require a quite different arrangement, it is left for future research.

¹⁵Although Keshk's (2003) two-stage probit least squares model offers some advantages over standard simultaneous equations models, it is still not appropriate for this study because its endogenous variables need to be dichotomous and continuous measures.

¹⁶This study also implements a peace-years correction (also known as logit splines) to take into consideration temporal dependence in the sanction data (Beck, Katz, and Tucker 1998). Since the results do not significantly deviate from those in Table 3, they are not reported in order to save space.

¹⁷Although many previous studies find no strong relationship between the Gini index and terrorist attacks (for example, Abadie 2006), recent research points in the opposite direction. For example, Derin-Güre (2009) uncovers some evidence that more terrorism is likely to occur in countries with high income inequality; and Lai (2007) shows that countries with higher levels of economic inequality are more likely to experience higher levels of terrorism.

As the Gini index suffers from many missing observations, GDP per capita may be proposed as another indicator of poverty. However, because the argument of this study is that economic sanctions have a particularly deleterious impact on distribution of wealth within a country—not overall poverty—and this prompts increased support of terrorism by aggrieved poor people, GDP per capita is not an accurate indicator, as it is a measure of overall poverty of each country or economic development.

¹⁸The limited data for income inequality is the main reason the two-step tests are not put forward as the main empirical analysis of this study.

¹⁹Models 1 and 2 in Appendix C replicate Models 1 and 4 in Table 1 after replacing the dependent variable from general terrorist incidents to United States victims of terrorism and the main independent variable from general economic sanctions to United States sanctions as well as adding a military intervention variable that measures the total number of United States military interventions into the country in question (Choi forthcoming; Pickering and Kisangani 2009).