Of terrorism and revenue: Why foreign aid exacerbates terrorism in personalist regimes

Abstract

States hosting terrorist groups often receive foreign aid from donors who have an interest in reducing the level of terrorism in these countries. However, existing work is inconclusive on the question of whether such aid is effective at bringing about favorable counterterrorism outcomes. Aid scholars argue that the political structure of the recipient regime is an important determinant of development aid effectiveness. I apply this logic to the topic of counterterrorism aid and argue that the effects of foreign aid on terrorism will be conditional on recipient political incentives. In particular, personalist dictatorships are unique in their reliance upon external sources of revenue with which to keep their regimes afloat. Thus, rents from foreign aid encourage these regimes to become counterterrorism “racketeers”, offering their services in exchange for a fee. But rather than fixing the problem, they perpetuate it, as their survival is conditional upon a perpetual security threat. Using a variety of data on regime type, terrorist attacks, and terrorist group duration, I find that in personalist regimes, US aid significantly increases levels of terrorist activity. This paper contributes to the literature linking foreign aid and terrorism by considering domestic politics as an important determinant of counterterrorism aid effectiveness.

Keywords

Authoritarian regimes counterterrorism foreign aid personalism terrorism

Introduction

With diplomatic facilities, businesses, and civilian and military personnel in nearly every country in the world, the US has truly global security interests, which also provide vulnerable and attractive targets for terrorist groups. In rare and extreme cases, the US intervenes militarily to protect these interests directly, but an enduring trait of American foreign policy is the desire to put boots on the ground as rarely as possible. This impulse manifests itself in a number of ways, one of which is the outsourcing of various security-related tasks to smaller powers (Lake, 2011). Thus, for day-to-day protection of American interests from terrorism, the US relies heavily upon the cooperation of trusted local agents—the police, military, and intelligence apparatus of the host country. The counterterrorism strategy of the Obama administration has heavily emphasized policies to “train, build capacity, and facilitate partner countries on the front lines” (White House, 2014), but this practice is by no means new. The 1961 Foreign Assistance Act formalized the allocation of aid explicitly for security purposes which included among its aims “assisting friendly countries to maintain internal security” and “promoting the independence of the recipient country from international communism” (US Congress, 1961). The 1982 Congressional testimony of then Under Secretary of State James L. Buckley is also illustrative: “The marginal U.S. dollar loaned under [Foreign Military Sales] to the Thai or Turkish Army or Pakistani Air Force is a dollar that we otherwise would have to spend outright on our own forces to do a job that the Turks and Thais and Pakistanis can do better and at less cost” (quoted in Siddiqa-Agha, 2001).

Furthermore, many of the countries in which the threat to US interests is greatest are also presided over by regimes with a host of other pressing security priorities (Byman, 2006). Asking governments to undertake drastic changes to their domestic security policies means reconciling with these other priorities, and foreign aid is thought to be a useful tool to compensate recipients for setting aside their own agenda in order to align themselves more closely with donor counterterrorism interests (Azam and Delacroix, 2006; Bandyopadhyay et al., 2011; Boutton and Carter, 2014; Fleck and Kilby, 2010). This mechanism was starkly illustrated by the revelation that the CIA had for years been literally delivering sacks of cash to the office of Afghan president Hamid Karzai in exchange for his cooperation against the Taliban (Rosenberg, 2013).

But is aid actually effective at changing incentives and purchasing counterterrorism cooperation? There are certainly examples of this strategy working as intended. US funding and training played a significant role in the energetic efforts of Indonesian counterterrorism forces against Jemaah Islamiya in the years following the 2002 Bali bombing (Guerin, 2007). Nearly the entirety of the group’s leadership had been arrested by 2012, and the threat from transnational terrorism in Indonesia virtually eliminated. Likewise, the US formalized a diplomatic initiative with Colombia in 2000, part of which entailed a large increase in US military assistance to combat the FARC and ELN. The 15 years that followed witnessed a sustained government offensive, along with a dramatic improvement in the security situation, such that a recently signed peace agreement has the potential to end the nearly 60 year conflict, largely on the government’s terms (Casey, 2016).

However, one can just as easily recall cases in which terrorist activity seemed only to spiral out of control, despite generous US assistance. In the wake of the 2011 American withdrawal from Iraq, President Nouri al-Maliki did little to quell the deteriorating security situation, at times even appearing to intentionally fan the flames. The apparently arbitrary arrest of Sunni opponents, purges of Sunnis from political and military positions, and use of the US-funded counterterrorism force against political rivals and peaceful protesters contributed to the escalating Sunni insurgency (Dodge, 2012). In 2014, as small bands of Islamic State fighters easily took control of major cities—notably Mosul, Ramadi, and Tikrit—entire Iraqi divisions were ordered to withdraw, despite having been trained, funded, and equipped with over $60 billion to deal with precisely these kinds of threats.

Similarly, Al-Qaeda in the Arabian Peninsula (AQAP) made a remarkable comeback in Yemen, thanks to a string of mysterious mass prison breaks beginning in 2006¹ and pardons of AQAP leaders granted by president Ali Abdullah Saleh. As AQAP activity escalated, so did US counterterrorism aid and training. Yet despite high levels of assistance, the group easily took over Abyan and Shabwa provinces in southern Yemen in 2011 as US-funded paramilitary units—controlled by Saleh’s son and nephew—withdrew from the areas (Johnsen, 2013).

What accounts for such disparate outcomes? Why does foreign aid make some governments generally cooperative and effective counterterrorists, while apparently making the problem worse in others? With this paper, I offer an explanation for this puzzle by incorporating the domestic political structure of recipient regimes as a mediating factor in the foreign aid–terrorism relationship.² However, the primary contribution is to offer a new way of understanding particular cases in which terrorism and instability seem intractable. While existing accounts of conflict duration and severity focus on military capacity, bargaining failures, rebel-centric factors such as grievances and recruitment, or exogenous factors such as terrain, I shift the emphasis to the incentives of the host government and relax the assumption that the goal of the government is always to eliminate terrorist and insurgent activity. With several exceptions (e.g. Bapat, 2011; Boutton, 2014; and Steinwand, 2015), most aid–terrorism research fails to adequately consider the role of host state political and strategic incentives.

Briefly, the provision by the US of counterterrorism aid³ has the potential to generate perverse incentives for recipient governments to perpetuate a terrorist threat rather than reduce it, as doing so may endanger future US aid levels. My argument is these incentives will be most striking in countries in which the regime maintains power through the distribution of patronage and private goods. These types of regimes are highly dependent on aid for survival, and also benefit from this particular type of aid, which includes weapons and training for small security force units, which can be deployed to protect the regime. Thus, in order to maintain the flow of assistance that is critical to their survival, personalist regimes will view a worsening terrorist threat as a source of leverage which can ensure continued assistance.⁴

The following section provides a short overview of previous literature on the role of politics in determining aid effectiveness. Next, I present my argument about the effect of personalist political institutions in the relationship between foreign aid and terrorism. In the third section, I derive two implications from the argument and test them empirically. Results indicate that higher levels of US foreign aid are associated with a significant increase in the number of anti-American terrorist attacks, as well as with longer terrorist campaigns in recipient countries characterized by personalist rule. A brief look at the final decade of Ali Abdullah Saleh’s rule in Yemen illustrates the inner workings of the argument. I offer some suggestions for future research in the concluding remarks.

Recipient politics and aid effectiveness

Most aid effectiveness research asks whether or not aid can promote economic growth. While there is currently no consensus on this question, a common theme throughout this literature is the notion that aid is more likely to have a productive impact in systems where the regime is accountable to a larger portion of its citizens (Boone, 1996; Smith, 2008), when the government pursues “good” policies (Burnside and Dollar, 2000) or in countries with institutionalized checks on government power (Svensson, 1999). These institutions ostensibly ensure that recipients will use aid for the benefit of the country, rather than to enrich themselves and their political allies. Thus, Wright and Winters (2010) rightly advocate for a focus on the politics of recipient states when thinking about aid effectiveness.

A leader whose ability to remain in office is a function of the approval of a party, military collective, or a relatively large subset of society should have greater incentive to use aid constructively. This approval is, in part, conditional upon the leader’s performance and provision of public goods, of which stability and counterterrorism comprise one. In the case of counterterrorism, such leaders should want to channel aid toward efforts to disarm terrorist groups and decrease overall levels of terrorism within the country, or else risk being replaced (Aksoy et al., 2015).

On the other hand, a dictator who requires the support of a smaller group in order to remain in office may use aid to solidify support within this coalition (Bueno de Mesquita et al., 2003). Indeed, Licht (2010) finds that aid increases the long-term durability of authoritarian regimes. According to Bueno de Mesquita and Smith (2009: 312), “If faced with a contradiction between actions that enhance their own political welfare and actions that advance societal well-being … recipient leaders will select those policies that benefit themselves”. In fact, there is even evidence that rent-seeking states actually reduce public goods provision when provided with aid (Boone, 1996; Djankov et al., 2008; Svensson, 2000; Wright, 2009), and that aid may have negative long-term effects on institutional development (Moss et al., 2006: 2–20). However, autocracies exhibit substantial variation. The following section examines in greater depth the survival strategies and the nature of political loyalty across dictatorships, and discusses how the institutional landscape in personalist regimes should lead to less effective counterterrorism aid.

Personalism, aid revenue, and patronage

Recent literature on authoritarian regimes generally agrees that dictators generate support and remain in power by relying on some combination of repression of dissent and distribution of patronage to key factions (Gandhi and Przeworski, 2006). The combination varies widely across regimes, and depends on availability of resources and the quality of institutions, as well as the position of the regime in relation to the military.

In a dictator’s ideal world, he would keep all rents for himself and simply rely on a competent, docile military to crush any anti-regime movements. Yet Svolik (2012) identifies a fundamental moral hazard in authoritarian regimes between a dictator and his armed forces: if the dictator uses the military to repress domestic opposition, this signals to the military that (a) the leader is vulnerable and (b) the military is indispensable. This knowledge enables the military to demand concessions from the leader, with the implicit threat that he will be overthrown if he does not grant them. Furthermore, empowering the armed forces to crush domestic threats also facilitates coordination among the security forces, making it easier for security forces to organize and execute a successful coup (Powell, 2014). These dilemmas temper dictators’ use of the army for internal security.⁵

This means that a civilian dictator seeking to remain in power while using repression sparingly must mobilize loyalty and support among elites and society through the distribution of rents. While elites in all regimes engage in rent-seeking in one form or another, the nature of the process, as well as the centrality of patronage to regime survival, are qualitatively different in personalist autocracies, and are the points of departure for this argument.

The most common and durable form of dictatorship for much of the post-war era has been the dominant-party autocracy (Geddes et al., 2014). These dominant parties recruit members early in their careers through grassroots organizations to carry out party-related tasks, while benefits for long-term party membership—political influence and material gain—are realized by senior party members later in their careers. Thus, party membership is an investment, and loyalty among key factions derives not just from material benefits (which can be distributed without a party) but rather from the promise of career advancement as a reward for past service and continued loyalty. The durable nature of the party lends this promise additional credibility, giving all members an enduring, rather than fleeting, stake in regime survival (Svolik, 2012). In the event of a crisis, the party “provides a credible guarantee to in-groups that their long-term interests are best-served by remaining loyal to the regime” (Smith, 2005: 431), thus preventing large-scale elite defection, even if short-term material benefits abate.⁶ Additionally, the higher levels of bureaucratic capacity (see Figure 1) found in party regimes enable them to compensate for revenue shortfalls by raising taxes (Escriba-Folch and Wright, 2010).

Figure 1.

Distributions of bureaucratic capacity (1984–2007) and logged coup probability (1970–2012) in personalist and non-personalist regimes. The bureaucratic capacity measure comes from Hendrix and Young (2014), while data on coup probability are taken from Ulfelder (2014).

Party members remain loyal to the regime for reasons other than immediate material gain. The knowledge that they are better off working within the party to realize their political ambitions than attempting to overthrow it makes loyalty the best strategy even if the flow of private goods wanes. As a result, leaders in party dictatorships rely less on pure material benefits to retain loyalty, and the consequences of crises and revenue shortfalls—such as aid cutoffs or reductions—are not necessarily fatal for the regime. The fact that Mexico’s ruling PRI continued to dominate Mexican politics for nearly 20 years after the 1982 Latin American debt crisis is a testament to the loyalty of elites in a party regime, even during a crisis (Magaloni, 2006).

Personalist regimes, by contrast, lack well-developed parties. Political power is concentrated in the hands of an individual, or small group, whose goal is to transfer society’s resources to themselves. Institutions such as parties and the military are marginalized, meaning that political privilege and career advancement depend upon a personal connection to the leader’s inner clique. With power limited to this inner circle, and without an institutionalized party to incentivize cooperation, acquiring political power entails overthrowing the regime. Politics thus resembles a prisoners’ dilemma in which “the rivalries and struggles of powerful and willful men, rather than impersonal institutions, ideologies, public policies, or class interests, are fundamental in shaping political life” (Jackson and Rosberg, 1984: 421).

Although personalist regimes are narrow and characterized by inefficient policies, they are able to forestall challenges to their rule by distributing selective incentives to key sectors of the elite and masses, usually using public resources and foreign aid. Indeed, Bratton and van de Walle (1997) argue that the distribution of rents and material inducements to political rivals is the defining feature of personalist regimes. As long as their loyalty to the leader is rewarded with access to patronage, rival factions have strong reasons not to overthrow the leader, and will mobilize their supporters to back the regime as well. These elites are complicit with the regime’s corruption and misrule, derive their livelihood from remaining loyal to the regime, and thus have nearly as much to lose as the leader should political order collapse (Bratton and van de Walle, 1997).

Effects of revenue shortfalls

In the event that revenue does wane as a result of aid cutoffs, sanctions, or some other shock, the effects on stability are most immediate and severe in personalist regimes for several reasons. First, these regimes lack the party infrastructure necessary to generate cooperation among elites. Whereas a leader in a democracy or party-based autocracy can make credible promises of future benefits during crises, personalist leaders cannot. The “glue” holding personalist regimes together is the network of rents and patronage that gives elite factions access to the spoils of the regime. Should these benefits diminish, no party or other mechanism exists to encourage elite cooperation; any promises by the leader of future rents are non-credible. Thus, rival factions no longer have a reason to support the regime, and they defect immediately, often mobilizing the masses to do the same (Escriba-Folch and Wright, 2010). Along the same lines, Escriba-Folch and Wright (2010) show that the destabilizing effect of sanctions is most acute in personalist regimes, since their lack of bureaucratic capacity renders them unable to compensate for revenue shortfalls.⁷ In other words, the institutions in non-personalist regimes lower the price of loyalty: a reduction in patronage resources in a personalist regime reduces the leader’s ability to retain support much more than the equivalent reduction in other regimes.

Second, in the event that a revenue shortage does lead to elite defection and mass protest, activating the military for repression is not a viable option for personalist leaders who wish to remain in power. Svolik’s moral hazard is particularly acute for them: the risk of regime change via coup is especially high (see Figures 1 and 2), particularly during mass unrest (Bell and Sudduth, 2015; Powell, 2014).

Figure 2.

Frequency of regime change vs reshuffling coups in dictatorships (from Aksoy et al., 2015).

Finally, personalist dictators’ inability to make credible promises, the absence of institutions, and their usually ruinous and exclusionary policies lead to disproportionately violent exits from office. Escriba-Folch (2013) shows that the political careers of personalist dictators culminate in imprisonment, exile, or death 61% of the time. By contrast, leaders in party autocracies and monarchies tend to enjoy the safest post-tenure fate. Leaders in military regimes often agree to rotate leadership among members of a junta, or simply return to the barracks after negotiating a transition to a civilian regime (Geddes, 2003). Democratic leaders, who are subject to regularized mechanisms governing the transfer of power, are also unlikely to be punished after leaving office.

All of this means that personalist leaders’ rule—more so than that of any other leader—rests on a razor’s edge: it is totally dependent upon a continual source of revenue with which to buy the support of rival factions, and to equip and train coup-proofing units. Because of the lack of other viable survival strategies, any disruption in the flow of patronage goods can lead to the rapid fall of the regime and possible death of the leader. The primary goal of personalist leaders must therefore be to procure foreign aid; welcoming or deliberately amplifying the presence of terrorism within their borders—particularly terrorism that targets wealthy donor states—is an effective way to do this.

It is worth taking a moment to discuss the assumption implicit in this argument that leaders in personalist regimes have little to fear from terrorism, and that the benefits of hosting terrorism are worth the potential risks. Literature on leaders in these regimes has reached a near consensus that their primary security concern is the threat of removal from office by regime insiders via coup. Indeed, this by far is the most common manner through which dictators lose office (Powell and Thyne, 2011), and leaders are willing to go to great lengths to defend against this threat. In this respect, my argument fits well with Powell (2014) and Roessler (2011), both of whom show that rational African leaders deliberately increase the likelihood of civil war in order to coup-proof their regimes. Recently, Aksoy et al. (2015) have shown that terrorism does not destabilize authoritarian regimes, although it can lead to a reshuffling of leadership in regimes headed by a collective, such as a military junta or a party elite, as a way to hold the leader accountable. However, these accountability mechanisms are absent in personalist regimes. By definition, personalist leaders have marginalized any individuals or entities who might hold them accountable. We can see this by looking again at Figure 2. These “reshuffling” coups are relatively rare in personalist regimes, and they fail more often than they succeed. By contrast, regime-change coups occur when rivals of the regime seek to completely replace the government. While personalist regimes experience these types of coups more frequently, Aksoy et al. (2015) show that they are not associated with terrorism. Rather, these occur more often as a result of mass-based protests against the regime, which typically stem from dwindling revenue for patronage and subsidies (Escriba-Folch and Wright, 2010). With all of this in mind, it becomes clearer why a personalist dictator might rationally trade stability for a continuous aid flow.

The following section discusses how donors’ inability to enforce aid conditions enables recipients to be deceptive about their counterterrorism efforts with relative impunity.

Conditionality and donor commitment issues

Why do donors not simply withdraw aid to punish partner countries who undermine their security interests? Donors nearly always place a set of conditions on aid packages that specify some effort level or outcome that must be achieved if aid is to continue (among others, Easterly, 2006; Collier, 2007). For these conditions to be effective, the donor’s threat to punish by withdrawing aid must appear credible to the recipient (i.e. Schelling, 1966). However, in counterterrorism cases, US officials often prefer sending good money after bad, rather than risk making the problem worse by withdrawing aid, and recipients often know this. Former Afghan president Hamid Karzai continued to receive assistance (and White House invitations), even after openly threatening to join the Taliban (Rubin, 2010). The knowledge that the aid flow will likely not be jeopardized as long as a terrorist threat is present—even if the recipient is thought to be unreliable—helps create the moral hazard that is central to this argument.

Hypotheses

The argument detailed above points to expectations about the actions that might be taken by a recipient government to either suppress or amplify terrorist activity inside its borders. The concept of “counterterrorism” remains ill-defined by scholars, spans across numerous policy areas, and can be difficult to observe. Counterterrorism “effort” could include overt activities such as military and police raids on terrorist hideouts, arrests, assassinations, or passage of anti-terrorism legislation. However, more covert, less observable actions, such as interrogations, changes in law enforcement/intelligence activity or budgets, or the freezing of financial assets, are no less pertinent.

Likewise, actions by the host government that exacerbate terrorism may be equally difficult to observe and measure, since the host would like such activity to remain unknown, or imperfectly known, to the donor. These activities likely consist of some combination of what Byman (2008) and Staniland (2012) would consider “passive” and “active cooperation”. The former includes tolerating or simply turning a blind eye toward terrorist activity, tacitly allowing groups to occupy certain territory, and agreeing to ceasefires. On the other hand, “active” cooperation or sponsorship suggests a more hands-on, overt state–terrorist group relationship, in which the two parties cooperate toward shared or complementary goals. This may entail, but is not limited to, the release of known terrorists from captivity, the provision of money, weapons, or logistics, or assistance with attacks.

The fact that counterterrorism contains a mix of observable and imperfectly observable actions that are unobservable and imperfectly observable presents an obstacle to conducting an unbiased empirical test of the theory. Furthermore, capturing a government’s attempts to covertly manipulate their level of terrorism—which is central to this argument—is impossible. Perhaps jailbreaks would happen more frequently, as we have seen in Libya, Iraq, Yemen, and Pakistan recently (Johnsen, 2013; Reuters, 2013; Salman and Parker, 2013), but as the discussion of the Yemen case will show, these actions are mostly hidden.

While reliable data on these concepts are not currently available, we can use levels of terrorism to draw inferences about these unobservable behaviors. Specifically, where aid generates perverse counterterrorism incentives, we would expect to see a prolonged terrorism threat, in the form of longer-lasting terrorist groups. This leads to my first hypothesis:

Hypothesis 1: Groups based in states governed by personalist regimes will be less likely to fail in a given year as US foreign aid increases.

My second hypothesis concerns how personalism conditions the effect of foreign aid on the level of activity by terrorist groups in recipient countries. The goal of US aid may not necessarily be to completely eliminate the groups, but rather to convince the regime to prevent attacks against American citizens and interests. The US may be indifferent to the prolonged presence of terrorist groups in recipient countries as long as its interests remain safe. Thus, personalist regimes may inflate their level of terrorism in order to ensure continued US assistance. This leads to the following hypothesis:

Hypothesis 2: Personalist regimes will experience more terrorist attacks in a given year as the amount of US foreign aid increases.

Research design

Dependent variables

To evaluate the group duration hypothesis, I use a modified version of the data used in Jones and Libicki (2008). For the second set of hypotheses in which terrorist attacks are the quantity of interest, I take data from the Global Terrorism Database, which contains information on terrorist attacks between 1970 and 2010. These variables are described in detail below.

Group duration

The Jones and Libicki (2008) data identify 603 internationally hosted, non-state actors between 1968 and 2006. In order to model group duration, I converted the data into binary time-series cross-sectional data, with one observation per year per country that the group was in existence.

In a few cases, groups had legitimate bases in multiple countries. When this was the case, I sought confirmation of different bases from US Congressional Research reports, from the Global Terrorism Database (START, 2012) and from group histories when available. Where appropriate, I coded a group’s presence in an additional country from the date that the secondary base was established. With few exceptions, these groups began operations in multiple countries more or less simultaneously. For example, the Salafist Group for Call and Combat has carried out attacks in Morocco, Algeria, Mali, and Mauritania and is coded accordingly. This coding strategy is appropriate because the US recognizes that these groups often traverse freely between countries, and its aid patterns reflect this (Ellis, 2004).

Because I seek to explain counterterrorism effort by the recipient state against these groups, my dependent variable is the failure of a group in year t as a result of actions taken by the state. The Jones and Libicki data code five types of group failure: accomplishment of group goals; entry into politics; defeat by external military intervention; defeat by local police and/or intelligence work; and internal dissolution/splintering. I consider only the latter two, since these are the result of actions taken by the host state, which is ostensibly what the US has in mind when training and aiding these states. Therefore, I combine police failures with splintering⁸ to form my dependent variable. This variable takes a value of 1 if the group fails in year t, and a 0 otherwise:

y_{i, t} = {\begin{matrix} 0 & if group is active in year t or " ends " for reasons unrelated to host state action \\ 1 & if group fails in year t due to internal dissolution or police force / intelligence \end{matrix}

When a group “fails” as a result of host state actions, it leaves the data and is not included in any subsequent years; there are 82 failures by police force/intelligence and 136 from splintering, together accounting for just over 45% of the 481 total group failures in the data. Groups which fail for any of the other three reasons remain in the data until they “fail”, at which point they drop out of the data without ever having “failed” in the model.

Attacks

Data on terrorist attacks are taken from the Global Terrorism Database (START, 2012), which is useful as it contains information on the location, year, and primary target of over 80,000 terrorist attacks from 1970 to 2012. I construct two separate dependent variables based on the primary target of each attack: (a) a count of the number of attacks directed at US interests in a host country in a given year; and (b) attacks directed at non-US targets. Descriptive statistics for the attack variables are presented in the Online Appendix.

Main independent variables

US aid

The theory outlined above specifies that levels of US foreign aid provided should influence the likelihood of a group failing in year t, conditional upon whether the recipient state is a personalist regime. I construct three separate explanatory aid variables, one each for levels of security, economic, and total aid (the first two categories added together). The argument I make above applies primarily to security-related assistance. This type of aid is delivered directly to the government or security forces of the recipient country, whereas economic aid often bypasses the recipient government and is implemented directly by NGOs or private contractors (Dietrich, 2013). Still, I test each category separately.

What I consider “security aid” consists of military aid (i.e. foreign military financing, loans, grants, and training programs), Anti-Terrorism Assistance (part of the Department of State’s Nonproliferation, Anti-terrorism, Demining, and Related Programs), Department of Defense Security Assistance, and Defense Department funding under sections 1206 and 1207. I assembled data on economic aid, military aid, Anti-Terrorism assistance, and Defense Department security assistance from the US Agency for International Development Greenbook (US Agency for International Development, 2014).

Section 1206 and 1207 funding programs began in 2007. Both operate through the Department of Defense, and are broadly meant to bolster the security forces of a smaller set of recipient countries (US General Accounting Office, 2010). I gathered data on these programs separately, as they are not listed in the Greenbook.

I make a couple of adjustments to these variables. First, aid levels exhibit considerable semi-random fluctuation from year to year. The argument here concerns the amount of aid that the regime has become accustomed to and thus needs to maintain. In order to capture this, for each year I calculate a moving average of the previous five years

{(\frac{A i d_{t - 1} + A i d_{t - 2} + A i d_{t - 3} + A i d_{t - 4} + A i d_{t - 5}}{5})}^{9}

Second, aid variables tend to be highly right-skewed. Values for total aid, for example, range from $ 0 to over $33 billion, with a mean of $203 million. To correct for this, I take the natural log of $(1 + m o v i n g a v e r a g e o f a i d)$ for each category. This smooths out the distribution of the aid variables, and also allows for the decreasing marginal returns to aid. Summaries of the aid variables are presented in an Online Appendix.

Personalism

To classify personalist regimes, I draw upon data on authoritarian regimes recently assembled by Geddes et al. (2014), who follow the classification scheme presented in Geddes (2003), updated to 2010. Like other regime types, personalist regimes are classified based on answers to a certain set of questions about that regime. The questions relevant to personalist regimes pertain to the role of the leader’s family and inner circle, the function of political parties, elections, military hierarchy and organization, and control over the security apparatus.¹⁰ Each country-year is coded as either personalist or not. Because I expect the effect of foreign aid to be conditional on personalism, I multiply the personalist indicator by each aid variable to create the interaction terms. In both the duration and count models, I include an array of terrorism-relevant control variables. Although these are reported in the tables below, I describe them in full detail in an Online Appendix.

Endogeneity

Azam and Delacroix (2006) report a positive relationship between aid and terrorism because donor countries allocate more aid to countries that experience high levels of terrorism, and which are likely to continue experiencing terrorist activity in the future. Any estimate of the effect of aid on terrorist activity that does not account for endogeneity will likely suffer from bias as a result. Most scholars have relied on instrumental variables and two-stage regressions in order to achieve a plausible claim of exogeneity in econometric models (Sovey and Green, 2011). However, this method has been shown to be problematic in non-linear regression (Terza et al., 2008). Thus, I emulate Azam and Delacroix (2006) and Young and Findley (2011) by estimating a series of linear regressions in which a set of exogenous, aid-relevant covariates are used to explain the endogenous variable (aid). The residuals from these reduced-form equations are then included in the equations for the attack-count models. The first-stage aid regressions are presented in an Online Appendix.

Results

Aid, personalism, and group duration

To analyze the conditional effect of US aid on group duration, I estimate a series of grouped duration models in which the outcome variable is either failure by host state intelligence/police work or internal dissolution (=1), or not (=0). All time-varying covariates are lagged one year. The estimation gives the effect of the independent variables on the probability of group failure in a given year, so a negative (positive) coefficient means that variable makes group failure less (more) likely. The fixed-effects models control for time-invariant, country-specific factors, and estimate within-country variation to ensure that the results are not simply a reflection of cross-national variation. The results of these models are shown in Table 1.¹¹

Table 1.

Aid averages, personalism, and group duration

Aid type	Security		Economic		Total
Aid type	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6
Aid to non-personalist regime	−0.008	0.018	0.045	0.027	0.037	0.023
	(0.028)	(0.014)	(0.026)	(0.017)	(0.029)	(0.016)
Aid to personalist regime	−0.198**	−0.111**	−0.271**	−0.107*	−0.255**	−0.102*
	(0.076)	(0.042)	(0.080)	(0.045)	(0.077)	(0.042)
Personalist regime, no aid	1.415	0.581	2.725*	0.720	2.717*	0.718
	(1.028)	(0.604)	(1.080)	(0.709)	(1.122)	(0.701)
Democracy	−0.631	−0.673**	−0.608	−0.605*	−0.640	−0.620*
	(0.513)	(0.254)	(0.516)	(0.249)	(0.517)	(0.250)
Military regime	−1.531*	−0.918*	−1.478*	−0.890*	−1.510*	−0.899*
	(0.721)	(0.399)	(0.722)	(0.398)	(0.725)	(0.399)
ln(GDP per capita)	0.010	0.353**	−0.079	0.364**	−0.076	0.359**
	(0.492)	(0.120)	(0.515)	(0.124)	(0.517)	(0.125)
ln(population)	2.944**	−0.067	2.805**	−0.090	2.812**	−0.079
	(0.906)	(0.136)	(0.898)	(0.132)	(0.905)	(0.133)
Territorial group	0.475	0.051	0.447	−0.006	0.449	0.031
	(0.257)	(0.208)	(0.257)	(0.210)	(0.257)	(0.208)
States sponsored	−0.300	−0.217	−0.289	−0.231	−0.292	−0.224
	(0.267)	(0.243)	(0.265)	(0.243)	(0.266)	(0.243)
Regime change group	0.411	0.211	0.366	0.198	0.369	0.216
	(0.232)	(0.190)	(0.233)	(0.191)	(0.233)	(0.190)
Status quo group	−0.156	−0.706	−0.222	−0.721	−0.223	−0.698
	(0.678)	(0.627)	(0.680)	(0.628)	(0.680)	(0.627)
Post-911	−0.060	−0.039	−0.090	−0.074	−0.057	−0.050
	(0.245)	(0.205)	(0.246)	(0.206)	(0.246)	(0.204)
Cold War	0.260	−0.079	0.134	−0.114	0.144	−0.105
	(0.273)	(0.183)	(0.278)	(0.182)	(0.277)	(0.182)
Recipient rivalry	0.102	−0.366	−0.166	−0.349	−0.038	−0.385*
	(0.598)	(0.193)	(0.615)	(0.186)	(0.604)	(0.195)
Country size		0.064		0.063		0.058
		(0.112)		(0.112)		(0.112)
Ruggedness		0.110		0.091		0.091
		(0.128)		(0.129)		(0.130)
Observations	5986	5986	5986	5986	5986	5986
Log Likelihood	−682.944	−736.820	−682.010	−737.282	−682.367	−737.346
Country fixed effects	✓		✓		✓

p < 0.05; **p < 0.01.

In the interest of space, I do not report estimates for time polynomials or continent controls. Full tables are available in the Online Appendix.

Table 1 contains the results of six logit models, two each (with and without fixed effects) for five-year moving averages of US security, economic, and total aid, estimated separately. Based on the estimates for Aid to non-personalist regime, none of the three categories of aid appears to have any significant effect on the outcome when allocated to non-personalist regimes.

The value of primary interest is the coefficient estimate for Aid to personalist regime, which gives us the effect of aid on group duration in personalist regimes. In all six of the models, the effect is negative and statistically significant, meaning that higher average aid levels over five years are associated with a decrease in the probability of terrorist group failure in these states. This suggests support for hypothesis 1.

The estimates for personalist regimes receiving no aid (Personalist, no aid) are positive—groups based in non-aid-receiving personalist regimes fail at higher rates, although the effect is only significant in Models 3 and 5. Given that, it is still the case that aid to personalist regimes is associated with even lower probabilities of failure than if they received no aid at all.

To get a sense of the substantive meaning of these findings, Figures 3 and 4 show what we can expect as US aid increases in personalist and non-personalist regimes.¹² Figure 3 shows that as US security assistance increases over its full range in non-personalist regimes (i.e. from $ 0 to $5.2 billion), the probability of group failure increases slightly, but the effect is nowhere near statistical significance (as indicated by the small and insignificant estimate for Aid, non-personalist regime). Thus, aid to non-personalist regimes makes little difference for the duration of terrorist groups in those states. However, Figure 4 shows that increases in security-related aid to personalist regimes are associated with a steep monotonic decrease in probability of failure. This translates into longer-lasting terrorist campaigns in these states, which is in line with my expectation that personalist regimes may seek to prolong terrorist activity as a source of external revenue, and lends further support for Hypothesis 1.

Figure 3.

Effect of US aid in non-personalist regimes.

Figure 4.

Effect of US aid in personalist regimes.

Aid, personalism, and terrorist attacks

Results in the previous section suggest that US foreign aid in some cases leads to longer terrorist campaigns. It is fair to point out, however, that the US may have more limited aims when allocating aid for counterterrorism purposes. Specifically, the goal may not necessarily be the complete elimination of terrorist groups, but may instead be preventing US interests in the recipient country from being attacked. Therefore, in order to maintain a level of threat to US security and ensure continued assistance, recipient states may facilitate attacks directly against US interests. This section contains the results of the count models assessing Hypothesis 2.

Table 2 contains the results of the count models estimating the number of terrorist attacks in a country in a given year. In Models 1–4, the dependent variable is attacks against non-US targets, while in Models 5–8, the DV is anti-US attacks.¹³ I also analyze security aid separately, and then together with economic aid in the total aid columns, and present fixed- and random-effects versions of each model.

Table 2.

Aid averages, personalism, and terrorist attacks

Aid type	Non-US attacks				Anti-US attacks
	Security		Total		Security		Total
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 6	Model 7	Model 8
Aid to non-personalist regime	−0.006	0.007	−0.010	−0.031**	0.055**	0.057**	−0.008	0.007
	(0.008)	(0.008)	(0.010)	(0.012)	(0.013)	(0.013)	(0.019)	(0.029)
Aid to personalist regime	0.040**	0.038**	0.176**	0.071*	0.195**	0.088*	0.518**	0.338*
	(0.014)	(0.015)	(0.027)	(0.028)	(0.038)	(0.040)	(0.116)	(0.145)
Personalist regime, no aid	−0.794**	−0.752**	−3.429**	−1.489**	−3.652**	−1.256*	−10.142**	−6.312**
	(0.233)	(0.208)	(0.490)	(0.490)	(0.631)	(0.641)	(2.197)	(2.792)
Democracy	0.139	−0.309**	0.083	−0.337**	−0.034	−0.094	0.014	−0.306
	(0.117)	(0.101)	(0.117)	(0.100)	(0.223)	(0.162)	(0.220)	(0.323)
Military regime	0.776**	0.562**	0.711**	0.542**	0.265	0.340	0.264	0.180
	(0.133)	(0.121)	(0.133)	(0.120)	(0.240)	(0.182)	(0.236)	(0.294)
Recipient rivalry	0.937***	0.852***	0.949***	0.859***	1.085***	1.110***	1.018***	1.188***
	(0.098)	(0.069)	(0.098)	(0.069)	(0.181)	(0.109)	(0.179)	(0.245)
ln(GDP per capita)	1.612**	0.298**	1.725**	0.377**	1.627**	0.475**	1.828**	0.501**
	(0.089)	(0.035)	(0.092)	(0.039)	(0.175)	(0.057)	(0.181)	(0.172)
ln(population)	5.419**	0.918**	5.284**	0.946**	3.285**	0.674**	3.344**	0.670**
	(0.157)	(0.033)	(0.157)	(0.033)	(0.267)	(0.056)	(0.262)	(0.170)
Civil war	1.824**	2.634**	1.817**	2.614**	0.856**	1.197**	0.790**	1.254**
	(0.080)	(0.080)	(0.080)	(0.080)	(0.140)	(0.120)	(0.139)	(0.233)
Coup	0.069	−0.040	0.063	−0.046	0.162	0.078	0.151	0.078
	(0.081)	(0.093)	(0.081)	(0.092)	(0.128)	(0.150)	(0.127)	(0.155)
Post-911	−1.061**	−0.376**	−1.140**	−0.452**	−0.524*	0.367	−0.673**	0.389
	(0.103)	(0.129)	(0.104)	(0.129)	(0.209)	(0.223)	(0.211)	(0.331)
Cold War	−0.108	−1.252**	−0.106	−1.217**	0.224	−0.520**	0.339*	−0.514**
	(0.085)	(0.085)	(0.085)	(0.085)	(0.141)	(0.131)	(0.141)	(0.183)
Media censorship	−0.101*	−0.041	−0.092	−0.055	−0.088	0.039	−0.079	−0.024
	(0.048)	(0.041)	(0.048)	(0.041)	(0.083)	(0.066)	(0.083)	(0.137)
Country size		−0.314**		−0.308**		−0.172**		−0.149
		(0.027)		(0.027)		(0.048)		(0.126)
Ruggedness		0.492**		0.472**		0.475**		0.450**
		(0.031)		(0.031)		(0.056)		(0.139)
Endogeneity	−0.030**	0.005	0.021	0.066**	−0.016	0.003	0.079**	0.078*
	(0.008)	(0.009)	(0.012)	(0.014	(0.014)	(0.015)	(0.023)	(0.034)
Constant	−109.515**	−11.449**	−108.510**	−12.957**	−79.263**	−16.203**	−82.547**	−16.813**
	(2.884)	(0.590)	(2.927)	(0.668)	(5.018)	(0.978)	(5.013)	(2.675)
Observations	6334	6334	6334	6334	6334	6334	6334	6334
Country fixed effects	✓		✓		✓		✓
Log likelihood	−11,720.680	−12,971.980	−11,721.710	−12,961.260	−2850.561	−3346.665	−2841.298	−3342.373

p < 0.05; **p < 0.01.

Full tables containing indicators for Global Terrorism Database collection methodology and continents can be found in the Online Appendix.

The effect of aid in non-personalist regimes (Aid to non-personalist regime ) is ambiguous: the signs on the coefficients are sometimes positive and sometimes negative. Even when the coefficient is significant, the estimated effect is small, and thus it is difficult to determine the effect of aid on terrorism in non-personalist regimes. In addition, the coefficient on the Personalist regime, no aid variable is large, negative, and significant in each model: personalist regimes receiving no aid experience fewer of both types of terrorist attacks than do non-personalist regimes. This finding has been documented in recent studies (e.g. Conrad et al., 2014; Piazza and Wilson, 2013), but the causal mechanism is unclear. Some argue that because personalist regimes generate fewer audience costs than other types of regimes, terrorism is less likely to result in policy change, and is thus viewed as an inferior tactic among those who might use it (Conrad et al., 2014). It may also simply be an artifact of under-reporting of terrorism in some personalist dictatorships. Regardless, this makes the positive coefficient on the Aid to personalist regime interaction term that much more meaningful. Together with the finding in Table 1 that groups fail faster in non-aid-recipient personalist regimes, this means that non-aid recipient personalist regimes are less susceptible to terrorism relative to other regime types. However, increases in US aid reverse this effect, leading to higher levels of terrorist activity. Furthermore, across the board, the marginal effects in the “Anti-US attacks” models are much larger than for Non-US attacks, suggesting that the effect is more pronounced for attacks against US interests. This is evidence in favor of the argument that personalist dictators strategically cultivate terrorism against certain targets.

Figure 5 helps paint a clearer picture of the conditional effect of aid on terrorism in both personalist and non-personalist regimes. The conditional marginal effect of US aid on anti-US attacks appears on the y-axis and is based on the coefficients in Model 7 in Table 2. While the effect of aid is essentially zero in non-personalist regimes, the effect is large and positive in personalist regimes (determined by adding $β_{Aid, non-personalist}$ + $β_{Aid, personalist}$ , or −0.008 + 0.518 = 0.51; Brambor et al., 2006). Although the effect of aid in non-personalist regimes is positive in two of the models, we can always expect aid to have a significantly larger effect on anti-US terrorism in personalist dictatorships.

Figure 5.

US aid and anti-American terrorism, 1970–2010.

Counterterrorism in Yemen under Ali Abdullah Saleh

This section includes a brief overview of developments in Yemen between roughly 1999 and 2012, a case which nicely illustrates the argument and the empirical results. The aim here is to examine the mechanisms along the causal chain between the provision of aid, perverse incentives, and the perpetuation/escalation of the threat. Doing so should provide a better understanding of the data generating process, and thus some insight into a recent case that has thus far baffled policymakers.

Former Yemeni president Ali Abdullah Saleh maintained his personalist rule for 33 years by expertly manipulating outside powers and navigating domestic political rivalries (“dancing on the heads of snakes”, as he described it; Clark, 2010). His power rested on a mix of elite-targeted patronage, fuel subsidies, and the protection of loyal paramilitary units. Saleh also played a central role in US counterterrorism policy beginning in the late 1990s following an flurry of terrorist attacks linked to the region, most notably the 1996 Khobar Towers attack in Saudi Arabia, the 1998 US embassy attacks in Kenya and Tanzania, the 2000 bombing of the USS Cole, and the attacks of September 11, 2001. The US increased its assistance to the Saleh regime, helping to establish and train the Central Security Force and Counter Terrorism Unit, paramilitary organizations headed by Saleh’s son and nephew, respectively, and created specifically to fight al-Qaeda (see Figure 6).

Figure 6.

US aid and terrorism in Yemen, 1990–2012.

Meanwhile, Saleh was alienating domestic political rivals and allies—including the powerful Hashid tribal confederation, as well as the popular General Ali Muhsin and his supporters—by concentrating power in the hands of his family, reneging on promises not to stand for re-election in 2006, and harassing challengers during the election (Johnsen, 2013: 179). Mounting opposition to Saleh necessitated damage-control efforts to protect his regime, and to ensure that his loyal paramilitary units were well-armed.

The US had prioritized counterterrorism in its foreign assistance efforts in Yemen and elsewhere in the region, and the conclusion Saleh drew from a 2005 visit to the White House was that without a credible terrorist threat, ’Yemen was just another poor country in a world of beggars’ (quoted in Johnsen, 2010). A series of prison breaks began in the mid-2000s—23 in total between 2006 and 2014—involving convicted high-value terrorists. The most consequential of these occurred in 2006, when 23 convicted militants tunneled out of a prison in Sana’a. According to the regime’s account, the prisoners used spoons to dig a 300-foot tunnel under the prison to a nearby mosque. Among the escapees were Jamal al-Badawi and Fawaz al-Rabaie—coordinators of the 2000 bombing of the USS Cole and the French tanker Limburg, respectively—as well as Nasir al-Wuhyahsi and others who would later form the leadership of a regenerated AQAP (Johnsen, 2013: 192–208). Evidence suggests that these escapes were staged by the Saleh regime (Phillips, 2011; Johnsen, 2013: 193–195), and typically coincided either with US refusals to meet Saleh’s aid demands or with internal challenges to the regime. In March 2011, in response to high-level elite defections (including General Ali Muhsin) and a US military aid suspension following a massacre of civilian protesters, Saleh secretly released 70 prisoners from the political security headquarters in the Yemeni capital (Scahill, 2012). During the height of the anti-regime protests in 2011 and 2012, 379 prisoners escaped from Yemeni prisons (Al-Moshki, 2014). Saleh used the escalating jihadist presence as leverage when bargaining with the US for more aid and military hardware: “If you don’t help, this country will become worse than Somalia” (Wikileaks, 2009).

Some evidence suggests that the regime’s involvement with AQAP may have extended to more direct assistance as well, including non-aggression pacts with released convicts, exchanges of territory, and provision of weapons and funding. US officials would learn that between 2003 and 2009, Saleh had released Fahd al-Quso (an AQAP deputy linked to the Cole bombing and the 9/11 attacks) and Mohammed al-Ahdal (another Limburg participant) from custody, and had also agreed not to capture Jamal al-Badawi and Jabir al-Banna in exchange for promises that AQAP would not attack Yemeni targets (Johnsen, 2013: 223).

Recently, an Al Jazeera report publicized claims made by a former AQAP informant for Yemeni intelligence. The informant alleged that he repeatedly warned Ammar Saleh and officers in the Political Security Organization months in advance, and in tactical detail, of both the 2007 suicide bombing which killed eight Spanish tourists (masterminded by one of the 2006 escapees), and the 2008 US Embassy attack which killed 19 (Ammar Saleh was the president’s nephew and intelligence chief, and acted as a key contact for US intelligence agencies). The latter attack was already viewed with suspicion after a road block in front of the embassy had been removed that day without explanation, giving the attackers access to the gate. In addition, both Saleh and his family had contacts with Qasim al-Raymi—another 2006 escapee and AQAP military commander—and facilitated attacks against Western targets by supplying funding and weapons through him. The informant further alleges that seven government officials knew the whereabouts of AQAP bomb maker Ibrahim al-Asiri and his laboratory, but showed no interest in acting on that knowledge (Jordan, 2015). According to a Yemeni political analyst, the Political Security Organization has detailed knowledge of AQAP leaders’ whereabouts, which allows police to arrest dozens of operatives within hours as needed (Kimball, 2012). After al-Asiri’s bombs were used in attempted attacks on US airliners in 2009 and 2010, the US promised more aid to Yemen. “In retrospect, the planned attacks were a blessing for us”, a Yemeni official reflected. “They have alarmed the West” (Smoltczyk and Windfuhr, 2010).

A 2015 UN investigation found that Saleh explicitly agreed to cede territory to local AQAP emir Sami Dayan during the 2011 anti-regime protests. The report states that Saleh ordered Yemeni forces to withdraw from al-Bayda and Abyan governorates in exchange for a non-aggression pledge from AQAP, which allowed the group to occupy the territory unimpeded (United Nations, 2015: 19). This is consistent with a separate account from General Muhammad al-Sumali, a Yemeni military commander whose brigade was charged with defending Abyan governorate. According to the general, the governor, police, and Central Security Force (the counterterrorism force headed by Yahya Saleh) fled the city as 400 AQAP fighters—many of whom had been recently released from prison—took control using equipment reclaimed from the CSF. Neither did the Republican Guard (another paramilitary unit headed by the president’s son, Ahmed) respond when General Sumali called for reinforcements (Scahill, 2012).

International terrorism became more central to US foreign policy in the late 1990s. The evidence discussed above is consistent with the argument that the leadership of the Yemeni government deliberately cultivated terrorism, once Saleh realized that a burgeoning jihadist presence in Yemen could capture the attention of the US. Meanwhile, by making token arrests and half-hearted military operations, Saleh was consistently able to maintain plausible deniability and convince the US that his regime was not supporting AQAP. After granting the US access to Yemeni territory, Saleh even shifted blame for any subsequent attacks onto the Americans themselves: “I have given you an open door on terrorism, so I am not responsible” (Wikileaks, 2009).

Although Saleh was eventually forced from power in early 2012 as part of a GCC agreement, the strategy of counterterrorism racketeering enabled him to cling to power for well over a decade after opposition to his rule began. His unpopularity intensified over time, but he held firmly to power until a cascade of elite defections and an assassination attempt during the 2011 protests made it untenable for him to remain. It seems likely that by strengthening Saleh’s personal paramilitary units and sustaining his patronage network, US security and other assistance¹⁴ prolonged his rule. Perhaps this was deliberate. Speaking about the possible fall of Saleh, US Secretary of Defense Robert Gates admitted that “if that government collapses, or is replaced by one who is dramatically more weak, then I think we’d face some additional challenges out of Yemen, there’s no question about it. It’s a real problem” (Agence France Presse, 2011).

Examining this case serves two purposes. First, it gives us a closer look at what the various mechanisms in the argument look like in the real world, and how they translate into the statistical results I present. Yemen is a prototypical personalist dictatorship, heavily reliant on patronage to maintain a coalition of competing political factions, and is also home to a branch of al-Qaeda. Saleh knew that the US relied upon him, and felt that his importance was conditional on not ending the terrorist threat. US officials’ credulity (e.g. Wikileaks, 2010) and the informational asymmetry between the two countries allowed Saleh to present himself publicly as a reliable, if imperfect, counterterrorism ally, while covertly inflaming the problem. Other regimes have demonstrated similar behavior at times—Afghanistan, Egypt, Haiti, Iraq, Uganda, to name a few—but the Yemen case most starkly demonstrates the moving parts of the argument.

Second, it underscores a point made earlier regarding the obstacles to any systematic empirical attempt to directly evaluate the effect of aid on recipient counterterrorism effort. The actions of the government are extremely difficult to observe; in fact, it is often important that they are not observed. Even in Yemen, arguably the most well-known case of this type of behavior, publicly available knowledge about Saleh’s actions comes almost exclusively from leaked documents and informants. Thus, even if some measure of counterterrorism effort based on readily observable actions did exist, it would give a misleading picture of how security aid alters the behavior of the recipient. On the surface, the Saleh regime often appeared to cooperate quite closely with the US.

Conclusion

In this paper, I demonstrate why counterterrorism aid can be counterproductive in some cases. Borrowing arguments from the development aid and comparative authoritarianism literature, I argue that recipient regimes dependent upon patronage to maintain political power will benefit by inflating the level of terrorism within their borders in order to continue receiving aid. Using data on group duration and terrorist attacks, I find support for my hypotheses: higher average US aid levels lead to longer-lasting terrorist campaigns in personalist regimes. Similarly, increasing aid levels to personalist regimes leads to more terrorist attacks—especially anti-US attacks—in a given year.

These findings suggest several pathways for further exploration of the causal chain between foreign aid and counterterrorism, and where political institutions fit within that process. First, it would be worth expanding this argument to include all donors and recipients, as the US is not the sole security assistance donor. The current Yemeni regime may be playing a similar game with Saudi Arabia and their fear of the Houthi, a northern separatist group. Anecdotal evidence also suggests that the Assad regime—a sectarian dictatorship facing intense internal challenges—has deliberately cultivated the threat from the Islamic State by releasing jihadist prisoners in 2011, and focusing its counterinsurgency efforts on more moderate and non-violent opposition groups instead (Weiss and Hassan, 2015). The rapid growth of the Islamic State has resulted in intensified Russian and Iranian support for Assad, and less strident US calls for him to step down.

In order to fully uncover the effects of foreign aid on recipient counterterrorism behavior, a better definition of counterterrorism is also needed, as are measurements and data on this concept. This would go a long way toward enabling us to find out which moral hazard mechanism is operative in the context of counterterrorism: is the recipient government actively sponsoring terrorism in hopes of continued aid? Or does the promise of future aid lead a personalist leader to harsh repression, thus also increasing terrorism? Both mechanisms would lead to the same outcome, but they are two very different processes that represent divergent responses to aid. Addressing these questions is essential if the goal is to understand what foreign aid does to recipient behavior, and efforts on both the theoretical and empirical fronts are critical in working toward answers.

These findings also raise questions about the “counterterrorism partnership” approach that has become so central to recent US counterterrorism policy. Although I do find that aid is associated with increased terrorism in some cases, the cause lies not with the aid itself, but rather with the way it is delivered and the nature of the donor–recipient relationship. According to the theory, the key issues here are (a) the need for clear and credible conditions placed on aid packages and (b) the willingness to withdraw aid or otherwise sanction recipients when conditions are not met. These are difficult to implement for a multitude of reasons, not the least of which are donors’ inability to monitor recipient security efforts, and the American preference for a military approach to international counterterrorism that does not involve putting US boots on the ground. As the Yemen case shows, even when officials in the donor country are aware that a “partner” is insincere, the unwillingness to downgrade the relationship only exacerbates the moral hazard. The findings indicate that the if the US (and aid donors more broadly) wishes to continue using economic and military assistance to combat terrorism, it needs to be more vigilant about how aid is used by recipient governments, and to carry out threats to withdraw aid if certain benchmarks are not met.

Footnotes

Acknowledgements

Thanks to David Carter, Whan Choi, Doug Lemke, Jim Piazza, Mark Schafer, Joe Wright, and the IR group at University of Texas for their insights and helpful suggestions on earlier drafts. All remaining mistakes are mine. Supplementary materials, which include data, appendices, and replication files for the main analysis and an Online Appendix, can be found on SAGE’s Conflict Management and Peace Science website.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Notes

References

Agence France Presse (2011) US says post-Saleh Yemen would pose ‘real problem’, 27 March. Available from: https://ph.news.yahoo.com/us-says-post-saleh-yemen-pose-real-problem-20110327–071932–729.html.

Aksoy

Carter

Wright

(2015) Terrorism and the fate of dictators. World Politics 67(3): 423–468.

Al-Moshki

(2014) 459 inmates have escaped Yemeni prisons since 2006. Yemen Times, 27 February.

Azam

J-P

Delacroix

(2006) Aid and the delegated fight against terrorism. Review of Development Economics 10(2): 330–344.

Bandyopadhyay

Sandler

Younas

(2011) Foreign aid as counterterrorism policy. Oxford Economic Papers 63: 423–447.

Bapat

(2011) Transnational terrorism, US military aid, and the incentive to misrepresent. Journal of Peace Research 48(3): 303–318.

Bell

Sudduth

(2015) The causes and outcomes of coups during civil war. Journal of Conflict Resolution, doi:10.1177/0022002715603098.

Boone

(1996) Politics and the effectiveness of foreign aid. European Economic Review 40: 289–329.

Boutton

(2014) Interstate rivalry, US foreign aid, and incentives for counterterrorism cooperation. Journal of Peace Research 51(6): 741–754.

10.

Boutton

Carter

(2014) Fair weather allies: Terrorism and the selective allocation of US foreign aid. Journal of Conflict Resolution 58(7): 1144–1173.

11.

Brambor

Roberts Clark

Golder

(2006) Understanding interaction models: Improving empirical analysis. Political Analysis 14(1): 63–82.

12.

Bratton

van de Walle

(1997) Democratic Experiments in Africa: Regime Transitions in Comparative Perspective. Cambridge: Cambridge University Press.

13.

Brownlee

(2004) Ruling parties and durable authoritarianism. CDDRL Working Paper 23. Available from: http://iis-db.stanford.edu/pubs/20761/brownlee.working_paper.pdf.

14.

Bueno de Mesquita

Smith

(2009) A political economy of aid. International Organization 63: 309–340.

15.

Bueno de Mesquita

Smith

Siverson

Morrow

(2003) The Logic of Political Survival. Cambridge, MA: MIT Press.

16.

Burnside

Dollar

(2000) Aid, policies and growth. The American Economic Review 90(September): 847–868.

17.

Byman

(2006) Friends like these: Counterinsurgency and the war on terrorism. International Security 31(2): 79–115.

18.

Byman

(2008) The changing nature of state sponsorship of terrorism. Saban Center for Middle East Policy at the Brookings Institution Analysis Paper no. 16.

19.

Casey

(2016) FARC rebels in Colombia reach cease-fire deal with government. New York Times. Available from: http://www.nytimes.com/2016/06/23/world/americas/colombia-farc-peace-deal-rebels-cease-fire-santos.html.

20.

Clark

(2010) Yemen: Dancing on the Heads of snakes. New Haven, CT: Yale University Press.

21.

Collier

(2007) The Bottom Billion: Why the Poorest Countries are Failing and What can be Done About It. Oxford: Oxford University Press.

22.

Conrad

Young

(2014) Tyrants and terrorism: Why some autocrats are terrorized while others are not. International Studies Quarterly 58(3): 539–549.

23.

Davenport

(2007) State repression and the tyrannical peace. Journal of Peace Research 44(4): 485–504.

24.

Dietrich

(2013) Bypass or engage? Explaining donor delivery tactics in foreign aid allocation. International Studies Quarterly 57(4): 698–712.

25.

Djankov

Montalvo

Reynol-Querol

(2008) The curse of aid. Journal of Economic Growth 13: 169–194.

26.

Dodge

(2012) Iraq’s road back to dictatorship. Survival 54(3): 147–168.

27.

Easterly

(2006) The White Man’s Burden: Why the West’s Efforts to Aid the Rest have Done so Much Ill and so Little Good. New York: Penguin Press.

28.

Ellis

(2004) Briefing: The pan-Sahel initiative. African Affairs 103: 459–464.

29.

Escriba-Folch

(2013) Accountable for what? Regime types, performance, and the fate of outgoing dictators, 1946–2004. Democratization 20(1): 160–185.

30.

Escriba-Folch

Wright

(2010) Dealing with tyranny: International sanctions and the survival of authoritarian rulers. International Studies Quarterly 54: 335–359.

31.

Fleck

Kilby

(2010) Changing aid regimes? U.S. foreign aid from the Cold War to the war on terror. Journal of Development Economics 91: 185–197.

32.

Gandhi

Przeworski

(2006) Cooperation, cooptation, and rebellion under dictatorships. Economics and Politics 18(1).

33.

Geddes

(2003) Paradigms and Sandcastles: Theory Building and Research Design in Comparative Politics. Ann Arbor, MI: University of Michigan Press.

34.

Geddes

Wright

Frantz

(2014) Autocratic breakdown and regime transitions: A new data set. Perspectives on Politics 12(2): 313–331.

35.

Guerin

(2007) Another success for Detachment 88. Asia Times Online, 16 June. Available from: http://www.atimes.com/atimes/Southeast_Asia/IF16Ae01.html.

36.

Hendrix

Young

(2014) State capacity and terrorism: A two-dimensional approach. Security Studies 23(2): 329–363.

37.

International Crisis Group (2011) Indonesian jihadism: Small groups, big plans. Asia Report no. 204.

38.

Jackson

Rosberg

(1984) Personal rule: Theory and practice in Africa. Comparative Politics 16(4): 421–442.

39.

Johnsen

(2010) Al-Qaeda in the Arabian Peninsula In Yemen. NPR Interview with Neal Conan, Talk of the Nation, 4 November. Available at: http://www.npr.org/templates/story/story.php?storyId=131071948

40.

Johnsen

(2013) The Last Refuge: Yemen, Al-Qaeda, and America’s War in Arabia. New York: W.W. Norton.

41.

Jones

Libicki

(2008) How Terrorist Groups End: Lessons for Countering al Qaeda. Washington, DC: RAND Corporation.

42.

Jordan

(2015) Informant says Yemen’s Saleh helped direct al-Qaeda. Al Jazeera. Available from: http://www.aljazeera.com/news/2015/06/informant-yemen-saleh-helped-direct-al-qaeda-150604073415522.html.

43.

Kimball

(2012) Whose side is Yemen on? Foreign Policy. Available from: http://www.foreignpolicy.com/articles/2012/08/29/whose_side_is_yemen_on.

44.

Lake

(2011) Hierarchy in International Relations. Ithaca, NY: Cornell University Press.

45.

Licht

(2010) Coming into money: The impact of foreign aid on leader survival. Journal of Conflict Resolution 54(1): 58–87.

46.

Magaloni

(2006) Voting for Autocracy: Hegemonic Party Survival and its Demise in Mexico. New York: Cambridge University Press.

47.

Moss

Pettersson

van de Walle

(2006) An aid-institutions paradox? A review essay on aid dependency and state-building in Sub-Saharan Africa. Center for Global Development working paper.

48.

Phillips

(2011) Yemen terrorists are pawns of power. The Australian. Available from: http://www.theaustralian.com.au/news/world/yemen-terrorists-are-pawns-of-power/story-e6frg6ux-1226033569159?nk=5a6491c13731d60d80dc27a0f5db5697.

49.

Piazza

Wilson

(2013) Autocracies and terrorism: Conditioning effects of authoritarian regime type on terrorism. American Journal of Political Science 57(4): 941–955.

50.

Powell

(2014) Trading coups for civil war: The strategic logic of tolerating rebellion. African Security Review 23(4): 213–224.

51.

Powell

Thyne

(2011) Global instances of coups from 1950–present. Journal of Peace Research 48(2): 249–259.

52.

Reuters (2013) Mass jailbreak in Pakistan as Taliban gunmen storm prison, 30 July. Available from: http://www.reuters.com/article/2013/07/30/us-pakistan-prison-attack-idUSBRE96S0WG20130730.

53.

Roessler

(2011) The enemy within: Personal rule, coups, and civil war in Africa. World Politics 63(2): 300–346.

54.

Rosenberg

(2013) Afghan leader confirms cash deliveries by CIA. New York Times, 29 April. Available from: http://www.nytimes.com/2013/04/30/world/asia/karzai-acknowledges-cash-deliveries-by-cia.html.

55.

Rubin

(2010) Karzai’s words leave few choices for the west. New York Times, 4 April. Available from: http://www.nytimes.com/2010/04/05/world/asia/05karzai.html.

56.

Salman

Parker

(2013) The great escape: How al Qaeda broke hundreds of bad guys out of the worlds most notorious jail—and what it means for America. Foreign Policy, 5 August. Available from: http://www.foreignpolicy.com/articles/2013/08/05/the_great_escape_abu_ghraib_prison_al_qaeda.

57.

Scahill

(2012) Washington’s war in Yemen backfires. The Nation, 14 February.

58.

Schelling

(1966) Arms and Influence. New Haven, CT: Yale University Press.

59.

Siddiqa-Agha

(2001) Pakistan’s Arms Procurement and Military Buildup, 1979–1999. New York: Palgrave.

60.

Smith

(2008) The perils of unearned income. Journal of Politics 70(3): 780–793.

61.

Smith

(2005) Life of the party: The origins of regime breakdown and persistence under single-party rule. World Politics 57(3): 421–451.

62.

Smoltczyk

Windfuhr

(2010) East of Aden: A president struggles to keep Yemen together. Spiegel Online, 11 November. Available from: http://www.spiegel.de/international/east-of-aden-a-president-struggles-to-keep-yemen-together-a-728361.html.

63.

Sovey

Green

(2011) Instrumental variables in political science: A reader’s guide. American Journal of Political Science 55(1): 188–200.

64.

Staniland

(2012) States, insurgents, and wartime political orders. Perspectives on Politics 10(2): 243–264.

65.

START (2012) Global Terrorism Database. Available from: http://www.start.umd.edu/start/.

66.

Steinwand

(2015) Foreign aid and political stability. Conflict Management and Peace Science 32(4): 395–424.

67.

Svensson

(1999) Aid, growth, and democracy. Economics and Politics 11 (3): 275–297.

68.

Svensson

(2000) Foreign aid and rent-seeking. Journal of International Economics 51: 437–461.

69.

Svolik

(2012) The Politics of Authoritarian Rule. Cambridge: Cambridge University Press.

70.

Terza

Basu

Rathouz

(2008) Two-stage residual inclusion estimation: Addressing endogeneity in health econometric modeling. Journal of Health Economics 27(3): 531–543.

71.

Ulfelder

(2014) Coup forecasts for 2014. Dart-throwing Chimp, 25 January. Available from: http://dartthrowingchimp.wordpress.com/2014/01/25/coup-forecasts-for-2014/.

72.

United Nations (2015) S/2015/125. United Nations Security Council, 20 February.

73.

US Agency for International Development (2014) USAID Greenbook: U.S. overseas loans and grants. Available from: http://gbk.eads.usaidallnet.gov/data/.

74.

US Congress (1961) Foreign Assistance Act of 1961. 87th US Congress, 1st session, S.1707 PL 87–195. Available from: http://www.gpo.gov/fdsys/pkg/BILLS-111s962es/pdf/BILLS-111s962es.pdf.

75.

US General Accounting Office (2010) DOD and state need to improve sustainment planning and monitoring and evaluation for Section 1206 and 1207 Assistance Programs.

76.

Weiss

Hassan

(2015) ISIS: Inside the Army of Terror. London: Regan Arts.

77.

White House (2014) Remarks by the President at the United States Military Academy Commencement Ceremony. Office of the Press Secretary, 28 May. Available from: https://www.whitehouse.gov/the-press-office/2014/05/28/remarks-president-united-states-military-academy-commencement-ceremony.

78.

Wikileaks (2009) Brennan–Saleh meeting, 6 September

79.

Wikileaks (2010) Saleh sticks to his CT guns, but offers a way out to AQAP rank and file, 11 January.

80.

Wright

(2009) Aid effectiveness and the politics of personalism. Comparative Political Studies 43(6): 735–762.

81.

Wright

Winters

(2010) The politics of effective foreign aid. Annual Review of Political Science 13: 61–80.

82.

Young

Findley

(2011) Can peace be purchased? A sectoral-level analysis of aid’s influence on transnational terrorism. Public Choice 149(3/4): 365–381.

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.

7.43 MB

0.00 MB