Is There a Rebel Resource Curse? Ideological Appeals, Material Incentives, and the Success of Rebel Organizations

Abstract

How do recruitment tactics affect the success of rebel groups? Scholars posit that recruitment strategies that rely more on ideological appeals, relative to material incentives, attract highly committed recruits who are more invested in the success of their groups than their materially motivated counterparts. An implication of this is that rebel organizations should be more successful when they rely more on ideological appeals than material incentives for recruitment. However, I argue that ideologically based recruitment strategies provide rebels with a double-edged sword. While ideological appeals help attract committed recruits, they are also associated with factors that introduce a variety of problems into the bargaining process, undermining the ability of rebels to earn negotiated settlements. Furthermore, ideological-based recruitment strategies often fail to build sufficiently powerful fighting forces, making it difficult for rebels to achieve outright victory. Thus, I expect that ideologically based recruitment strategies do not help rebels achieve long-term success. I test this claim with novel data on the recruitment practices of rebel organizations that were active across the world between 1989 and 2011. I do not find evidence that greater reliance on ideological recruitment appeals, relative to material incentives, increases or decreases the probability that rebels achieve favorable outcomes.

Resumen

¿Cómo influyen las tácticas en materia de reclutamiento sobre el éxito de los grupos rebeldes? Los académicos postulan que las estrategias de reclutamiento que se basan más en atractivos ideológicos que en incentivos materiales tienden a atraer reclutas altamente comprometidos que están más interesados en el éxito de sus grupos que sus homólogos motivados por incentivos materiales. Una implicación de esto es que las organizaciones rebeldes deberían tener más éxito cuando utilizan atractivos ideológicos en mayor medida que incentivos materiales para el reclutamiento. Sin embargo, argumentamos que las estrategias de reclutamiento basadas en ideologías son un arma de doble filo para los rebeldes. Si bien es cierto que los llamamientos ideológicos ayudan a atraer reclutas comprometidos, estos también están asociados con factores que introducen una variedad de problemas en el proceso de negociación, debilitando la capacidad de los rebeldes de lograr acuerdos negociados. Además, las estrategias de reclutamiento basadas en ideologías, a menudo, no logran construir fuerzas combatientes lo suficientemente poderosas, lo que dificulta que los rebeldes logren una victoria absoluta. En consecuencia, hipotetizamos que las estrategias de reclutamiento basadas en ideologías no ayudan a los rebeldes a lograr el éxito a largo plazo. Ponemos a prueba esta hipótesis utilizando datos novedosos en materia de las prácticas de reclutamiento de las organizaciones rebeldes que estuvieron activas en todo el mundo entre 1989 y 2011. No encontramos evidencia alguna de que una mayor confianza en los incentivos ideológicos de reclutamiento, en comparación con los incentivos materiales, aumente o disminuya la probabilidad de que los rebeldes logren resultados favorables.

Résumé

Quel est l'effet des tactiques de recrutement sur la réussite des groupes rebelles? Les chercheurs postulent que les stratégies de recrutement qui font davantage appel à l'idéologie, par rapport aux motivations matérielles, attirent des recrues extrêmement motivées, qui s'impliquent davantage dans la réussite de leur groupe que leurs homologues aux motivations matérielles. Une implication en découle : les organisations rebelles devraient réussir davantage quand elles se reposent plus sur l'appel à l'idéologie que sur les motivations matérielles pour recruter. Néanmoins, j'affirme que les stratégies de recrutement qui se fondent sur l'idéologie sont à double tranchant pour les rebelles. Bien que l'appel à l'idéologie permette d'attirer des recrues impliquées, il s'assortit aussi de facteurs qui introduisent un éventail de problèmes dans le processus de négociation. Aussi les rebelles obtiendront-ils plus difficilement des accords négociés. En outre, les stratégies de recrutement fondées sur l'idéologie ne permettent souvent pas de bâtir des forces armées suffisamment puissantes, ce qui complique la victoire incontestée des rebelles. Par conséquent, je pense que les stratégies de recrutement fondées sur l'idéologie n'aident pas les rebelles à atteindre une réussite long terme. J’évalue cette affirmation à l'aide de données inédites sur les pratiques de recrutement des organisations rebelles actives entre 1989 et 2011 dans le monde entier. Je ne trouve pas d’éléments probants pour confirmer le fait qu'un recrutement fondé sur l'appel à l'idéologie, plutôt que sur des motivations matérielles, augmente ou diminue la probabilité d'issues favorables aux rebelles.

Introduction

In his groundbreaking works, Weinstein (2005, 2007) posited that material wealth is actually a curse for rebel organizations. While such resources can help organizations fund their operations, Weinstein argues that when rebels mobilize around material incentives, they tend to attract opportunistic, uncommitted individuals who seek short-term material gain, but who are not invested in helping militant organizations achieve their long-term goals. In contrast, Weinstein asserts that in the absence of economic endowments, armed groups can instead mobilize recruits around social endowments, such as shared ideologies and identities. Weinstein expects that because groups that mobilize around ideological and identity-based ties must make credible commitments about their ability to deliver benefits in the future, they tend to attract only highly committed individuals who are willing to forgo short-term gains in exchange for long-term benefits.

Based on the premise that material incentives attract recruits with perverse incentives, Weinstein posits that large economic endowments present a “curse” for rebel organizations. Much of the scholarship that builds off the idea focuses on how rebels that mobilize extensively around economic endowments tend to be more isolated from, and abusive toward, civilian populations (e.g., Weinstein 2005, 2007; Wood 2010; Sarkar and Sarkar 2017; Fortna, Lotito, and Rubin 2018; Walsh et al. 2018). However, another lesser-explored implication of Weinstein’s argument is that heavy reliance on material incentives for recruitment should undermine the long-term success of rebel organizations. Indeed, if materially motivated recruits care less about the success of their organizations and are more likely to desert and defect (Weinstein 2005, 2007), then it will be challenging for groups that organize around economic incentives to be successful.

There has been some acknowledgment of the potential deleterious effects that material-based mobilization could have on rebel groups’ long-term prospects. Indeed, Weinstein (2005, 600) writes that “while natural resources and other economic endowments initially appear as a blessing to would-be rebels, they attract recruits that are possibly ill-suited to the long-term goal of capturing state power.” Weinstein (2005, 603) further elaborates that “At the early stages of rebellion, the presence of uncommitted soldiers can irreparably harm a movement and lead to its quick defeat. Hence, rebel leaders wish to recruit high-commitment as opposed to low-commitment individuals.”

However, since this initial connection was made, there has been a dearth of work that theoretically and empirically unpacks the connection between rebel groups’ recruitment tactics and their long-term success. This contrasts with the growing body of work that examines how the broader ideologies of rebel organizations (but not their recruitment strategies specifically) affect their longevity and success (e.g., Svensson 2007; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022). In response, I develop a theoretical argument that addresses how reliance on ideological recruitment appeals, relative to material incentives, affects the success of rebel organizations. I then evaluate the association between mobilization strategies and rebel group outcomes with systematically collected, cross-group data on rebel recruitment practices.

I posit that previous work has overlooked some of the distinct drawbacks of ideological-based recruitment strategies and that, overall, such tactics do not provide rebels with a distinct advantage relative to those that rely more heavily on material incentives for recruitment. Specifically, I argue that heavy reliance on ideological recruitment appeals undermines the ability of rebels to both earn negotiated settlements and achieve outright victory over the governments they are fighting.

I highlight how both ideological and material recruitment appeals are associated with features of rebel organizations that complicate the bargaining process, making it difficult for rebels to enter negotiations and achieve concessions. First, I posit that both ideological- and material-based recruitment strategies are associated with greater capacity on some dimensions, but weaker capacity on others, which creates uncertainty about their capabilities. This complicates the bargaining process, making it difficult for rebels to enter negotiated settlements. Furthermore, because neither material nor ideological-based recruitment strategies have a clear advantage in improving rebel groups’ capacity, neither will help increase militants’ probability of total victory over government forces.

Second, for various reasons, groups that rely more on ideological appeals and those that recruit more heavily with material incentives both have difficulty credibly committing to abiding by peace agreements, making it difficult for them to reach negotiated settlements (e.g., Fearon 1995). Third, groups that rely heavily on ideological appeals, and subsequently attract highly ideologically committed individuals, might increase the probability that issue indivisibilities plague negotiations, making it challenging for such groups to reach peace agreements.

Importantly, I am not arguing that groups will be more successful if they employ material recruitment incentives. Indeed, these groups tend to be plagued by a lack of commitment, desertion, defection, and in-fighting among the ranks (e.g., Weinstein 2005, 2007). Rebels that pursue material gain are also often unwilling to compromise, as they can profit more from wartime activities than peace settlements (Fearon 2004). These issues can make it difficult for rebels that heavily rely on material incentives to either enter negotiated settlements or achieve outright victory over government forces. Rather, my central argument is that there are empirical and theoretical reasons to believe that ideological-based mobilization strategies, relative to materially centered ones, do not significantly improve rebels’ chances of success. Instead, as the results of this study indicate, rebel group capacity is a much stronger predictor of rebel group success than their recruitment tactics.

Using data from the Rebel Appeals and Incentives Dataset (RAID) (Soules 2023) and the Uppsala Conflict Data Program’s (UCDP) Conflict Termination Dataset (Kreutz 2010), I quantitatively test the association between rebel group recruitment tactics and the outcomes they experience. Despite the implication in prior literature that recruitment strategies affect the success of militant movements, I do not find a statistically significant association between the degree to which groups rely on ideological appeals, and the probability that they achieve favorable outcomes, such as negotiated settlements or total victory. Additionally, across a variety of alternative tests, I do not find support for the idea that material-based recruitment strategies significantly undermine the success of rebel organizations.

This paper makes at least two contributions. First, it makes a theoretical contribution by discussing the ways in which ideological- and material-based recruitment strategies can both undermine the success of armed organizations. Specifically, it challenges the underexplored implication of prior work that material-based recruitment strategies uniquely harm rebels’ ability to be successful. Instead, I posit that there are also many disadvantages to relying on ideological appeals for recruitment and that these recruitment strategies ultimately do not make groups more successful. Again, I am not arguing that it is more beneficial for groups to recruit with material incentives, as there are many disadvantages associated with these appeals as well (e.g., Weinstein 2005, 2007). However, the benefits of nonmaterial recruitment strategies have been overstated in some prior work.

Existing work has examined how variation in rebel ideology, as well as ideological incompatibilities between rebels and governments, helps explain the long-term success of militant movements (e.g., Svensson 2007; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022). While valuable, this work does not evaluate the claim that material-based recruitment strategies are more detrimental to rebels’ long-term success than ideological-based mobilization strategies. Relatedly, I am not arguing that there is no important variation in how different ideologies affect the success of armed organizations. Instead, I am assessing variation in the extent to which groups rely on ideological appeals or material incentives for recruitment.

Second, this paper provides, to the best of my knowledge, the first cross-rebel group quantitative analysis of how the material and ideological recruitment appeals rebels employ affect the outcomes they experience. While valuable data exist on the ideologies of rebel organizations (e.g., Polo and Gleditsch 2016; Wood and Thomas 2017; Basedau, Deitch, and Zellman 2022), and their material resources (e.g., Sawyer, Gallagher Cunningham, and Reed 2017; Walsh et al. 2018), these datasets do not capture the actual mobilization strategies of rebel organizations. This is significant because the material and ideological resources of militant organizations do not always translate into their recruitment strategies (e.g., Herbst 2000; Soules 2023). Thus, this study provides more direct evidence of the consequences of recruitment strategies than other cross-group quantitative analyses on how ideological and material resources affect rebel groups’ longevity and success (Svensson 2007; Sawyer, Gallagher Cunningham, and Reed 2017; Conrad et al. 2019; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022).

In the next section, I will discuss what “success” is in the context of this paper. Following this, I review the literature on the consequences of rebel group recruitment strategies to highlight that a testable implication of existing literature is that groups will have a higher likelihood of success when they rely more on ideological appeals than material incentives for recruitment. Next, I highlight a variety of drawbacks of ideological-based recruitment strategies that challenge this implication. I then present the research design and results. Following this, I attempt to empirically test some of the proposed theoretical mechanisms by examining the association between rebel recruitment tactics and organizational capacity. After this, I conduct a variety of robustness checks. I conclude by discussing the implications of the findings.

Defining Success

In analyzing how recruitment strategies affect rebel success, it is important to first define success. In the context of civil war, success could mean a variety of achievements, including the retention of troops or the successful execution of military operations, among other outcomes. However, to analyze the full extent to which recruitment tactics help or hinder rebels, I argue that we must understand how they affect the ability of groups to achieve their goals, as military success is the central goal of all rebel organizations (Gates 2002, 112).

Indeed, as noted above, Weinstein (2005) also predicts that groups that mobilize predominately around material incentives will struggle to achieve their long-term goals and face a significant risk of defeat. Building off prior work on rebel success, I examine how recruitment tactics affect the probability that rebels experience favorable outcomes, such as entering peace agreements or achieving total victory over government forces, or unfavorable outcomes, such as suffering total defeat by government forces or by ending through low levels of activity (e.g., Cunningham, Gleditsch, and Salehyan 2009; Kreutz 2010; Fortna 2015; Braithwaite and Ruiz 2018; Greig, David Mason, and Hamner 2018; Silverman, Acosta, and Huang 2023).

Benefits of Ideological Recruitment

Scholars have detailed multiple benefits associated with relying more on ideological appeals than material incentives for recruitment. First, as noted above, ideological-based recruitment strategies are expected to attract highly committed individuals who are willing to forgo short-term benefits for long-term gains (Weinstein 2005, 2007). Consequently, ideologically committed recruits are expected to be less likely to desert and defect (e.g., Gutiérrez-Sanín 2008; Oppenheim et al. 2015; Altier et al. 2017; Riley and Schneider 2022).

Second, ideological-based recruitment strategies can help improve rebel organizations’ popularity. Indeed, nonviolent propaganda by militant movements that includes ideological and grievance-based messaging helps increase support for the group (Mitts, Phillips, and Walter 2022). Third, Humphreys and Weinstein (2006) posit that materially driven recruits are more likely to pursue personal, material gain over actions that benefit their groups. As a result, they are more likely to engage in abusive behavior toward civilians, particularly when it facilitates the acquisition of material resources (e.g., looting). In contrast, they expect that ideologically motivated recruits will be more likely to engage in activities that benefit the group as a whole and are less interested in personal enrichment.

Thus, an expectation in the existing literature is that rebel groups that rely more on ideological appeals, relative to material incentives, for recruitment, will be more likely to attract highly committed recruits who are less likely to desert and defect and who work for the good of the group. Furthermore, at least certain types of ideological appeals are expected to increase the popularity of rebel groups. This all suggests that groups should be more likely to achieve their long-term goals if they have recruits who are more committed to the cause and who engage in behavior to benefit their group as a whole. Specifically, a testable implication derived from this literature is that: 

However, there are several disadvantages to ideological-based recruitment strategies that have been overlooked in prior scholarship. I posit that such problems make it difficult for rebel groups to either earn negotiated settlements or achieve outright victory. I turn to these issues in the next section.

Bargaining Issues and Negotiated Settlements

Private information about the capabilities and resolve of actors, incentives to misrepresent this information, credible commitment problems, and issue indivisibilities all undermine the ability of actors to reach peaceful negotiated settlements instead of fighting (Fearon 1995). Prior scholarship has examined how a variety of types of rebel ideologies heighten these bargaining issues (e.g., Svensson 2007; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022). However, again, such work has not compared the effects of ideological and material-based recruitment strategies on the ability of rebel organizations to earn negotiated settlements.

Uncertainty about Rebel Group Capacity

Uncertainty about the military capacity of opponents creates issues for the bargaining process. This uncertainty means that actors could underestimate or overestimate the strength of their opponents, leading them to inefficiently bargain (Fearon 1995). For the reasons detailed below, I expect that both ideological- and material-based recruitment strategies can generate significant uncertainty about the capabilities of rebel organizations.

According to Weinstein (2005, 2007), rebels that mobilize around ideological and identity-based ties often do so in the absence of substantial economic endowments. While groups with strong social endowments might be able to attract highly committed recruits, the dearth of economic resources could also make it difficult for such organizations to fight effectively enough to coerce governments into making concessions or to defeat them outright. Indeed, strong rebel organizations with substantial resources, including the ability to procure arms, are more likely to swiftly achieve favorable outcomes in civil wars (Cunningham, Gleditsch, and Salehyan 2009). Thus, even if groups that have substantially more social endowments than economic endowments attract more committed recruits, the former might not have the same material capacity to fight government forces as the latter.

An important caveat is that I am not arguing that governments directly interpret groups’ use of ideological recruitment appeals as a signal of reduced material capacity. Rather, I am positing that the dearth of material resources that such groups have will translate to their battlefield performance, which governments can more easily interpret (Mattes and Savun 2010). However, mobilizing predominantly around ideological appeals can help attract widespread popular support, which governments could also interpret as a sign of strength.

Specifically, rebels’ ideological platforms might help mobilize civilians with significant grievances, particularly those who have not previously felt well-represented by other dissident movements (e.g., Tokdemir et al. 2021; Mitts, Phillips, and Walter 2022; Schwab 2023). Indeed, public and highly visible ideological propaganda can help rebels gain significant support (Mitts, Phillips, and Walter 2022). Thus, the successful formation of a rebel group around ideological appeals might signal to the government that they are facing an aggrieved public, which could make them view the rebels as more of a threat.

Furthermore, ideological platforms can help rebels gain support from ethnic and religious diaspora communities, as well as ideologically sympathetic foreign governments (e.g., left-wing foreign governments sponsoring Marxist rebel groups) (e.g., Kalyvas and Balcells 2010; Salehyan et al. 2011; Nilsson and Svensson 2021). Indeed, the potential for external support from ideologically aligned actors creates uncertainty about the capacity of rebels, which can prolong conflicts (Nilsson and Svensson 2021). This argument is less about the actual external support rebels receive, and more about their potential to receive it in the future, as this generates uncertainty about the capabilities of militant groups (Nilsson and Svensson 2021).

For these reasons, governments might have a difficult time determining the capacity of rebel organizations and their corresponding level of threat. However, there can also be significant ambiguity surrounding the capacity of groups that mobilize primarily around material incentives. Related to the argument above, groups with significant economic endowments are often effective at sustaining their rebellions because they can pay for soldiers, weapons, other supplies, etc. (e.g., Conrad et al. 2019). These resources can translate to battlefield performance that signals the strength of groups that mobilize around economic endowments.

However, recruiting primarily with material incentives might also be associated with factors that signal the weakness of rebel groups. Indeed, there is evidence to suggest that materially motivated rebels are more likely to desert and defect than their ideologically driven counterparts (Oppenheim et al. 2015). Thus, a potential consequence of recruiting with material incentives is that it results in higher rates of troops deserting and defecting, which can signal rebel weakness.

Tying these issues together, both material- and ideological-based recruitment strategies can send unclear signals about the capacity of rebel organizations. Uncertainty about the capability of actors can make it difficult to reach a peace agreement during civil war, as it can be challenging to determine how long one’s adversary is able to hold out for (Walter 1999). Thus, information asymmetries can present a barrier to rebels achieving a favorable outcome through negotiated settlement.

Credible Commitments

Conflicts become difficult to resolve if all sides cannot credibly commit to abiding by the terms of a peace agreement (Fearon 1995; Walter 1997). Conflict actors have a difficult time credibly committing that they will uphold their end of a peace agreement if they have the incentive to renege (Fearon 1995). I expect that there are characteristics of both ideological- and material-based recruitment strategies that can exacerbate credible commitment problems.

In broad terms, rebel movements with a higher percentage of ideologically motivated recruits will have a difficult time credibly committing that they will be satisfied enough by a compromise to abide by the terms of a settlement. Ideologically committed recruits are particularly likely to defect in response to groups not maintaining their original ideological goals (Oppenheim et al. 2015). Keels and Wiegand (2020) argue that when ideological issues are particularly salient and polarizing in a conflict, rebel leaders have a difficult time credibly committing to peace. This is because leaders could alienate their cadres if the former perceives the latter to be selling out by accepting certain compromises.

Furthermore, even if rebels initially accept some concessions, governments might fear that the rebels will renege and pursue more concessions in the future. This is particularly salient for groups that rely heavily on ideological recruitment appeals, given evidence that the most ideologically committed recruits are the ones more likely to remobilize (Mironova, Alhamad, and Whitt 2020). Thus, because they have difficulty credibly committing to peace, groups that mobilize extensively around ideological appeals will have difficulty achieving long-term success because governments will not want to compromise with them (undermining the opportunity to gain concessions).

However, groups that mobilize around economic endowments also have characteristics that can make it difficult for them to credibly commit to peace. Indeed, groups that profit substantially from wartime activities (e.g., looting, taxing, etc.) can have a difficult time credibly committing to peace, as they might accrue more material benefits during wartime than during periods of peace (e.g., Fearon 2004). Thus, groups that organize around economic resources might also struggle to garner favorable outcomes through negotiations because of the difficulties they have credibly committing to peace.

Issue Indivisibilities

Another salient problem that ideological-based recruitment introduces to the bargaining process is the exacerbation of issue indivisibilities. Fearon (1995) explains that issue indivisibilities arise when (1) disputing parties view any compromise on an issue as too drastically diminishing its value to make compromise worth it and/or (2) there are no other concessions that can be made in other areas to substitute for proposed concessions on an issue that at least one side views as indivisible.

Indeed, Svensson (2007) argues that religious ideologies in civil wars exacerbate both core problems associated with issue indivisibilities. Recruits with religious (Hassner 2003; Svensson 2007; Toft and Zhukov 2015; Nilsson and Svensson 2020; Basedau, Deitch, and Zellman 2022) and/or (ethno)nationalist motives (Hasner 2003; Goddard 2006; Wucherpfennig et al. 2012) particularly view certain issues as indivisible. Relatedly, the opposition will often crack down hard on rebels that are able to effectively use ideology to mobilize recruits out of the fear that such rebels are unwilling to compromise but are effective and devoted (Balcells and Kalyvas 2015; Hafez 2018). This poses further obstacles to groups achieving their goals.

We see similar evidence at the level of individual recruits. In the Syrian Civil War, ex-fighters that were ideologically committed to fighting the Assad regime or establishing an Islamic state were more likely to remobilize (Mironova, Alhamad, and Whitt 2020). These recruits are also more likely to remobilize when they believe the group will pursue total victory over negotiated settlements (Mironova, Alhamad, and Whitt 2020). Other studies also present individual-level evidence that ideologically motivated recruits are less likely to desert or defect (Gutiérrez-Sanín 2008; Oppenheim et al. 2015; Altier et al. 2017). While this unwillingness to surrender might seem like a benefit for rebel groups on the surface, it can also mean that ideologically driven recruits are less likely to support compromise and/or accept offers of amnesty, making it difficult for their organizations to eventually be granted concessions as part of the negotiating process.

A potential implication of this discussion is that the presence of issue indivisibilities actually means that groups that rely more heavily on ideological appeals will be less likely to enter negotiated settlements with governments. However, as noted above, I still expect that mobilizing around material incentives will often result in rebel groups being unwilling to compromise, as they acquire more material resources through wartime operations than by compromising with governments (Fearon 2004). Furthermore, some ideologies (e.g., religious and ethnonationalist beliefs) are more likely to frame certain issues as indivisible than others. This would suggest that the type of ideology matters more than the extent to which groups rely on ideological appeals, relative to material incentives, for recruitment. Issue indivisibilities are still important to consider, however, as they might help explain why groups that heavily mobilize around ideological appeals do not have a distinct advantage in their efforts to earn peace agreements.

Mobilizing Sufficient Support

Many of the aforementioned issues associated with ideological recruitment undermine rebels’ ability to bargain, and thus, their subsequent ability to achieve negotiated settlements. It is possible, however, that while ideologically driven groups might be less likely to negotiate (and thus win at least some concessions), they will fight hard enough to achieve outright victory. However, I also expect that groups will struggle to achieve total victory when they mobilize heavily around ideological appeals. While ideological appeals might attract highly committed recruits, they do not necessarily sufficiently strengthen the group enough to produce total victory, as ideological appeals do not necessarily have a distinct advantage over material incentives in their ability to mobilize powerful fighting forces.

Indeed, as discussed above, groups with extensive social endowments often lack significant economic resources (Weinstein 2005, 2007), which can make it difficult for them to effectively fight government forces. Thus, more committed recruits do not necessarily translate to the resources needed (e.g., weapons and other supplies) to achieve total military victory. Relatedly, groups with extensive economic endowments might have the resources needed to achieve victory on the battlefield, but they are more likely to be plagued by low internal cohesion and high levels of disobedience, desertion, and defection among the troops (Weinstein 2005, 2007). Consequently, neither recruitment strategy provides a clear advantage for groups’ efforts to achieve total victory over government forces.

Ideological Recruitment and Rebel Success

Again, an implication stemming from the prior literature is that rebel groups, on average, will be more successful when they recruit more heavily with ideological appeals than material incentives. However, I do not expect to find support for this argument. Tying together the various dynamics discussed earlier, both material- and ideological-based recruitment strategies have various advantages and disadvantages, resulting in neither type of mobilization tactic substantially increasing (or decreasing) the probability of rebel group success.

As noted earlier, ideologically driven recruits, relative to their materially motivated counterparts, tend to be more committed to the cause and, consequently, are less likely to desert or defect and are more likely to behave in ways that benefit their organizations. Thus, the devotion and discipline of ideologically motivated recruits can provide a distinct benefit for rebel organizations (Weinstein 2005, 2007).

However, I also posit that ideological-based recruitment strategies, and the ideologically motivated recruits they tend to attract, will introduce a variety of problems to the bargaining process that undermine the ability of rebel groups to garner negotiated settlements. Again, groups that heavily rely on ideological appeals tend to have a dearth of material resources, which can undermine their battlefield performance (e.g., because they have a more difficult time acquiring arms). Consequently, I expect that groups that mobilize heavily around ideological appeals do not have a clear advantage on the battlefield, because while they have highly devoted recruits, they also tend to lack substantial material resources. This lack of clear battlefield advantage creates uncertainty about rebel group capabilities, which can undermine the bargaining process (Mattes and Savun 2010). Furthermore, as discussed above, the presence of a high number of ideologically committed recruits makes it challenging for rebel movements to credibly commit to peace agreements. This ideological devotion can also create issue indivisibilities, further undermining the possibility of achieving negotiated settlements. Again, I also expect that ideologically driven rebel groups will not be more likely to achieve total victory, as they often lack material resources to be consistently successful on the battlefield.

Thus, ideological-based recruitment strategies have important advantages and disadvantages that ultimately do not clearly help, nor clearly hinder, rebels’ ability to achieve their goals. Said differently, the advantages and disadvantages of ideological-based recruitment strategies somewhat balance each other out. Indeed, the ideological devotion that drives recruits to be more disciplined and less likely to desert and defect can also exacerbate bargaining issues, such as credible commitment problems and issue indivisibilities. Furthermore, the dearth of economic endowments that encourages forming strong ideological platforms to mobilize troops, also means that such groups can struggle on the battlefield due to their lack of material resources.

However, I do not expect that groups will be more (or less) successful if they rely heavily on material incentives for recruitment. As detailed earlier, groups that mobilize extensively with material appeals also tend to have features that disrupt the bargaining process (i.e., inconsistent battlefield performance and difficulty credibly committing to peace agreements over continuing to acquire wartime profits), making it hard for these groups to win concessions through negotiated settlements. The high rates of disobedience, desertion, and defection that these organizations face (e.g., Weinstein 2005, 2007; Oppenheim et al. 2015) also dampen the chances that groups have for total victory. These groups, however, tend to have much more extensive material resources, which provide them with some advantages. Consequently, similar to my argument about heavy reliance on ideological appeals, the combined advantages and disadvantages of material-based mobilization also balance each other out in a way that means such recruitment tactics do not have a clear effect on rebel group success.

Thus, despite the arguments by some scholars that rebel recruitment strategies significantly influence the quality of recruits they attract (e.g., Weinstein 2005, 2007; Oppenheim et al. 2015; Sarkar and Sarkar 2017), I do not expect that the extent to which groups rely on ideological or material appeals for recruitment will have a significant impact on their long-term success. This is because both types of recruitment strategies have associated (dis)advantages, rendering both as neither a more (nor less) effective mobilization tactic. Instead, the overall strength of rebel organizations (e.g., Cunningham, Gleditsch, and Salehyan 2009) and their subsequent ability to impose costs on the government (e.g., Thomas 2014), among other factors, are likely to play a more significant role in explaining rebel group success. This leads me to the hypothesis that: 

In the next section, I thus investigate how the extent to which groups rely on ideological appeals, relative to material incentives, affects the outcomes they experience. I compare this factor to other commonly proposed correlates of rebel groups success, such as organizational capacity (e.g., Cunningham, Gleditsch, and Salehyan 2009).

Research Design

Sample

To test how groups’ reliance on ideological and material recruitment appeals affects their long-term success, I rely on data from the RAID, which provides information about the recruitment strategies of rebel organizations (Soules 2023). RAID covers a global sample of 232 rebel movements that were active at some point between 1989 and 2011. Organizations can enter the dataset before 1989 and/or leave after 2011; they just had to have been active at some point during this period to enter RAID. The sample of actors in RAID is taken from the non-state actor (NSA) dataset (Cunningham et al. 2013). While other valuable datasets exist that cover the ideological and/or material resources of rebel organizations (e.g., Polo and Gleditsch 2016; Wood and Thomas 2017; Walsh et al. 2018; Braithwaite and Cunningham 2020; Basedau, Deitch, and Zellman 2022), RAID is the only dataset, to the best of my knowledge, that contains measures that directly capture the recruitment practices of militant groups.

Dependent Variable and Estimation Strategy

To test both hypotheses, I employ data on the ways in which rebel groups end. Following the precedent of several studies on civil war termination (e.g., Fortna 2015; Gurses 2015; Greig, David Mason, and Hamner 2018; Phayal, Mason, and Gurses 2019), I use data from the UCDP Conflict Termination Dataset (version 3-2021), updated from the original version developed by Kreutz (2010). This dataset contains information on multiple types of outcomes, including whether (1) a rebel group ends because of a peace agreement or ceasefire, (2) a rebel group achieves total victory, (3) a rebel is defeated by government forces, or (4) the group ends through low levels of activity (i.e., “fizzling out”). The former two outcomes are considered favorable outcomes, while the latter two are viewed as unfavorable (Fortna 2015).

I employ competing-risk duration models because they allow for analysis of the probability that conflicts end in specific ways, relative to other potential types of outcomes (e.g., Fortna 2015; Phayal, Mason, and Gurses 2019, 490). These models are also helpful because the data are right-censored, as not all groups have terminated by the end of the analysis period, and because the models help address issues related to temporal dependence. Additionally, the duration and outcomes of conflicts vary together and thus are important to model together (e.g., Brandt et al. 2008). Similar to Wood and Allemang (2022), I use a dyad-episode unit of analysis because the main explanatory variable, as noted below, is time-invariant. While there are 232 groups in RAID, there are 424 observations in the sample because some rebel organizations engaged in multiple conflict episodes (Kreutz 2010) (some of these 424 observations are dropped in the analysis because of missing data for some of the control variables). The models capture the time, in years, between the beginning of the episode and its specific form of termination.

Independent Variable

The main explanatory variable of this study is taken from RAID and is a five-point ordinal measure of the extent to which groups rely on ideological appeals, relative to material incentives, for recruitment. Specifically, this scale captures groups that rely entirely on material incentives (0), mostly on material incentives and very little on ideological appeals (1); a relatively even combination of ideological and material appeals (2); mostly on ideological appeals and very little on material incentives (3); or entirely on ideological appeals (4). Thus, higher values of the variable indicate greater reliance on ideological appeals, while lower values translate to greater reliance on material incentives. Greater values of this variable should translate to a higher percentage of members being ideologically committed, which some previous scholarship implies should be associated with an increased probability of success for such groups.

To construct RAID, detailed qualitative narratives were written about the recruitment practices of all groups in the sample. These narratives included not only details on the specific types of recruitment appeals employed by groups, but also the relative frequency at which they used them. Based on this information, it was first determined whether the group used only ideological appeals or only material incentives for recruitment, or if they employed both. If only one broad type of appeal was used, the group was coded as being on one of the respective far ends of the spectrum. If the group made some combination of material and ideological appeals, it was then determined whether the group employed both at the same relative frequency (2 on the ordinal scale) or if one type of appeal was used more frequently or systematically than the other (1 or 3 on the ordinal scale) (Soules 2023). This variable is time-invariant due to difficulties in gathering detailed information on the recruitment practices of rebels (Soules 2023).

Control Variables

I hold a variety of potentially confounding factors constant. First, the ideological foundations of rebel groups help shape their recruitment strategies (Weinstein 2005, 2007). Rebels with certain ideologies, particularly illiberal ones, are more likely to wage intractable conflicts and experience unfavorable outcomes (e.g., Svensson 2007; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022). Thus, using data from both the Foundations of Rebel Group Emergence (FORGE) dataset (Braithwaite and Cunningham 2020) and the Women in Armed Rebellion Dataset (Wood and Thomas 2017), I include two separate binary indicators: one measuring whether the group has a radical Islamist ideology and the other if it has any kind of left-wing ideology. Given that including these measures along with the main explanatory variable could lead to issues associated with multicollinearity, I conduct robustness checks in which I exclude these two control variables. The core findings do not change. Relatedly, secessionist conflicts tend to be difficult to resolve (e.g., Fearon 2004) and rebels’ goals might also shape their recruitment strategies. Thus, using data from sources including WARD and FORGE, I control for a dichotomous indicator of whether a group has secessionist aims.

Rebels’ capacity, including their natural resource wealth and external patronage, also affects their recruitment strategies (e.g., Weinstein 2005, 2007), as well as their success (e.g., Cunningham, Gleditsch, and Salehyan 2009). I control for several variables to account for these dynamics. Using data from the NSA dataset, I include a five-point ordinal measure that captures rebel groups’ strength relative to the governments they are fighting. Additionally, with data from Sawyer, Gallagher Cunningham, and Reed (2017), I include a binary measure of whether an external third party provides any form of support to the group. Furthermore, using data from the Rebel Contraband Dataset (Walsh et al. 2018), I include a dichotomous measure of whether a group exploited natural resources for profit. Finally, state capacity and regime type also affect the operations of rebel groups and the outcomes they experience (Cunningham, Gleditsch, and Salehyan 2009). Thus, I control for the measure of electoral democracy in the Varieties of Democracy (V-Dem) dataset (Coppedge et al. 2024). I also include a logged measure of the country’s per capita GDP, with data from Fariss et al. (2022).

Results

The results are displayed in Table 1. The standard errors are clustered on the rebel group in every model, and the hazard ratios are reported. As the results show, I do not find support for the implication derived from previous literature that rebel groups will be more likely to achieve favorable outcomes, such as negotiated settlements and total victory, when they rely more on ideological appeals than material incentives for recruitment (Hypothesis 1). Instead, in line with Hypothesis 2, I do not find evidence that reliance on ideological recruitment appeals, relative to material incentives, increases or decreases the probability of rebel group success.

In the models for the favorable outcomes (negotiated settlements and rebel victory), the measure of reliance on ideological appeals has hazard ratios above one, indicating that these groups are more likely to achieve these outcomes. However, this association never reaches statistical significance. There is also not a statistically significant association between reliance on ideological appeals and the probability of experiencing unfavorable outcomes (government victory and fizzling out). Thus, I did not find evidence that groups are more or less likely to be successful when they rely more on ideological appeals than material incentives for recruitment.

The other covariates suggest that other features of rebel organizations have a more substantial effect on their long-term success. Consistent with prior scholarship, rebel groups that have higher relative levels of power are more likely to achieve both peace agreements and total victory over government forces (Cunningham, Gleditsch, and Salehyan 2009). Relatedly, rebel groups that receive support from external actors have a higher likelihood of earning peace agreements. The results also suggest that such organizations are more likely to stave off total defeat by government forces but that such groups are also less likely to defeat government forces. The latter result is surprising, though it might be driven by the fact that state sponsorship can lead to harsher counterinsurgency responses toward recipient groups, decreasing the probability that militants are victorious (Carter 2012). Rebels with natural resource funding are also more likely to totally defeat government forces.

Thus, overall, stronger rebels with more resources have a higher likelihood of achieving favorable outcomes. As discussed above, rebel groups that lack economic resources are often the ones that mobilize heavily around ideological appeals (Weinstein 2005, 2007). Given the above evidence that stronger, better-resourced groups are more likely to achieve favorable outcomes, groups that mobilize heavily around ideological appeals might have a difficult time achieving long-term success. However, again, high rates of disobedience, desertion, and defection among groups that mobilize around material incentives can also undermine rebel group success (Weinstein 2005, 2007). Ultimately, both material- and ideological-based recruitment strategies have potential disadvantages that can harm the prospects of rebel organizations. Instead, strength and resources appear to be better predictors of rebel group success.

Consistent with existing literature, the results also suggest that radical Islamist groups are less likely to achieve peace agreements (Svensson 2007). This finding highlights another potential issue with the implication in prior literature that greater reliance on ideological appeals will increase the probability of rebel success. Specifically, this literature ignores how some ideologies might complicate the bargaining process more than others, undermining rebels’ efforts to achieve their goals. Said differently, making the assumption that ideological appeals are better for rebels’ long-term success than material incentives ignores important variation in ideology that affects their longevity and success (Svensson 2007; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022).

Rebel Recruitment and Strength

In this section, I attempt to empirically examine the aforementioned mechanism that both ideological- and material-based recruitment strategies have associated characteristics that send unclear signals about the strength of rebel organizations. On the one hand, groups that rely heavily on ideological appeals often lack significant material resources (Weinstein 2005, 2007), and weaker groups struggle to achieve favorable outcomes (Cunningham, Gleditsch, and Salehyan 2009). However, on the other hand, mobilizing around ideological appeals can increase popular support for groups (Mitts, Phillips, and Walter 2022), and they draw in recruits who are less likely to desert or defect (Weinstein 2005, 2007).

To examine the association between rebel recruitment tactics and organizational capacity, I conduct a series of cross-sectional analyses of the groups in RAID to unpack this association further. More details about the analyses are available in the online appendix. I use several measures of rebel strength from the aforementioned NSA dataset (Cunningham et al. 2013). The NSA dataset contains two measures that help capture rebels’ ability to perform on the battlefield. The first is an ordinal indicator of the ability of rebels to procure arms, relative to the government, and the second is an ordinal measure of rebels’ fighting capacity, relative to the government. If rebel groups that rely on ideological appeals really are poorly resourced (Weinstein 2005, 2007), then they might have difficulty acquiring the arms and achieving the overall fighting capacity necessary to effectively fight government forces.

The NSA dataset also contains an ordinal measure of the ability of rebel groups to mobilize personnel relative to the government. It is possible that the popular appeal provided by mobilizing around ideological appeals will make rebels more effective at mobilizing personnel, which could contribute to their long-term success. All three ordinal measures capture low, moderate, and high capacity on these different dimensions. I transform these measures into binary indicators of whether a group has moderate or high levels of capacity on the dimension in question. I do so because a very small percentage of groups are coded as having a high capacity on any of the dimensions (e.g., while approximately 32 percent of the sample is coded as having a moderate capacity to procure arms, only about 0.92 percent are classified as having a high capacity to do so). I conduct a series of logistic regression analyses in which these variables alternate as the outcome variables, and the five-point ordinal indicator of reliance on ideological appeals from RAID is the main explanatory variable. I do this to examine the association between reliance on ideological appeals and a variety of dimensions of rebel group capacity (online appendix, Tables A1–A3).¹

Across a series of models, I find a negative and statistically significant association between the extent to which groups rely on ideological appeals and both their ability to procure arms and their overall fighting capacity, relative to the government. Said differently, I find evidence that groups that rely more on ideological appeals, relative to material incentives, have a lower overall fighting capacity and a more difficult time acquiring arms. In contrast, I find a positive and statistically significant association between reliance on ideological appeals and the mobilization capacity of rebel groups, relative to governments. Marginal effects for these analyses are displayed in Figure 1a–c.

These results provide some evidence for the argument that different recruitment strategies affect different dimensions of rebel capacity in different ways. This is complemented by previous subnational, quantitative analysis, which shows that ideologically motivated recruits are less likely to desert and defect (Oppenheim et al. 2015). Thus, both ideological- and material-based recruitment strategies could be associated with difficulties in interpreting the strength of rebel organizations. Such ambiguities can decrease the probability of negotiated settlements occurring (Walter 1999). Additionally, if the extent to which groups rely on ideological or material appeals for recruitment does not have a clear effect on their capacity, then reliance on ideological appeals might not be a strong predictor of the probability of achieving outright victory over government forces. These findings support my earlier argument that there are both advantages and disadvantages of ideological-based recruitment strategies, as they improve rebel groups’ capacity on some dimensions, but undermine their capacity in other ways.

Do Mixed Recruitment Strategies Work Better?

A potential, alternative theoretical story is that groups that employ a combination of ideological and material appeals, rather than relying almost exclusively on just one type, are more likely to be successful. It is possible, for instance, that such groups have the material resources to perform well on the battlefield and recruits who are ideologically motivated enough to not desert or defect in large numbers. Said differently, rebel groups might be the most effective when they mobilize around both ideological and material resources, and do not come to overly rely on one and can garner advantages associated with both.

However, there are theoretical reasons to expect that such a recruitment strategy also does not provide clear advantages, relative to other types of mobilization tactics. First, there might still be significant uncertainty surrounding the capabilities of these groups. Indeed, these groups are unlikely to have the same level of material resources as groups that can primarily mobilize around economic incentives and thus might not have the same capacity to perform on the battlefield. Relatedly, while there might be some ideologically committed recruits in these organizations, they may not have a high percentage of recruits who are as deeply committed as groups that mobilize primarily around ideological appeals do. Said differently, groups that combine ideological and material recruitment appeals might have middling levels of both material and ideological resources and thus might not be substantially stronger than other types of rebel organizations.

Second, such groups might also struggle to credibly commit to peace agreements. Just as groups with ideologically committed recruits struggle to credibly commit that their members will be satisfied with compromising on ideological issues and not take up arms again, groups that mobilize around both ideological and material appeals might have trouble credibly committing that most members will be satisfied with the terms of peace agreements. Specifically, if groups have some members who are more materially motivated, and others who are more ideologically committed, then it might be difficult for rebels to achieve peace agreements that satisfy both types of recruits. Consequently, it might be difficult for rebel leaders of groups that employ a relatively even combination of ideological and material appeals to credibly commit that all factions of their movement will be satisfied by, and obey, a peace agreement.

However, to account for this possibility, I conduct analyses that attempt to capture differences between recruitment strategies that rely (1) exclusively on material incentives, (2) only on ideological appeals, or (3) some combination of the two. Specifically, I created two binary indicators, one measuring groups that rely exclusively on material incentives (i.e., those coded as 0 on the original ordinal scale) and the other capturing groups that only employ ideological appeals (i.e., those coded as 4 on the ordinal scale). I reconduct the main analyses, using these binary indicators as the main explanatory variables instead of the original measure of reliance on ideological appeals. The baseline excluded category is thus groups that employ some combination of both ideological and material recruitment appeals (online appendix, Table A7).

Neither full reliance on material incentives nor exclusive use of ideological appeals has a statistically significant association with ending through negotiated settlements or total rebel victory. These variables do not have a consistently statistically significant association with rebel groups ending through low levels of activity. However, both explanatory variables have a statistically significant association with rebels ending by total defeat by government forces. Said differently, groups that employ various combinations of ideological and material recruitment appeals are less likely to experience total defeat at the hands of the government than groups that rely exclusively on one type of mobilization tactic. This suggests that combining material and ideological appeals might help rebels stave off the worst possible outcome, but I do not find evidence that such recruitment strategies help increase the probability that militants achieve favorable outcomes.

Robustness Checks

I conduct a variety of robustness checks to assess whether there is any other evidence of an association between ideological-based recruitment strategies and rebel group success. The results are available in the supplemental materials.

Alternative Outcome Variables

First, I consider other ways to measure rebel group success. One potential issue is that the conflict termination data used in the main analysis does not capture rebels’ efforts to enter negotiations nor the breadth of concessions that they receive (Thomas 2014). In response to this, I replicate two studies that measure rebel negotiations and concessions, adding in the measure of relative reliance on ideological recruitment appeals.

I begin by replicating a study by Thomas (2014), who examines how terrorism affects both the probability that rebel groups enter negotiations with governments and the number of concessions rebels receive from governments. Her sample covers 106 African rebel organizations between 1989 and 2010. If material-based recruitment strategies are systematically more detrimental, we should see evidence that rebel groups are more likely to enter negotiations, and earn concessions, when they mobilize predominately around ideological appeals.

Using replication data and code from Thomas (2014), I rerun the main analysis, adding in the five-point ordinal measure of reliance on ideological appeals. I do not find evidence of a statistically significant relationship between reliance on ideological appeals and the probability rebels enter negotiations (online appendix, Table A8), nor do I find a statistically significant association between this explanatory variable and any of the counts of concessions built by Thomas (online appendix, Table A9). Thus, even using more fine-grained data, I do not find support for the idea that material-based recruitment strategies uniquely “curse” rebel organizations.

There is a possibility that I find only null results because the sample discussed above is restricted to only African rebel groups. In response, I replicate research by Cunningham and Sawyer (2019), which uses a global sample of rebel groups to examine how the means by which rebel leaders rise to power within their organizations influence the probability that the groups enter negotiations. I also use replication data and code from this study to rerun analyses that include the ordinal measure of reliance on ideological appeals (online appendix, Table A10). I do not find evidence that reliance on ideological appeals affects the probability that rebels enter negotiations, which is often an important step toward garnering concessions (Thomas 2014).

Alternative Explanatory Variables

Another issue that could be driving the results is how the central explanatory variable is measured. Indeed, there was a subjective element of determining what position on the ordinal scale that any one rebel group fell (Soules 2023). I take a couple of different measures in response. First, I employ a simplified three-point version of the original five-point ordinal measure of reliance on ideological appeals. Specifically, this variable collapses the original measure into three categories: exclusive reliance on material incentives (0), any combination of material and ideological appeals (1), and total reliance on ideological appeals (2). I rerun the main analysis using this measure and still do not find a statistically significant association between reliance on ideological appeals and the probability of experiencing any of the outcomes (online appendix, Table A11).

Next, RAID contains a measure of the degree of confidence associated with coding the five-point ordinal indicator of reliance on ideological appeals for each group. This variable captures whether there was low (1), moderate (2), or high (3) confidence in the coding. I reconduct the main analysis, excluding all observations that received the lowest level of certainty for the coding of the main independent variable. I once again do not find evidence of a statistically significant association between reliance on ideological recruitment appeals and any of the outcomes of interest (online appendix, Table A12).

Accounting for Variation in Ideology

Yet another potential issue is that, among groups that recruit heavily with ideological appeals, there is significant variation in the types of ideologies that they organize around. Before empirically delving into this issue further, it is important to reemphasize why I do not examine variation in types of ideology in the main analysis. While this variation in ideology helps explain the outcomes rebels experience (e.g., Keels and Wiegand 2020), the underexplored implication, based on existing theories, that ideological-based recruitment strategies produce more favorable outcomes than material-based mobilization tactics has largely been untested. This school of thought focuses on variation in reliance on material and ideological appeals, not variation within ideological appeals. However, variation within ideological appeals also helps challenge the notion that ideological-based recruitment strategies are generally preferable, as certain ideologies lead rebels to wage more intractable and less successful fights (e.g., Svensson 2007; Keels and Wiegand 2020; Nilsson and Svensson 2020, 2021; Basedau, Deitch, and Zellman 2022).

It is still important, however, to account for the possibility that some dimension of ideological variation is driving the core findings. To start, in the main models, I control for whether rebel groups have a radical Islamist or left-wing ideology. However, these variables might be highly correlated with the main explanatory variable, introducing issues related to multicollinearity. To account for this, I rerun the main analysis, removing these two control variables (online appendix, Table A13). Reliance on ideological appeals does not have a statistically significant association with any of the types of outcomes even when the controls for ideology are excluded. Next, I reconduct the main analysis twice, once excluding all groups with a radical Islamist ideology (online appendix, Table A14) and once excluding all groups with a left-wing ideology (online appendix, Table A15). I do this because such ideologies can make conflicts more intractable, even relative to some other types of ideology (Basedau, Deitch, and Zellman 2022). When I do this, I still do not find evidence of a statistically significant association between reliance on ideological recruitment appeals and any of the outcomes in question.

Alternative Control Variables

In the main analysis, I controlled for the V-Dem dataset’s measure of electoral democracy (Coppedge et al. 2024). However, in the model in which the outcome variable captures whether groups ended through negotiated settlement (Table 1, Model 2), the coefficient and standard error for this variable are very high. To ensure that any potential issues with this measurement are not driving the results, I reconduct the main analysis, substituting the Polity V project’s 21-point ordinal indicator of how democratic a country is (Marshall and Gurr 2020). I do not use this measure in the main analysis because it has more missing observations than the V-Dem measure of electoral democracy. However, even when I control for this alternative measure of regime type, I do not find any evidence of a statistically significant association between reliance on ideological recruitment appeals and any of the outcomes under study (online appendix, Table A16).

Additionally, in the main analysis, I limited the number of control variables I included because of the relatively small sample size. However, I conduct additional analyses that include more control variables. Specifically, using data from the NSA dataset (Cunningham et al. 2013), I include a binary indicator of whether a rebel group controls territory. Using data from Fariss et al. (2022), I also control the logged population of the country. Both factors are expected to be associated with the longevity and success of rebel organizations (Cunningham, Gleditsch, and Salehyan 2009). I reconduct the main analyses, including these variables, but still do not find a statistically significant association between reliance on ideological appeals and any of the outcomes in question (online appendix, Table A17).

Alternative Estimation Technique

Another potential issue is that many of the covariates used in the competing-risk model are time invariant, despite these models being used, in large part, to account for temporal dynamics. In response, I reconduct the main analysis, employing multinomial logistic regression analysis instead of competing risk models. The unit of analysis for these tests is still the dyad episode, and the outcomes being examined are still rebels (1) earning a peace agreement, (2) achieving total victory over government forces, (3) being totally defeated by government forces, or (4) ending through low levels of activity. The baseline excluded category is groups that are still active at the end of the dyad episode. Even using these alternative tests, I still do not find evidence that reliance on ideological recruitment appeals increases or decreases the probability that rebels experience favorable or unfavorable outcomes (online appendix, Table A18).

Overall, even when I subject the first hypothesis to a battery of alternative tests, I do not find consistent evidence supporting it. Thus, I do not find substantial evidence in support of the notion that rebel organizations are more likely to be successful if they rely more heavily on ideological appeals than material incentives for recruitment.

Conclusion

An important implication derived from the literature on the “rebel resource curse” is that rebel organizations that mobilize recruits with material incentives are cursed, as they tend to attract uncommitted, opportunistic individuals who undermine the long-term goals of their groups (Weinstein 2005, 2007). However, there is a dearth of quantitative literature investigating how recruitment practices affect the success of rebel movements.

In this paper, I posit that there are several significant, but often overlooked, drawbacks of ideological-based recruitment strategies. These issues challenge the idea that ideological appeals are the superior recruitment tool. Specifically, ideological appeals and recruits can introduce a variety of problems into the bargaining process with governments, all of which undermine rebels’ chances of earning negotiated settlements. Furthermore, ideological-based recruitment strategies do not provide a clear advantage, relative to material incentives, in building powerful fighting forces, and thus, ideologically driven groups will not be more likely to achieve total victory over government forces.

Using novel data on rebel recruitment practices, I examine how armed groups’ reliance on ideological recruitment appeals, relative to material incentives, affects their fate. Contrary to the implications of prior theories on rebel recruitment, I do not find evidence that ideologically based recruitment strategies increase or decrease the probability of success for militant movements. Again, I do not expect that rebels will be more successful if they employ material incentives to mobilize recruits, nor do the results show this. Instead, there is no clear evidence of material incentives or ideological appeals having greater long-term benefits for rebels’ efforts to achieve their goals.

Of course, null results mean that I did not find evidence for the argument that ideological-based recruitment strategies help make rebel groups more successful, not that I find evidence against it. However, the ideas underpinning the “rebel resource curse” have largely been unchallenged both theoretically and empirically. Thus, it is worth evaluating whether the extent to which groups rely on ideological recruitment appeals, relative to material incentives, affects the success of rebel organizations. Future work should continue to interrogate the ways in which rebel groups’ recruitment strategies affect the outcomes of the experience. Indeed, while material wealth might sometimes curse rebel groups, ideology can too.

Footnotes

References