How Rebel Governance Affects Civilian Behavior: Evidence from Field Experiments

• In Event: Political Order in Rebel Governance

Abstract

Rebel actors often aim to capture and secure territory from the state in order to undermine the state’s legitimacy and extract resources to continue the war effort. Stronger, ideologically oriented groups tend to utilize territorial control as an opportunity to signal their capability to govern as well. Yet the extent to which rebel groups control and govern within territory varies substantially. Territory is, at times, fully controlled by rebel actors, while on other occasions it is either contested between rebel forces and the government or is firmly controlled by the government. While studies that employ observational data have analyzed when, how, and why rebel governance occurs, we know very little about the local political and psychosocial consequences of experiencing the contestation of governing territory at the individual level. Recent research has emphasized how local social networks and collective action capabilities can shape rebel governance choices and responsiveness, while other work has focused on how experiencing violence affects long-term behavioral traits related to interpersonal trust. Very few studies have examined how rebel rule or contested governance shapes the collective behavior of civilians, inter-personal trust and attitudes toward authority.

How does the contestation of territory and exposure to non-state governance by rebel actors affect collective action by citizens, trust between them, and attitudes towards de jure governing entities? Answering this question is important because the psychosocial effects of conflict processes affect rebel governance and are at the root of many local-level peacebuilding challenges during and after civil wars. Despite the importance of this topic, we still do not understand whether and what dimensions of a local social fabric serves as a resistor to certain types of governance or how governance (by state or non-state actors) affects local social fabrics.

We argue that there is a U-shaped relationship between the level of rebel control and the degree to which civilians engage in collective action, trust each other, and have trust in authority. Greater political uncertainty in contested areas induces free-riding by civilians, eroded trust between them and reduces trust in authority. By contrast, higher certainty and longer time-horizons associated with collective interaction in rebel-controlled areas and those under state control curbs free-riding, promotes inter-personal trust, and increases trust in authority.

To test these causal claims, we conducted a series of lab-in-the-field experiments across villages, towns and cities in three neighboring districts in Eastern India that have experienced varying levels of control by Maoist rebels (i.e., Naxalites): Vaishali (Bihar), Samastipur (Bihar), and Dhanbad (Jharkhand). We used detailed data from the census and other sources to identify these districts as having similar demographic, economic, and political features. However, Vaishali represents a district with a strong and sustained state presence, Samastipur represents a territory contested between the Indian state and Maoist insurgents, and Dhanbad currently exhibits state control but was previously controlled by rebels. We fielded our experiments across three clusters in each district, resulting in 1,100 local participants.

The lab-in-the-field experiment includes two games: a repeated Public Goods Game with Voluntary Contributions Mechanism and the Simple Trust Game in which the randomly selected sender acts as an authority that provides an initial endowment to receivers. The first is used to measure inclination towards contributing resources and effort for generating Common Pool Resources (CPRs). The second game is used to measure trust, trustworthiness, and dispositions regarding reciprocity.

Our findings suggest that independent of individual experiences of violence by state or rebel forces, living under contested governance reduces voluntary contributions and thus CPRs, decreases trust in authority, and trust in one another. Contrary to our theoretical expectations, civilians in rebel-controlled areas engage in free-riding and have low levels of inter-personal trust, but exhibit moderate levels of trust toward authority. In state-controlled areas, we find that voluntary contributions are significantly high, while the degree of trust in one another and in authority is also high. Our lab-in-the-field experiments contribute to a deeper understanding of the micro-foundations of conflict processes like civilian agency while also speaking to larger questions about the challenges associated with reintroducing the state into a contested territory during or after a protracted conflict; the long-term consequences of local state absence; and the strategic gains and pitfalls for rebel groups that use service provision, violence, or both to control or garner support among civilians.

Authors

• Brandon Lee Bolte, University of Illinois at Springfield

• Vineeta Yadav, Pennsylvania State University

• Bumba Mukherjee, Pennsylvania State University