The Non-Economic Logic of Rebel Taxation: Evidence from an Islamic State-Controlled District

Mara Revkin, Yale University

Previous studies of rebel governance predict that armed groups with access to exploitable resources will tend to engage in short-term, opportunistic looting rather than invest in building the complex bureaucracies that are necessary for taxation and long-term governance (Weinstein 2006; Collier and Hoeffler 2004; Olson 1993). A related argument from the “resource curse” literature is that complex tax systems are most likely to be found in resource-poor states where governments have no alternative but to extract revenue from their populations (Ross 2004; Fearon and Laitin 2003). Such claims rely on an untested assumption that the sole purpose of taxation is to collect revenue. I challenge this assumption with new evidence from the case of the Islamic State (IS), an insurgent group that taxes civilians in oil-rich areas of Syria, Iraq, and Libya at levels that do not appear to be economically rational. For example, a Deir Ezzor newspaper reported that tax rates are higher under IS rule than they were previously under the Syrian regime, even though the local unemployment rate is at least 75%, according to interviewees, and residents cannot afford to buy food, much less pay taxes. One interviewee from Deir Ezzor was puzzled by IS’s decision to tax so heavily in an area where “they are already rich from oil. It makes no sense.”

IS’s behavior in Deir Ezzor is surprising, because the imposition of taxes is associated with significant economic and political costs. First, effective taxation requires an extensive bureaucratic apparatus that is capable of calculating and collecting taxable assets (Tilly 1975; North 1981; Scott 1998). Second, taxes are universally disliked and are therefore a source of friction between state and society (Levi 1989). Given these costs, why would a rebel group go through the trouble and expense of collecting taxes in areas where abundant resources should greatly reduce or eliminate the need for tax revenue? I argue that the puzzle of taxation in resource-rich areas can be explained by taking into account the non-economic functions of taxation in governance and state-building. Scholars of taxation have long recognized that states derive non-economic benefits from taxation (Heer 1937) and I apply this insight to rebel groups with state-like ambitions for territorial control and sovereignty. I theorize that taxation, in addition to raising revenue, may facilitate rebel governance and state-building through three additional mechanisms: social control, collective identity formation, and demographic engineering. This theory of the non-economic functions of rebel taxation resonates with other research in this POMEPS series on the ways in which Islamist-governed municipalities promote particular identities and values (Marschall, Shalaby, and Unal 2017).

If I am correct that taxes are imposed for non-economic reasons in addition to their obvious role in the collection of revenue, I should find that IS is statistically just as likely to impose taxes in resource-rich districts as in resource-poor districts because, even though taxation is not economically necessary in the latter areas, it still provides important non-economic benefits for state-building. Second, I should find qualitative evidence of the theorized non-economic mechanisms. In a new working paper, I evaluate this theory quantitatively with an original dataset on seven different types of tax policies from the 18 Syrian districts (rural and urban) that have been governed by IS since 2013. Given space constraints, this memo focuses on a case study of the district of al-Mayadin, where I find support for my theory in qualitative evidence. The data on al-Mayadin is drawn from a 45-month dataset that differentiates between seven types of tax policies: income taxes, border taxes, excise taxes, fines, licensing fees, property taxes, and service fees. For each district, I determined whether or not a given tax policy was being implemented (as a binary variable) during a particular month by triangulating among social media data, local newspapers, archival IS documents, and interviews with 112 key informants conducted over the course of five months of fieldwork in southern Turkey. Each observation is supported by at least one document (and as many as seven) and corroborated by an interviewee from the relevant district. A selection of documents from the larger dataset that are cited in this memo are available online at the following link:

Case study of al-Mayadin

The Syrian district of al-Mayadin is home to the country’s largest and most profitable oil field (al-Omar) and is therefore a “hard test” of my theory. If rival theories are correct, then we should find low levels of taxation in al-Mayadin because IS derives ample revenue from oil. Another reason to expect low levels of taxation is that IS has faced significant opposition in the form of nonviolent protests as well as armed resistance in the district. Since IS captured the district in July 2014, there have been at least two peaceful anti-IS demonstrations and eight armed attacks against IS personnel or buildings including once incident in which assailants set fire to IS police cars near a sheep market (Dataset ID #1166). During the same period, there were at least two peaceful demonstrations: one in which a group of women gathered outside of the IS court in al-Mayadin to demand the release of their imprisoned sons and another in which protesters raised the flag of the Free Syrian Army, with which IS is at war (Dataset ID #496, 539). Since taxation tends to exacerbate unrest, a rational sovereign would be unlikely to risk further provoking a population that is already on the verge of rebellion.

However, this case study demonstrates just the opposite: Not only did IS collect taxes during 29 of the 35 months that it has controlled the district, but it has also imposed seven different types of tax policies. If IS were taxing the oil industry itself, then this data might still be consistent with rival theories. But of the 45 instances of taxation in the dataset, only one was related to oil: a 2.5 percent zakāt tax imposed on the owners of independent refineries (Dataset ID #791). Regardless, rival theories cannot explain why IS imposes taxes on so many other types of activities and commodities that are unrelated to oil including: service fees for internet, phones, electricity, sanitation, and roadwork (Dataset ID #531, 535, 868); licensing fees for motorcycles and vegetable market stalls (Dataset ID #890, 528); and fines for smoking cigarettes, shaving beards, and littering (Dataset ID #859, 861, 1142). I use this case study to illustrate the non-economic dimensions of IS’s tax policies. 

Collective identity formation

The first tax introduced in al-Mayadin was a border tax of 5,000 Syrian pounds (then about $23.33) per month on buses transporting passengers from other areas of Syria in August 2014 (Dataset ID #945). Border taxes play an important role in the symbolic construction of sovereignty by demarcating the spatial boundaries of rebel-controlled territory and demonstrating the group’s capacity to regulate the movement of people and commodities across those boundaries. The next tax policies to appear were fees on drinking water in September 2014 and household gas in October 2014 as IS began to provide these basic services (Dataset ID #947). The provision of services in exchange for tax payments is an observable indicator of the presence of a social contract (Arjona 2016). In al-Mayadin, the introduction of these fees was symbolically significant because it indicated that IS had attained the status of a de facto state with the ability to provide services and the coercive capacity necessary to collect payments from the beneficiaries of those services.

In January 2015, IS opened an income tax collection office known as a diwan az-zakāt. During the Friday sermon, an imam announced that civilians would be required to pay 2.5 percent of their annual earnings and warned that these taxes would be collected “by force” if not given voluntarily. Shortly thereafter, tax collectors began visiting the shops of goldsmiths and other merchants to calculate their taxable assets (Dataset ID# 1157). According to official IS doctrine, the payment of zakāt is a religious obligation, and therefore tax evasion is considered an act of apostasy (Islamic State 2015). The fact that these taxes are imposed universally on all but the poorest Muslim civilians contributes to a sense of membership in a common political and religious community. IS has erected billboards publicizing its tax policies and warning of the punishment for tax evasion. At least some residents viewed these taxes as fair and legitimate based on the services and charity being provided by IS. According to one, “At first, we were annoyed by the collection of zakāt, but when we saw [IS] registering the names of the poor and giving them money and food, our admiration for them increased … The money goes back to the people through the services that [IS] provides.”

By March 2015, IS had introduced a property tax of 200 Syrian pounds (then about $0.93) per dunam (0.25 acres) of land (Dataset ID #1098). In al-Mayadin, the introduction of a universal income tax and property tax was not only a means of revenue collection but also a symbolic assertion of sovereignty over people and territory. It is noteworthy that IS did not impose these taxes immediately upon capturing al-Mayadin in July of 2014 but instead waited many months until it had built up sufficient coercive capacity to be able to enforce the policies. During these six months, IS was also creating a judicial system and providing services in an effort to legitimize its rule. If the goal of taxation was simply to generate as much revenue as possible, why would IS limit the rate of income taxation to only 2.5 percent? And why not impose income and property taxes immediately, rather than waiting six months? The answer, I argue, is that insurgent taxation is more than a means of extracting resources from civilians. Taxation turns civilians into citizens of a new political community. Multiple interviewees referred independently to the “symbolic” significance of taxes. According to one man who had paid taxes to IS in al-Mayadin, “They tax because that is what states do, and they want to be a state.”

Social control

 Interviewees from al-Mayadin described the IS tax system as an instrument of social control. One former resident said of his encounters with tax collectors, “It wasn’t about the money… They tax us to show that they are in control.” Several interviewee reported being fined for violations including wearing colorful shoes, listening to the radio, and failing to close shops during prayer times in order “to teach us a lesson,” in the words of one man. In some cases, fines were accompanied by publicly administered corporal punishments including whipping with leather straps and a form of public shaming in which the alleged wrongdoer is forced to wear a sign stating the violation and corresponding punishment. In one example from al-Mayadin, a man received 200 lashes and a fine for possessing a satellite dish (Dataset ID #1155). These cases suggest that the revenue-generating function of fines is secondary to their function as a tool for social control and discipline. It is possible that IS collects such fines for the sole purpose of raising revenue, but I believe it is more likely that these policies are implemented with the intent of conditioning the behavior of civilians in ways that are consistent with IS’s strategic or ideological objectives. For example, imposing fines to deter civilians from using satellite dishes is a means of regulating their access to information and communications with the outside world.

Demographic engineering

The case of al-Mayadin also supports my claim that rebel groups use tax policies as a tool for demographic engineering. In October 2014, IS announced that it would impose a jizya tax on Christians in exchange for guaranteeing their protection as non-Muslim subjects of the caliphate. At the time, there was only one family of Christians living in the district and IS presented them with three options: convert to Islam, pay the jizya tax, or leave IS territory. After the Christians chose to leave, IS confiscated the property they left behind (Dataset ID #1145). The jizya tax, by imposing costs on non-Muslims, has the effect of discouraging and stigmatizing the practice of minority religions. In the case of al-Mayadin, the tax led to the departure of the only remaining Christians.

IS also appears to be using border taxes to manage the demography of al-Mayadin. As it becomes more difficult for IS to attract foreign recruits, the group has become more focused on preventing human capital flight out of the areas that it controls (Revkin 2016). In an apparent attempt to discourage out-migration, IS imposed a tax of approximately $100 on every resident traveling out of Menbij, a district adjacent to al-Mayadin, in January 2016 (Dataset ID #363). A resident of al-Mayadin reported that similar “exit taxes” were levied there.

Finally, women expressed the view that IS police were deliberately harassing them with gender-based fines to discourage them from going outside. According to one woman, “The threat of these fines and other punishments made us prisoners in our own homes,” where their only role is to serve their husbands and raise children. In this rural district with many family-owned farms, women have traditionally worked alongside men in fields and pastures. When IS took control, however, the religious police began issuing fines to female farmers who did not comply with the new dress code (Dataset ID #925). One female farmer was arrested for wearing “shoes that drew attention” and brought before the local IS court, where the judge ordered her to pay a fine and gave her a new pair of dark shoes to wear. As a result of these fines, “most women prefer to stay home and not go outside,” she said (Dataset ID #1087). In some cases, IS punished husbands for violations committed by their wives. For example, one man was publicly whipped and fined because his wife was not wearing the mandatory face veil (Dataset ID #1138). Such punishments create incentives for husbands to monitor and control their wives. IS envisions a society in which women exist only to raise the caliphate’s next generation, and the group appears to be using certain tax policies to enforce particular gender roles.


Much of the literature on rebel governance has been shaped by an untested assumption that the sole purpose of taxation is to collect revenue. This memo has challenged that assumption by theorizing that tax policies may promote state-building through three non-economic mechanisms: social control, collective identity formation, and demographic engineering. The memo suggests several directions for future research. First, what is the relationship between rebel taxation and legitimacy? Does taxation, through the mechanism of collective identity formation, make civilians more inclined to support the insurgency? Or does taxation lead to increased resistance by generating demands for representation and participation? As IS’s footprint in Syria and Iraq continues to shrink, it will become possible to collect new and higher quality data to address these questions. 


Arjona, Ana. 2016. Rebelocracy: Social Order in the Colombian Civil War. New York, NY: Cambridge University Press.

Collier, Paul, and Anke Hoeffler. 2004. “Greed and Grievance in Civil War.” Oxford Economic Papers 56 (4): 563-595.

Fearon, James and David Laitin. 2003. “Ethnicity, insurgency, and civil war.” American Political Science Review 97 (1): 75-90.

Heer, Clarence. 1937. “Taxation as an Instrument of Social Control.” American Journal of Sociology 42 (4): 484-492.

Islamic State. 2015. “And they Gave Zakāt.” video-message-from-the-islamic-state-and-they-gave-zakah/.

Levi, Margaret. 1988. Of Rule and Revenue. Berkeley, CA: University of California Press.

Marschall, Melissa, Marwa Shalaby, and Saadet Konak Unal. 2017. “Municipal Service Delivery, Identity Politics and Islamist Parties in Turkey.” Presented at POMEPS Workshop on Islamists and Local Politics, Brastad Sweden, June 13-15.

North, Douglass C. 1981. Structure and Change in Economic History. New York, NY: W.W. Norton.

Olson, Mancur. 1993. “Dictatorship, Democracy, and Development.” American Political Science Review 87 (3): 567-576.

Revkin, Mara. 2016. “Does the Islamic State Need Territory to Survive?” New York Times, October 21.

Ross, Michael L. 2004. “What Do We Know About Natural Resources and Civil War?” Journal of Peace Research 41 (3): 337-356.

Tilly, Charles. 1975. “Western State-making and Theories of Political Transformation.” In Charles Tilly, ed., The Formation of National States in Western Europe. Princeton, NJ: Princeton University Press, 601-686.

Weinstein, Jeremy. 2006. Inside Rebellion. New York, NY: Cambridge University Press. 


Table I is a simplified rendering of the shape of the dataset with letters—I, B, E, F, L, P, and S—representing the seven types of tax policies, income taxes, border taxes, excise taxes, fines, licensing fees, property taxes, and service fees. The unit of analysis is the district-month, with columns for each district and rows for each month. Cells are grayed in for months in which IS did not control a particular district, either because the group had not yet captured the area or because it had lost control.

A selection of documents from the larger dataset that are cited in this memo are available in PDF form at the following link:

Each file name follows the format: “Dataset ID # – Province Name – District Name – Type of Tax Policy (any notes about the policy) – Month Year.pdf.”

Table I. Visualization of the Dataset