The implementation of democracy in Iraq in 2004 shifted the balance of power between Sunnis and Shias, creating a security dilemma in which the Shia-dominated state struggled to credibly commit to positive future treatment of the newly weakened Shias. I propose that police integration addresses this commitment problem because it gives included groups reason to expect better treatment, and because it is costly for the state to renege on police integration once implemented. I first show that Iraqi Sunnis, who believe the police are integrated interpret service provision as fairer, perceive that their group has better access to government jobs and expect less future repression by the police or the government. These outcomes map onto commonly identified motives for conflict, including horizontal grievances, unemployment, and the commitment problem. I then test the link between integration and support for violence. Using an experiment embedded within the survey, I find that Sunnis exposed to information that the police are integrated are significantly less likely to support anti-state violence.

This chapter provides an overview of policing in modern Iraq, with a special focus on the role of sectarian identity in citizen–police relations. Following the 2003 US invasion and the overthrow of Saddam Hussein, the transitional government initially purged Sunnis from the state security forces. However, the government soon faced domestic and international pressures to reintegrate Sunnis into the police as a counterinsurgency measure. In some areas like Anbar and parts of Baghdad, these pressures led to substantial inclusion of Sunnis in the police. In other areas, local politicians resisted Sunni integration so they could distribute police jobs as political patronage. Drawing on interviews with Iraqi citizens, I show just how salient the police’s makeup is to ordinary citizens’ relationships with the state. Citizens see access to positions in the police as a path toward empowerment and, at times, self-preservation. At the same time, they perceive that officers behave differently depending on their sect, with shared identity smoothing interactions between citizens and the police.

In Israel, police integration is limited mainly by the supply of Arab citizens who wish to join the police. Many Arab Israelis express skepticism that participation in the criminal justice system will improve services, worrying that Arab officers will instead be used as collaborators in anti-Arab repression. Consequently, interviews with police officers paint a picture of an institution that wishes to provide better services to minorities but struggles to do so due to a lack of citizen buy-in. I explore detailed station-level data covering a six-year period on police officers’ religious identities, and I uncover processes through which the police assign officers from different groups to different locations through interviews with high-ranking police officers.

This chapter highlights key findings and proposes a path forward for the implementation of police integration in power sharing and post-conflict agreements. I identify several key lessons of police integration. In particular, it has the potential to provide societywide net benefits, making it a promising first step toward reconciliation where intergroup trust is especially low. When it comes to implementation, police integration is applicable to an unusually wide range of settings because unlike legislatures and cabinets, the police are no less effective in autocracies compared to democracies. I close by proposing a potential extension of my argument: Integration of other service-providing bureaucracies such as public education or healthcare may have similarly positive impacts on citizen–state relations by helping the government credibly signal its intentions to historically marginalized citizens.

Given the advantages of police integration, why do leaders not integrate the police in every divided society? I first explore the possibility that dominant-group civilians oppose integration, for example out of concern for a loss of jobs or fear that newly-integrated minorities will renew hostilities once integrated. Instead, I find that an overwhelming majority of Iraqi Shias and Israeli Jews support police integration. On the other hand, politicians and community leaders often oppose police integration for personal gain. Iraq, local politicians resist sectarian integration for fear that it will limit their opportunities to distribute jobs in the police as patronage in return for political support. In Israel, Arab community leaders who seek to create an independent Palestinian state view participation in the police as a dangerous normalization of the status quo. To discourage Arabs from joining the police, they portray participation in the police as collaboration with the enemy. Thus, implementation of police integration tends to be limited not by mass-opposition but by instrumental political concerns of political elites.

In Israel, the relationship between Arab citizens and the police centers on a desire for better service provision. This chapter asks how police integration affects the quality and distribution of crime prevention. First, I use original survey data to show that Arab citizens who perceive the police as more integrated are more willing to report crimes to the police. Next, I argue that the police use this information to allocate crime-fighting resources more efficiently. Using panel data on officer demographics and a nationwide crime victimization survey, I show that within a given location, increases in the religious diversity of officers are associated with decreases in crime. This effect occurs only in localities with majority non-Jewish populations. However, within those localities, the decrease in crime holds for both Jewish and non-Jewish citizens. These results point to the potential for police integration to benefit all members of society, making it a powerful first-step toward reconciliation in communities with deep-seated identity divisions.

This chapter begins with an overview of police integration worldwide. Nearly 40 percent of modern peace agreements include police reforms, and the makeup of the police rank and file is often central to these reforms. Yet, despite the police’s central role in peace-building, existing literature on power sharing pays little attention to this critical institution. I argue that in divided societies, the police rank-and-file’s unique combination of three characteristics – visibility, discretion, and capacity for violence – makes it especially influential in shaping citizen–state relations. The second half of the chapter details causal mechanisms linking police integration with citizens’ attitudes and expectations. For example, language differences between officers and citizens may make citizens feel less comfortable communicating with officers, or citizens may fear that cultural differences will prevent officers from understanding their situation. When the police are integrated at the patrol-level, officers can monitor their colleagues’ treatment of coethnic citizens and deter bad behavior.

The police are a critical but understudied institution in divided societies. I argue that integrating marginalized groups into the police improves citizens’ attitudes toward the state, not just in terms of their experiences today but also their expectations of the future. Integrating the police signals the government’s commitment to included groups’ security. The signal is critical because police departments are difficult to purge. Once integration is implemented, it is difficult for the state to undo. I propose that the difficulty of purging the police, and consequently the credibility of the government’s signal to an included group, depends not just on how many officers come from each group but also how those officers are distributed. This chapter also introduces the two main cases for analysis, Iraq and Israel, and summarizes the main findings.

This chapter tests two intermediate links in the chain connecting police integration and citizens’ behavior. Interviews and focus groups reveal that citizens use cues like officers’ accents, names, and facial hair styles to determine their identities. Comparing survey responses with data on officer demographics suggests that citizens’ perceptions correlate closely with actual local integration. On the other hand, marginalized-group citizens appear not to translate their local-level observations of integration into perceptions that the institution overall is integrated. I demonstrate that citizens process observations of police demographics through their accumulated experiences with the police and the state. In the shadow of historical exclusion, citizens update perceptions of the institution incrementally.