An open access publication of the American Academy of Arts & Sciences
Fall 2022

Trust & Models of Policing

Tracey L. Meares
View PDF

The notion of trust has become central to the discussion of policing and its transformation over the last decade. Scholars, policy-makers, and the agents who purport to carry out public safety projects on behalf of the public now commonly point to trust as one of the central goals of the relationship between policing agencies and members of the public they serve, in contrast to the more common and familiar notion of crime reduction. This essay highlights three common mechanisms agencies and the individuals they comprise use to attempt to improve the public’s trust in police: changing policy, training of police, and citizen oversight boards. Focusing on the conceptual framework that the social psychology of procedural justice offers, the essay then turns to a less common target for change: the very laws police enforce. Changing the police will require not only transforming how its members carry out the job but also the laws they are sworn to uphold.

Tracey L. Meares, a Fellow of the American Academy since 2019, is the Walton Hale Hamilton Professor and a Founding Director of the Justice Collaboratory at Yale Law School. She is the editor of the Annual Review of Criminology (with Robert J. Sampson) and has published in such journals as The Annals of Political and Social ­Research, Journal of Social Issues, and Journal of Experimental Criminology.

Over the last decade and certainly since 2015, it has become common for scholars, policy-makers, and those in public media outlets to highlight the importance of the public’s perception of police as a legitimate authority. The Final Report of the President’s Task Force on 21st Century Policing contains, perhaps, one of the most prominent statements of this idea.1 President Barack Obama convened the Task Force in 2014 after Michael Brown was killed by police officers in Ferguson, Missouri, and Eric Garner was killed by police officers in New York City. The foundation of the Task Force’s fifty-nine recommendations for research, policy, and action is the report’s first pillar, “Building Trust and Legitimacy.” This initial step builds on extensive research of the concept of empirically assessed legitimacy and its close connection to the social psychology of trust and procedural justice. Beginning with psychologist Tom Tyler’s seminal work on procedural justice, there is now an extensive literature demonstrating that when police focus on process, or how they treat members of the public, as opposed to outcomes, such as lowering crime rates no matter the approach, people are more likely to perceive them as legitimate and trustworthy.2 But legitimacy and trust as measured in these studies are not one and the same; rather, perceptions of trustworthiness are important precursors to the public’s conclusions regarding legitimacy. A growing literature also demonstrates that procedural justice, or process-based fairness, is associated with greater trust in police, as well as increased perceptions of legitimacy.3 The recent turn to improving trust relationships has not meant that former goals such as crime reduction are no longer important. Instead, the research demonstrates that agencies can continue to pursue such goals while also treating members of the communities they serve with dignity and respect. With this in mind, we explore recent efforts to address the pervasive lack of public trust across institutions of criminal legal processing and, in particular, distrust of the police.

Although the analysis here does not conceptualize trust as mere “confidence” in the relevant public actor, Gallup polling concerning confidence in police is a useful starting point. Gallup has tracked public confidence in a random sample of adults in a range of institutions, including police, for just over a quarter of a century.4 An examination of these data reveals two striking facts. First, there is a large and distinct gap between levels of confidence in police among white adults as contrasted with Black adults (Figure 1). Among white adults, levels of confidence have hovered right around 60 percent since 1993. With respect to Black adults, confidence rates are approximately half that level, or 30 percent, during the same time period, with the lowest rate of 19 percent posted in 2020. Second, among all adults, these levels have remained largely flat over time (Figure 2).5 Even among Black adults, the confidence rate in police for 2021 was 27 percent.6 The fact that confidence rates have remained largely flat over time, even accounting for the very large gap between white and Black adults, is notable and relevant to the analytical approach of this essay.

In recent decades, police have committed themselves to crime reduction and styled themselves as “warriors” against crime. This orientation fueled a raft of proactive policing strategies, including the notorious “stop, question, and frisk” approach that was constitutionally challenged in 2013 in Floyd v. City of New York.7 During the period that police actively employed these strategies between 1990 and 2016, violent crime declined dramatically, leading scholars to hypothesize about the relative importance of police activity as a contributing factor.8 The group of people most likely to be victims of homicides, young Black men, experienced many fewer homicides during this period of decline. Sociologist Patrick Sharkey expressed this demographic change in the following way, “The impact of the decline in homicide on the life expectancy of Black men [as a group] is roughly equivalent to the impact of eliminating obesity altogether.”9 To the extent that police can claim even some credit for this decline—and some believe that they can claim a great deal of credit—the Gallup polls described above are puzzling and call into question the relationship between public confidence and crime-fighting. If the primary reason for public confidence in police was their effectiveness at crime-fighting, we would expect confidence to rise during that time rather than to remain flat. Moreover, we would expect that the group who received the most benefits of crime-fighting, Black adults, would register increasing ratings of confidence, even accounting for low base rates. But that is not what we see, which suggests that members of the public care about other factors beyond police effectiveness at reducing violence when coming to conclusions about confidence and trust. Tyler’s process-based approach provides a way to account for what we see in the Gallup data.

In their pathbreaking book The Social Psychology of Procedural Justice, psychologists Tom Tyler and Allan Lind develop what they call the “group value theory” of procedural justice, which explains that people understand the ways legal authorities treat them as information about how those authorities view them, as opposed to information about outcome control, which was the prevailing view before Tyler and Lind developed their theory.10 People tend to place much more weight on how authorities exercise power than the ends for which that power is exercised. Across institutional contexts (courts, businesses, schools), researchers have demonstrated consistent findings: public conclusions regarding legitimacy are tied more closely to judgments about the fairness of actions than to evaluations of the fairness of outcomes or the effectiveness of actors in achieving outcomes. In other words, the relationship between the public and how authorities make decisions is inherently relational instead of instrumental.

Procedural justice turns out to be key in members of the public’s determinations of whether they consider legal authorities to have behaved fairly. Their perceptions of procedural justice depend on four factors.11 The first factor centers participation, particularly “voice.” People report higher levels of satisfaction in encounters with authorities when they have an opportunity to explain their situation and perspective. This is true even when people are aware that their participation will not impact the outcome; they nonetheless want to be listened to and taken seriously. Fairness of decision-making by authorities—that is, aspects of fair process—is the second factor. People focus on indications of a decision-maker’s neutrality, objectivity, factuality, consistency, and transparency. In the specific context of policing, it matters whether legal authorities, in their interactions with the public, take the time to explain what they are doing and why. The third factor is related to the first: people care a great deal about how they are treated by legal authorities, such as police officers. Specifically, people desire to be treated with dignity, respect for their rights, and politeness. Being listened to and taken seriously is obviously related to this factor. Fourth and finally, in their interactions with authorities, people want to believe that authorities are acting out of a sense of benevolence toward them. That is, people attempt to discern why authorities are acting the way they are by assessing how they are acting. They want to trust that the motivations of the authorities are sincere and well-intentioned. Basically, members of the public want to believe that the authority they are dealing with believes that they count and cares about them. In relationships with law enforcement, the public makes this assessment by evaluating how police officers treat them.

Research connecting these ideas to policing has demonstrated great benefits for the police agencies that employ them. For example, when people perceive that legal authorities are treating them fairly, they say they are more likely to comply, cooperate, and engage with the law and authorities’ directives.12 Importantly, when policing agencies emphasize process-based approaches, they need not choose between crime reduction and promoting trust.

The public conversation around policing has begun to center trust as opposed to merely police effectiveness at reducing crime. Again, a statement from President Obama’s Task Force is instructive: “Crime reduction is not self-justifying.”13 To that end, in the last several years, police departments have promoted and implemented numerous strategies ranging from changes in policy to active bystander training for officers, to greater civilian involvement in setting policy, goals, and projects for their agencies.

Changing policy is a major starting point for many policing agencies attempting to establish trust through the behavior of officers in their interactions with civilians. For example, recent consent decrees adopted by the United States Department of Justice and the State of Illinois have prioritized requirements that agencies adopt formal policies promoting “Fair and Impartial Policing,” key aspects of which highlight the importance of procedural justice as described above.14 These policies are critical precursors to officer training; instructors will say that they “train to policy.” While new policies concerning fair and impartial policing are predicates for training, it is important to note that they also serve as a basis for discipline in situations in which officers fail to comport their behavior in accordance with such policies upon completion of training.15

Changes in policy that emphasize fairness can help enhance trust between officers and members of the public by not only signaling to officers the importance of fair and impartial behavior as a key part of their mission but also by clarifying important legal constraints on everyday policing activity such as stops, arrests, and searches, which in too many places are governed by vague constitutional standards that are not constrained by department policy. As social psychologist Jack Glaser and public policy scholar Amanda Charbonneau have recently explained, vague constitutional standards combined with unspecific policy provide a foundation for broad police discretion to engage in behaviors subject to bias.16 They emphasize the role that uncertainty plays in exacerbating individual biases and therefore the negative consequences of that bias, explaining that narrowing the scope of police behaviors can reduce rates of racial disparity that police action can produce. They argue that policy change is a step in this process.

It is difficult to assess the extent to which these policy changes are successful independent of evaluations of subsequent training. Since policy change is a bedrock component of the Department of Justice’s “pattern or practice” program, one might look to assessment of the impact of consent decrees for some sense of the success rates of policy change. In his recent review of two decades’ worth of federal consent decrees, historian and criminologist Samuel Walker notes that it is fair to conclude that decrees yield positive results, especially with respect to use of force and discrimination claims.17 In support of his conclusion, Walker provides evidence of public polling among different racial groups in Los Angeles rating the LAPD much more highly after a consent decree. Since racially disparate treatment undermines trust, we can safely argue that changes in policy that limit ordinary policing in terms of prevalence and depth, along with increases in yield for searches, should be associated with increases in trust and confidence.18

The most important aspect of promoting trust-based approaches is the training of agency personnel. I use importance in two senses here. First, training is the most common approach to establish the importance of trust among agency personnel. Second, agency leaders see training as the most likely mechanism to lead to behavioral change in a world where changing agency personnel is difficult to implement due to union rules and regulations, and where legal liability, whether criminal or civil, is rare. Overall, both systematic reviews of procedural justice theory and meta-analyses of the existing evidence find positive associations between the procedural justice or injustice people experience when dealing with the police and their perceptions of the police, their support for cooperative behaviors, and whether they say they trust the police.19 Individuals who perceive interactions with the police as more procedural have more positive perceptions of legitimacy, as well as increased satisfaction with police services, disposition in interactions, and trust and confidence in the police. In their review of the research examining police-led intervention programs that aim to strengthen police legitimacy, criminologist Lorraine Mazerolle and colleagues found that interventions improve perceptions of procedural justice, as well as satisfaction with, confidence in, and compliance and cooperation with the police.20 Given that the goal of this essay is to explore potential mechanisms for enhancing public trust in police, the focus here will be on trainings as interventions that could exploit the theoretical framing laid out above.

There is academic literature covering rigorously studied effects of procedural justice training on officer and civilian attitudes, officer behaviors, and administrative policing outcomes. Evaluations of procedural justice trainings include studies of script-based trainings, whereby police are taught to use brief, procedurally just scripts (that is, texts) in traffic stops or other similar settings in which interactions are short and relatively homogeneous. Other studies have evaluated trainings that focus more broadly on the development of procedurally just policing practices through the use of lectures, discussions, and exercises that offer participants the opportunity to practice and refine these skills. Some research demonstrates that procedural justice training can positively influence officers’ attitudes about the importance of procedural justice in their work—a critical first step in the process of motivating officers to value enhancing trust over their efforts to reduce crime at all costs.21 As an example, political scientist Wesley Skogan and colleagues evaluated the short- and long-term effects of a police training program in the Chicago Police Department that aimed “to present procedural justice principles to officers as tactics that would encourage the public to recognize the police as a legitimate source of authority, resulting in improved officer safety, more compliance with their instructions, and greater cooperation from the public.”22 Short-run survey-based comparisons for approximately 2,700 officers suggested that training had a positive and statistically significant effect on officers’ perceptions of the importance of various procedurally just behaviors (neutrality, respect, trust, and voice). A longer-term survey (with a 28 percent response rate) suggested that these attitudinal changes persisted. Similarly encouraging findings are reported on a suite of trainings coordinated in six cities across the country by the National Initiative for Building Community Trust and Justice. The evaluation found statistically significant improvements in officers’ self-reported attitudes toward procedural justice.23

Additional studies have found that civilians’ views of police are more positive after interactions with officers trained in procedural justice principles. For example, the Queensland Community Engagement Trial (QCET) in Australia is the first randomized field trial to test the effect of procedural justice training on citizen views of the police. In their field study, Lorraine Mazerolle and colleagues have explored how brief, positive, and procedurally just police-citizen interactions influence people’s legitimacy-related perceptions of the police, as well as their satisfaction and willingness to cooperate with the police.24 They found that respectful treatment positively influenced specific and general views of the police. A subsequent analysis of QCET found that Australian drivers exposed to the experimental group of officers in the trial had higher trust and confidence in police than those exposed to the control group, though no significant differences were found for obligation to obey police or willingness to cooperate with law enforcement.25

Another set of studies reveals the impacts of procedural justice training on administratively measured outcomes of police-citizen interactions. Criminologist Emily Owens and colleagues evaluated a procedural justice intervention in which supervisors were instructed to treat officers in a patient, respectful, and procedurally just manner. The intervention appeared to impact officers’ encounters with citizens, as reflected by decreased arrest rates. The treatment group was less likely to resort to arrests in the week following their meeting (by 25 percent, relative to a pre-intervention incident-level arrest rate of 6 percent). Looking six weeks before and after their meeting, this result diminished, but officers who completed the training still demonstrated a 12 percent reduction in arrests.26

Further, sociologist George Wood and colleagues evaluated the implementation over four years of a one-day procedural justice training in Chicago, which emphasized policing strategies that create appropriate voice, neutrality, respect, and trustworthiness in community interactions.27 Nearly 8,500 officers participated in the training program. Taking advantage of the phased rollout across the department, the researchers evaluated whether the training had effects on cluster-level outcomes, including complaint records relating to officer conduct, civil litigation settlement payouts, and officer use of force. Significant treatment options were identified for each of these outcomes and, over two years, treatment reduced complaints filed against officers by 10 percent and use-of-force reports by 6 percent (corresponding to 11.6 fewer complaints and 7.5 fewer use-of-force reports per one hundred trained officers).

In addition to procedural justice, we can examine another popular recommendation that aims to address trust-related problems between civilians and policing organizations: the process of civilian review. A common prompt for such a recommendation has been an incident of violence between a police officer and a member of the public, and the public perception that the officer in question has not been held accountable for their actions. In response, members of the public sometimes point to civilian review of official disciplinary channels as a remedy. Indeed, more than two hundred of these boards have proliferated across the country in the last decade.28

In reality, civilian review boards rarely achieve the ends that such public calls seek. In many cases when an organization’s leader seeks to separate an officer on the basis of violating conduct, civilian review boards will overturn the executive’s decision. It is rare for a civilian board, even when empowered to do so, to impose more stringent punishment than organizational leaders will. But civilian review boards’ focus on punishment instead of policy, and individual instances of misconduct instead of widespread organizational and policy change, is backward-looking rather than forward-looking. It is exceedingly difficult to change behavior in an organization by addressing individual instances of misconduct in contrast to imposing regulations that seek to change behavior in a forward-looking way by imposing high standards, which, of course, is highly relevant to policy change. A recent article published by the Council of Criminal Justice illustrates the ways in which even the civilians who push for these institutional mechanisms are increasingly disillusioned by them. Their survey of oversight agencies demonstrates a large majority (78 percent) reporting that police executives listen carefully to their recommendations; however, less than half (46 percent) of the respondents believe that police executives frequently implement the recommendations.29

A different approach is to promote policy-making through these boards. In contrast to the backward-looking and more individual-centered proposals that civilian review boards regularly undertake, what legal scholars Barry Friedman and Julian Clark call “community advisory boards” provide more front-end accountability by being more broadly engaged with a town’s policing agency, building trust relationships between the agency and citizens and collaborating with the agency to help solve problems.30 Most of these boards are volunteer-oriented and advisory only. It is rare for these boards to proactively create or pass binding policies and directives that police departments they engage must follow. Even when these boards do have that power, it appears such powers are rarely used.31

One of the most well-known boards that possesses binding authority on an agency is the Board of Police Commissioners in Los Angeles, California. Another board was recently inaugurated in Chicago, the country’s third-largest city.32 And smaller cities also have begun to establish forward-looking policy-making boards. In November 2020, Portlanders (Oregon) overwhelmingly voted to pass the Police Oversight Board Charter Amendment, establishing a new police oversight board with the power to recommend new police policies and directives for the City Council—not the police bureau—to approve. This new board also restricts membership to individuals who lack either employment or familial ties to law enforcement. Despite its two-year existence, Portland’s board has yet to make any significant policy recommendations.

So far, the strategies I have outlined that address trust in policing have focused on how the policing service carries out tasks long associated with a primary goal of the agency, law enforcement, and shaping and structuring those tasks in ways that enhance public legitimacy according to the ideas of social psychology described above. The reality, however, is that normative conceptions of legitimacy that we might seek to measure through positive empirical methods are challenged by the role that settlement and chattel enslavement—and their ideological counterpart, race—have played in the construction of the very laws that the policing service has historically enforced and still does. Thus, an important consideration for improving trust between police and members of the public is reform or elimination of the laws police officers are sworn to uphold.

While the passage of the Reconstruction Amendments at the end of the Civil War and the subsequently passed civil rights laws of the 1960s could be said to have removed white supremacy from the literal text of the criminal law, the structure and attendant culture of racial caste that three hundred and fifty years of law had already built remained embedded in legislation and law enforcement. For example, in the antebellum period, when the state was involved in punishment of an enslaved person—or more commonly, reserving “justice” for the enslaved as a task to be meted out privately—the law made formal distinctions in punishment for the same conduct as between enslaved Black people and whites.33 Imprisonment was reserved for white people, as the punishment of liberty deprivation required a person be free and recognized as a citizen of the state. These formal distinctions were removed from state criminal law after the Civil War when Confederate leaders rewrote their state constitutions. Nonetheless, even after the removal of formal distinctions by race in the criminal code, distinctions by race still were encoded in the law sub rosa and also through the law’s operation.

As is well-known today, the practice of convict leasing functionally re-created slavery in many parts of the South for decades after Reconstruction through enforcement of criminal law, as the Thirteenth Amendment to the Constitution permits involuntary servitude for convicted criminal offenders.34 Less well-known is the role that state constitutions played in formally legitimizing discrimination that the U.S. Constitution did not prohibit. This provision of South Carolina’s 1895 Constitution is illustrative:

The Penitentiary and the convicts thereto sentenced shall forever be under the supervision and control of officers employed by the State; and in case any convicts are hired or farmed out, as may be provided by law, their maintenance, support, medical attendance, and discipline shall be under the direction of officers detailed for those duties by the authorities of the Penitentiary.35

Just as corrections officers and facilities were given broad discretion over those imprisoned, so, too, were police officers given discretion to enforce the law in ways that typically yielded broad inequities by race. Well into the 1960s, police officers used both “field interrogations” and vaguely worded loitering and vagrancy statutes to both harass and intimidate people of color and other groups the officers deemed marginal.36 Even after the Supreme Court invalidated loitering and vagrancy laws as inconsistent with federal due process guarantees in 1972, while also placing constitutional limits on the ability of police to utilize field interrogation at will in the famous 1968 case Terry v. Ohio, the turn toward policing for crime reduction put pressure on law enforcement agencies to engage in proactive policing strategies that relied, at least in part, on tactics such as stop and frisk. Stops and frisks are consistent with the Constitution so long as a police officer has a reasonable belief that the person they are about to detain is about to engage or has engaged in a crime, and every state and locality in this country has managed to criminalize all manner of low-level behaviors—with specificity.

Thus, although the operation of criminal legal processing is formally democratic, it operates in conjunction with antidemocratic structures and culture. Low-level criminal laws might appear on their face to be devoted to public safety in service of a goal that majorities support through processes typically considered democratic. But these laws in operation and through their very DNA perpetuate and support structures of inequality. Consider that while laws prohibiting loitering and vagrancy have long been identified as suspect in the context of advancing the democratic project, contemporary ordinances designed with specificity—prohibitions on selling loose cigarettes on a street corner, or limits on grass lawns exceeding certain lengths—have not typically been considered to fall into this suspect category because their specificity has traditionally been thought to resist the expansion of police discretion that was the clear concern of those who sought to abolish the vagrancy and loitering laws of old.37 The proliferation of even specific prohibitions vastly expands the power of enforcers to enforce laws, and this is an overlooked reality. The old concern about the potential for a police officer to create law and then enforce it is not the problem.38 Rather, the rule of law becomes a mockery of itself when the enforcer has a smorgasbord of petty laws to choose from to validate a forcible arrest. Eric Garner lost his life for selling loose cigarettes on a street corner. As my colleague Stephen Carter puts it, “Every new law requires enforcement; every act of enforcement requires the possibility of violence.”39

But Carter is only half right. He is correct that laws require enforcement, but it does not follow that every act of enforcement requires the possibility of violence. That would only be true if one believed that compliance requires the threat of violence, which in turn is true only if one believes, as the theory of deterrence specifies, that people comply with the law because they fear the consequences of failing to do so. Tyler’s foundational and long-standing work points in a different direction: one that places emphasis on trust and legitimacy in securing compliance, cooperation, and engagement with legal authorities.40

Recent reform efforts support Tyler’s view emphasizing the importance of trust in securing compliance as well as disabling the threat of violence as central to law enforcement. That is, many agencies now embrace strategies that enhance trust as a better, more efficient, and less expensive way to achieve crime reduction and law compliance. To that end, the Baltimore Police Department is now largely prohibited from using forcible arrest to enforce a number of low-level crimes.41 In Berkeley, California, police are now limited in their power to enforce ordinary traffic violations, the consensus being that police intervention should be reserved for dangerous drivers.42 In Virginia, the General Assembly banned police from pulling people over for exclusively minor traffic violations.43 While legalization of possession of marijuana is now the majority position among the fifty states, Oregon is the first state to decriminalize possession of small amounts of drugs such as heroin and methamphetamine. Shifting attention to legislative efforts that restrain the initial grant of power police have over citizens is probably the most important effort we can undertake to enhance trust in government and therefore the state. Best of all: these efforts will not detract from, but very likely enhance, the goal of having safer and healthier communities.

Building trust between legal authorities and members of the public is a cornerstone of strategies that promote safe communities, but trust does more than just satisfy these instrumental goals. Police officers are state authorities who play a critical role in helping people decipher their environment and where they fit in society. Criminologist Ian Loader and sociologist Aogán Mulcahy put it this way:

[Police are] an interpretive lens through which people make sense of, and give order to, their world . . . a vehicle that enables individuals and groups to make sense of their past, form judgements on the present, and project various imagined futures. As an institution intimately concerned with the viability of the state . . . policing remains closely tied to the maintenance of ontological security, the production of subjectivities, and the articulation of collective identities.44

To ensure equality among all members of society, we need a better understanding of how and when critical state actors and members of the public create and maintain trust relationships. Elsewhere, historian Benjamin Justice and I have argued that the project of criminal justice—a normative project, not a descriptive one—must be concerned with the civic educational implications of the law and how it is enforced, as opposed to mere compliance. Given the role of police encounters in shaping individuals’ perceptions of civic identity, it is time for us to contemplate just how police educate citizens and the potential for that education to foster trust.45