Cities and Policing for Crime Prevention: Refocusing the Agenda to Maximize Benefits and Minimize Unintended Harms

Abstract

Over the past 20 years, there has been growing concern over a purported link between proactive policing to control crime and unfair, biased, and abusive policing approaches. Overly aggressive and indiscriminate policing initiatives run the risk of driving a wedge between police and the communities they serve, with residents of disadvantaged neighborhoods feeling less like partners and more like targets. At the same time, research suggests that effective policing has an important impact on public safety. We present a conceptual framework for how police can prevent crime while minimizing unintended harm. We focus particularly on cities and urban communities, arguing that the benefits of policing can be maximized and the costs of policing can be minimized when the police respect individual rights and dignity and focus on community, and that problem solving should begin with a focus on risky people in the places that generate the most crime.

Long-standing rifts divide the police and many of the communities they are charged to serve and protect. These ongoing tensions heighten in the aftermath of tragic citizen deaths during police encounters, as evidenced by the intense protests and calls for police reform following events in Ferguson, Missouri; New York City; Chicago; Minneapolis; and elsewhere in the United States. Over the past decade, the policing profession has been embroiled in a legitimacy crisis. Police legitimacy is achieved when community members agree that police play an appropriate role in making and implementing rules governing public conduct (Tyler 2006). While many factors influence police legitimacy (Bottoms and Tankebe 2012), police departments need to develop, implement, and sustain crime prevention practices that are fair, lawful, and effective. As suggested by the National Research Council’s Committee to Review Research on Police Policy and Practices (2004, 2), “policing that is perceived as just is more effective in fostering a law-abiding society, and that success in reducing crime enhances police legitimacy.” Former President Obama’s Task Force on 21st Century Policing (2015, 42) also noted that, “law enforcement’s obligation is not only to reduce crime but also to do so fairly while protecting the rights of citizens.”

Black Americans and, to a lesser extent, Hispanic Americans are, on average, more likely than other groups to live in neighborhoods characterized by higher spatial concentrations of poverty, unemployment, joblessness, family disruption, and geographic isolation (Sampson et al. 2018). With respect to lethal violence, young minority men face especially large and disproportionate burdens. In 2020, Black males between the ages of 18 and 34 accounted for just 2 percent of the U.S. population but suffered roughly 40 percent of gun homicide victimizations (Braga and Cook 2023). In some communities in cities like Chicago and Philadelphia, young men face a greater risk of being shot than members of a U.S. combat infantry brigade deployed to Iraq during the height of the U.S.-led war in that country (del Pozo et al. 2022). This long-standing problem was exacerbated in the aftermath of the COVID-19 pandemic, as the bulk of the 2020–2021 increase in shootings was largely concentrated in the top decile of census block groups in Los Angeles, New York City, and Philadelphia, with the shooting victims in these hot-spot areas disproportionately Black and Hispanic (MacDonald et al. 2022).

Given the dramatic degree of racial inequality in lethal violence, reductions in gun violence in U.S. cities will disproportionately benefit economically disadvantaged minority communities. Research finds that when cities expand the size of their police force, for example, homicide rates fall to a greater degree among Black than among white Americans (Chalfin et al. 2022). More generally, a litany of evidence in criminology and economics indicates that crime declines when cities hire more police officers (Chalfin and McCrary 2017; Evans and Owens 2007) and concentrate them in high-crime areas (Braga and D. Weisburd 2022; MacDonald et al. 2016). While research suggests that staffing levels matter, evidence also suggests that the types of crime prevention strategies used by the police play the largest role in achieving lasting crime reduction (Kim et al. 2024; Lee et al. 2016). Whatever our approach—whether maintaining appropriate staffing or investing in more evidence-based strategies—we are left with the enduring question of how public policy can secure the benefits of the police and minimize harmful intrusions on the public they are entrusted to serve. This issue is urgently important for all Americans, but it is especially critical for the economically disadvantaged as the effects of policing—both good and bad—are inevitably magnified by social inequalities.

A litany of social science evidence suggests that police can be more effective in controlling crime by focusing on the relatively small number of high-risk people and high-risk places that generate a widely disproportionate share of urban crime problems (Braga 2008; MacDonald 2024; D. Weisburd 2015). At the same time, due to the potentially high burdens of police activity in a community, policymakers and social critics have suggested that police executives and policymakers should be encouraged to think critically about policing tactics that focus outsize attention on particular people and places (Young and Petersilia 2016). Critics contend that excessive surveillance and enforcement are practiced by police departments in too many urban neighborhoods that already suffer from serious crime problems; this, they say, contributes to diminished police legitimacy and to mass incarceration (Alexander 2010). Beyond the normative judgments that underlie this perspective, the empirical literature shows considerable support for the premise that disparities in police practices cannot be attributed exclusively to crime control considerations. When a city grows its police force, for example, racial disparities in arrest rates for lower-level crimes appear to grow (Chalfin et al. 2022), an outcome that widens the reach of the criminal justice system to include young minority men with relatively low risk of future offending (Graef et al. 2023). This dynamic is particularly true when policing is subjected to a “crime numbers game” managerial mindset, associated with the proliferation of CompStat across multiple police departments, that promotes yearly increases in arrest, summons, and investigatory stop actions as key performance measures (Eterno and Silverman 2012).

The burdens of this kind of enforcement ultimately fall disproportionately on those who are least able to sustain a costly shock, and recent research finds that individuals who are arrested, detained, and prosecuted, on average, end up being more criminally active in the future (Leslie and Pope 2017), an unwelcome “backfire” effect that runs the risk of ending up with the worst of both possible worlds. Such backfire effects might be worth sustaining if arresting large numbers of people were to generate an equally large deterrence effect among the remainder of the population, but there is little evidence that this is the case. The available research has largely found that aggressive enforcement practices associated with large numbers of stops and low-level arrests have limited benefits when it comes to public safety (see Braga, Brunson, et al. 2019; Tebes and Fagan 2024). There are likewise concerns that when so many people have acrimonious engagements with police officers, the mission of police to provide for public safety is compromised, either by reducing legitimacy (Tyler 2006) or by reducing ordinary citizens’ willingness to report crimes or serve as witnesses (Brunson and Wade 2019). A systematic review of research studies of the widespread use of police stops suggests that individuals stopped by the police are more likely to experience mental and physical health issues, such as depression, anxiety, and sleep deprivation, relative to individuals who were not stopped (D. Weisburd et al. 2023); this finding suggests that for the marginal person not engaged in criminal activity, a police stop may generate negative health consequences.

While the preceding discussion is grim, a growing body of research evidence suggests that the police do not have to choose between excessive policing of low-level crimes and disengagement from vigilant activity in communities of concentrated disadvantage. Both incarceration and crime rates can be reduced by policing strategies that change active offenders’ perceptions of apprehension risk and that modify opportunities for crime. Durlauf and Nagin (2011) conceive of such an outcome as a “double dividend,” whereby good policing can generate dual benefits for society—less crime and fewer people in prison—with few tradeoffs.

Such a claim starts from an important insight from several recent papers showing that when cities expand the size of their police force, crime declines modestly, without an increase in arrests for the types of serious crimes that account for most prison sentences (Chalfin et al. 2022). This effect follows, in part, from the fact that only a small share of arrests is for serious crimes. It also reflects the possibility that greater police presence generates a deterrence effect by dissuading people from committing crimes in the first place (S. Weisburd 2021). Additionally, because crime is so highly concentrated among a small number of young men within a few communities, a marginal increase in the arrest rate of these individuals can generate considerable crime reductions. When it comes to serious crimes like murders and shootings, even in large cities like New York or Chicago, only a few thousand people among millions drive much of the risk (Tracy et al. 2016). The concentration of crime by people and place implies that policing could become more tailored and generate many crime control benefits while subjecting fewer people to costly enforcement actions. A focus on high-rate offenders and the places where they operate is not limited to the most serious crimes, like murders and shootings. Even for less serious crimes, like burglary and shoplifting, research consistently finds that a small number of people drive a large share of the problem. With respect to burglary or theft from a vehicle, many of these are crimes of opportunity—the result of an open window or an unlocked car door. However, a large share of burglaries will inevitably be driven by a small number of highly productive offenders, professional burglars who hit several houses per week (J. Roman et al. 2009). Likewise, with respect to shoplifting, while many Americans have stolen something from a store at some point in their lifetime, most of the problem is caused by a small number of serial thieves operating as part of organized theft rings (Hegarty 2010).

Strategies that focus on the concentration of crime among high-rate offenders can focus on a relatively small number of people. When coupled with a strong commitment to community and problem-oriented policing, “high-risk places, high-risk people” strategies can also improve police legitimacy by exposing far fewer and less serious offenders to the criminal justice system. For example, a district attorney might find that many low-level offenders can be diverted without driving up crime to a substantial degree so long as law enforcement is able to successfully prosecute the key drivers of specific crime problems. In the remainder of this article, we review the available scientific research on crime and policing and suggest a roadmap for how police resources should be allocated to maintain public safety in U.S. cities. We focus on a small number of best practices that are most likely to be cost-effective and transmittable across different policy contexts—and thus that can have an impact on crime at scale.

High-Risk Places (and Times)

Crime is highly concentrated in specific hot-spot locations within cities. These hot spots are typically defined as clusters of street addresses, groups of street blocks, and particular street intersections and street segments (D. Weisburd et al. 2009). For instance, only 3 percent of the addresses in Minneapolis accounted for 50 percent of the crime calls to the police (Sherman et al. 1989). Researchers have further demonstrated that the concentration of crime at specific places is stable over time (e.g., see D. Weisburd et al. 2012). Indeed, some 5 percent of street corners and block faces in Boston generated 74 percent of the city’s fatal and nonfatal shootings between 1980 and 2008, with the most-active 65 locations experiencing more than 1,000 shootings during that period (Braga et al. 2010). This long-term stability suggests that high-crime places are not likely to disappear without focused attention. Together, these findings have been described by D. Weisburd (2015) as an empirical “law of crime concentration” (see also D. Weisburd et al. 2024).

The persistent concentration of crime at specific small places is largely generated by highly localized crime opportunity structures (e.g., see Braga and Clarke 2014; D. Weisburd et al. 2012). The localization is such that even within a given high-crime neighborhood, large and stable differences appear in crime risk from block to block. The characteristics and dynamics of places are key factors in explaining clusters of criminal events (Eck and D. Weisburd 1995; Felson 2006). For example, a poorly lit street corner with an abandoned building, located near a major thoroughfare, provides an ideal location for a drug market (Braga 2008). The lack of proper lighting, an abundance of “stash” locations around the derelict property, a steady flow of potential customers on the thoroughfare, and a lack of defensive ownership (informal social control) at the place all generate an attractive opportunity for drug sellers. The compelling criminal opportunities at the place attract sellers and buyers and thus sustain the market. Similarly, prospective robbers may repeatedly visit a bus stop to identify distracted and/or intoxicated commuters who pass through a dark alley to reach their eventual destination. The dark alley provides a location to commit robberies that conceals the offender’s immediate whereabouts from commuters and is shielded from the watchful eye of the police and neighborhood residents. Prospective robbers may consider the risks of apprehension (e.g., police presence, neighbor surveillance, presence of video cameras [St. Jean 2007]), the effort required to complete the robbery (e.g., strong-arming or using an available weapon), and rewards of success (e.g., wallet with cash and credit cards, smartphone) that follow their assessment of the situational dynamics in a bus stop and alley area (Clarke 1997).

Crime problems are often concentrated during specific days of the week and times of the day that arise from the situational characteristics and dynamics of places (Felson and Poulsen 2003). These temporal patterns of crime can be powerfully tied to the place-level dynamics of “activity spaces” (Felson 2006) and the idea of crime hot spots. Juveniles, for instance, are attracted to very specific activities such as “hanging out” at malls and movie theaters. The timing of school hours and business hours of popular hangout places can lead to concentrations of potential offenders and potential targets in space and time. Research on juvenile crime in Seattle provides strong confirmation of the relevance of juvenile activity spaces to understanding the concentration of juvenile arrests over time at specific places (D. Weisburd et al. 2012). The most active juvenile crime hot spots were much more likely to have arrest incidents committed at schools, youth centers, shops/malls, and restaurants, as compared to lower-activity juvenile crime hot spots. Research, for example, shows that when underperforming high schools close, crime drops significantly at those locations (Steinberg et al. 2019).

High-Risk People

Empirical research has demonstrated that a small proportion of the population, and of offenders, generate large shares of crime problems. A systematic review of 73 studies in the U.S. and other countries found that 10 percent of the most criminally active people from populations that included offenders and nonoffenders accounted for around 66 percent of crime (Martinez et al. 2017). The review also found that the most active 10 percent of offenders accounted for around 41 percent of crime committed by all offenders (Martinez et al. 2017). Highly active offenders are not specialists and tend to commit a wide range of crimes (Spelman 1990). For instance, the criminal careers of most gun offenders are characterized by a wide range of offense types, including armed and unarmed violent offenses, illegal gun possession offenses, property offenses, drug offenses, and disorder offenses (Braga and Cook 2016).

High-rate offenders often commit their crimes with associates or “co-offenders.” Nearly half of U.S. robberies are committed by groups of offenders, and the chances of a successful robbery are greatly enhanced when accomplices are involved and when guns are used in the commission of the act (Cook 2009). Youth often commit crimes in groups and account for a disproportionate share of serious crimes (Zimring 1981). Most studies find between 50 and 75 percent of juvenile crimes are committed in the company of others (Warr 2002). Youth gun violence is usually concentrated among groups of serious offenders, and conflicts between youth street gangs have long been noted to fuel much of the serious street violence in major cities (Howell and Griffiths 2018). A study of more than 20 cities found that, on average, less than 1 percent of a city’s population was involved in gangs, drug crews, and other criminally active groups, but these violent groups were connected to more than 50 percent of a city’s shootings and homicides (Lurie 2019).

A small number of people also suffers a disproportionate number of victimizations. A systematic review of 20 victimization studies completed in the U.S. and other countries found that only 5 percent of the total population experienced nearly 62 percent of all victimizations reported (O et al. 2017). Exposure to potential offenders varies with the characteristics of the victim (e.g., age, race, place of residence) and lifestyle routines (Hindelang et al. 1978). Drinking alcohol in public or using public transportation late at night, for example, can increase an individual’s risk of victimization. For individuals with particularly risky lifestyles, the probability of victimization can be quite high. For example, during Boston’s youth homicide epidemic in the early to mid-1990s, youth gang members stood a roughly one-in-seven chance of being killed at some point during an average nine-year period of gang membership (Kennedy et al. 1996). Youth gunshot-wound victims treated in Boston emergency rooms often had scars from past gun and knife wounds (Rich and Stone 1996). In a study of one disadvantaged Boston community, roughly 85 percent of all gunshot victims were in a single co-offending network dominated by gang members and representing less than 5 percent of the community’s population (Papachristos et al. 2012).

Repeat victimization and repeat offending patterns influence the spatial distribution of crime in cities. It is well known that crime hot spots are often populated by specific targets that repeatedly attract offenders (Braga 2008). For instance, an apartment building could have a particular unit that is burglarized repeatedly. The same apartment building could have another unit that generates multiple calls to the police because of a violent husband who repeatedly abuses his wife and children.

How Risks Are Policed Matters

These crime concentration patterns represent important opportunities for more effective crime prevention and control. If police departments can organize themselves to control the small number of risky places, risky times, and risky people who generate the bulk of a city’s crime problems, they can more effectively manage citywide crime trends. However, how the police choose to address these recurring problems may either improve or damage their relationships with residents in the places with the highest concentrations of crime. Police departments can adopt crime prevention strategies that seek to engage the community in changing the underlying conditions, situations, and dynamics that cause crime problems to recur. Alternatively, police departments can simply “put cops on dots” through directed patrols or carry out enforcement blitzes aimed at potential offenders in high-crime areas. Unfortunately, these kinds of enforcement initiatives sometimes become unfocused in practice as entire neighborhoods may be defined as hot spots and young minority males simply using these public spaces may be regarded as potential high-risk offenders.

Citizens’ appraisals of the police are influenced by the style of policing in their communities. Policing strategies that emphasize increased investigative stops, criminal summons, and misdemeanor arrests across jurisdictions have been shown to generate notable racial disparities (Braga, Brunson, et al. 2019). And, when these sorts of tactics are deployed excessively, this results in a phenomenon the popular culture refers to as overpolicing. When overpolicing is combined with occasional disrespectful treatment by the police, negative views of the police can intensify (Brunson 2007). Hence, aggressive policing strategies for minor offenses can set the stage for increased acrimony between the police and the public they are charged with serving and protecting. Moreover, if the police are focused too much on low-level offenses, they may underpolice problems of violence that are concentrated in high-risk places and among high-risk people and reduce their ability to mitigate upticks in violence (Brunson 2020). For instance, a police officer who has processed an arrest for a lower-level crime may be unavailable to deter or otherwise deal with a more serious crime several hours later. Black Americans living in U.S. cities frequently express dissatisfaction regarding delayed response times, uncertain prioritization of calls for service, and the overall perception that police are not committed to solving serious crimes that have been reported in their community (Brunson 2007).

It is not surprising that residents of distressed, high-crime neighborhoods consistently report higher levels of dissatisfaction with the police and often blame the police for persistent crime and disorder problems (Weitzer and Tuch 2006). At the same time, residents of crime-plagued neighborhoods often call for greater police presence. In fact, one study found that 85 percent of Hispanics and 88 percent of Black Americans favored more police surveillance of high-crime areas (Weitzer 2010). Another more recent study suggests that Black Americans preferred to maintain (or increase) police patrols even if crime was declining and new police reforms were not enacted (Balcarová et al. 2024). All communities regardless of income or ethnicity seem to want more effective police and equal treatment under the law.

How Risks Should Be Managed by Police

Both criminological theory and the available empirical evidence suggest that the benefits of police are maximized when they are focused on high-risk people and places. What specifically should the police do to ensure that focused policing is successful? Is there a general set of principles and best practices that will apply broadly to many types of places and agencies?

First, community policing should be the foundation of any focused crime prevention strategy. While simply engaging the community doesn’t seem to translate directly into crime reduction gains (Gill et al. 2014; NASEM 2018), developing close relationships with community members helps the police gather information about crime and disorder problems, understand the nature of these problems, and develop specific strategies to address them. Community members can also help with key components of strategies tailored to specific problems by making improvements to the physical environment and engaging in informal social control of high-risk people. Indeed, a central idea in community policing is to engage residents so they can exert more control over situations and dynamics that contribute to their own potential for victimization.

Community engagement in developing appropriately focused strategies helps to safeguard against indiscriminate and overly aggressive enforcement tactics, which, in turn, erode the community’s trust and confidence in the police. Collaborative partnerships between police and community members improve the transparency of law enforcement actions and provide residents with a much-needed voice in crime prevention work. Ongoing conversations with the community can ensure that day-to-day police-citizen interactions are conducted in a procedurally just manner that enhances community trust and compliance with the law (Tyler 2006; D. Weisburd et al. 2022). In short, community engagement forms the basis for effective policing and is a necessary component of an effective and sustainable law enforcement strategy (for a discussion, see Lum et al. 2016).

Second, policing is more effective when law enforcement leaders and mid-level managers rely on data to identify and analyze the concentration of crime by place and people. Data might refer to highly detailed administrative data on the location of crimes, the modus operandi of offenders, or connections between suspects from prior law enforcement engagements. A focus on data can also mean that a department is investing in better information flow within the agency or is developing mechanisms to communicate more effectively with members of the community, with other law enforcement personnel, and with prosecutors. Practical prevention frameworks, such as problem-oriented policing, can be used to diagnose and respond to high-crime places, times, and offenders. Problem-oriented policing—which uses the iterative process of problem identification, analysis, response development, and assessment to generate crime prevention impacts (Goldstein 1990)—should be a central part of community policing efforts (Skogan 2019). Research shows that problem-oriented policing is an effective strategy for reducing a wide range of crime and disorder problems (NASEM 2018).

Third, it is important to recognize that effective police crime prevention efforts are characterized by changing the perceptions of potential offenders of apprehension risk and by modifying criminal opportunities (Nagin et al. 2015). While arrests are inevitable, police should be oriented toward preventing crimes from happening in the first place. A department that prioritizes deterrence will value having officers remain visible in high-crime areas more than having them make an arrest for simple drug possession—an action that takes an officer’s eyes off the street for several hours or more. Similarly, a department that prioritizes deterrence will train officers to prioritize their attention on community problems that have a potential connection to serious violence or loss of life. An out-of-control after-hours party that spills out on to the street might initially generate only a noise complaint, but a well-trained officer will understand that this is precisely the type of event that carries an outsize risk for violence.

Policing at crime hot spots

Hot-spots policing works to reduce crime without displacing offenders to nearby places (NASEM 2018). An ongoing systematic review of 65 controlled studies (including 27 randomized experiments and 38 quasi experiments) finds that hot-spots policing is an effective crime prevention strategy (Braga, Turchan, et al. 2019). A more recent meta-analysis of hot-spots policing found that this tactic was associated with a 16 percent reduction in crime outcomes at treated hot spots relative to control hot spots (Braga and D. Weisburd 2022). These crime reductions tend to diffuse into surrounding areas rather than causing significant spatial crime displacement. Because offenders tend to be strongly tied to place for a variety of reasons (D. Weisburd 2015), even rational offenders are not likely to be easily displaced to another social environment. In line with the deep ties between crime and place that have been observed in the research, large-scale randomized controlled trials and agent-based simulation models seem to suggest that scaled-up hot-spots policing programs can be used to control jurisdiction-wide crime levels (Ariel, Sherman, et al. 2019; D. Weisburd et al. 2017).

Problem-oriented policing to change the underlying environmental conditions that cause an area to become a crime hot spot appears to be a particularly effective strategy (Braga and Bond 2008; Braga, Turchan, et al. 2019). Police presence in crime hot spots can change offender perceptions of risk without generating mass arrests or subjecting large numbers of people to investigative stops. While in these places, police can work with other city agencies and community groups to change the physical and spatial characteristics of hot spots, such as poor lighting, abandoned buildings, disorderly bars, and the like, that attract potential offenders. Relative to traditional police tactics, such as increased patrol, police efforts to change the underlying crime opportunity structures may also produce longer-lasting crime prevention impacts.

Chronic problems at hot spots may require an initial response that leans heavily on enforcement activities, but the proximal causes of crime in those communities should be addressed at the same time through the development of problem-solving interventions aimed at crime reduction. For instance, recurring gun violence takes a devastating toll on urban communities, and police need to respond immediately to outbreaks of shootings to prevent further loss of life. Solid evidence indicates that stops and frisks of potential offenders in gun violence hot spots increases gun seizures, reduces shootings, and are welcomed by residents (Cohen and Ludwig 2003; McGarrell et al. 2001; Rosenfeld et al. 2014; Sherman and Rogan 1995). These highly focused stops can be used in the short run to generate much-needed immediate relief to residents suffering from the ongoing trauma of repeated shootings. To minimize the negative impacts of stops on the well-being of stopped individuals, it is critically important for police departments to focus their actions on high-risk offenders in gun violence hot spots during time periods when shootings are at the greatest risk. This strategy requires that police engage in additional intelligence-gathering and analysis as well as getting to know area residents and others who routinely use these public spaces. Upfront community engagement, training in lawful and bias-free policing, and close supervision of officers assigned to gun violence hot spots are also essential. Such steps are necessary to avoid indiscriminate and unfocused enforcement efforts that will undermine police legitimacy. As place-based, problem-oriented interventions are implemented and safer public spaces created, police can be more economical in their use of stops to deter high-risk people from carrying guns.

Policing high-risk offenders

Police can also help reduce crime when they focus on the high-risk people who generate a disproportionate amount of offending and victimization in cities. For instance, during the early 2010s, in the aftermath of a federal court decision that curtailed the ability of the New York City Police Department to continue to rely on very large numbers of street stops, especially in low-income minority communities, the department embraced a “precision policing” strategy designed to reduce shootings by using highly coordinated and targeted “takedown” operations on violent gangs in public housing (Bratton and Murad 2018). Recognizing that gang members represented the bulk of shooting victims and offenders in and around public housing developments in the city, the department doubled the size of its antigang unit and invested more in its capacity to identify the most violent gangs and launch major criminal investigations. Targeted gang members were arrested on felony and criminal conspiracy charges and, as appropriate, prosecuted in both state and federal court. The New York City Police Department’s more precise and targeted policing strategy reduced shootings in and around public housing developments by about 25 percent between 2012 and 2020 (Chalfin et al. 2021).

Recognizing the importance of targeted enforcement as well as the promise of the community’s “moral voice,” focused deterrence represents another problem-oriented approach to changing the criminal behaviors of repeat individual offenders, gangs, and other criminally active group members through the strategic application of law enforcement, social service, and community-based action (Braga and Kennedy 2021). Relative to law enforcement takedowns, focused deterrence is more economical in its use of sanctions against offending groups as swiftness and certainty are emphasized over severity. Additionally, focused deterrence strategies explicitly seek to avoid indiscriminate and unfocused enforcement efforts and can improve police legitimacy in the highest-crime places (Braga et al. 2018; Brunson 2015; Meares 2009). Community members (including family), street outreach workers, and churches are also involved in focused deterrence efforts to provide a moral voice that assists in empowering informal social controls by the community and to offer repeat offenders help. A key element of focused deterrence involves the communication of potential enforcement consequences associated with continued offending and the availability of services, opportunities, and other forms of help. In the eyes of community members, there is an inherent fairness in offering targeted offenders a choice and providing resources to support their transition away from violent behavior rather than simply arresting and prosecuting them.

Based on a meta-analysis of 24 quasi experiments, the most recent systematic review of focused deterrence studies concluded that these programs generated moderate crime reduction impacts (Braga et al. 2018). Focused deterrence programs have produced significant violent crime reductions in challenging urban environments such as Baltimore ¹ ; Chicago (Papachristos and Kirk 2015); Oakland, California (Braga, Zimmerman, et al. 2019); Philadelphia (C. Roman et al. 2019); and Stockton, California (Braga et al. 2024). Several randomized controlled trials of focused deterrence have been completed. In New York City, Aboaba et al. (2023) found individuals who were called in to attend notification forums (antiviolence meetings with representatives from law enforcement, social service, and community-based agencies) were substantially less likely to violate their parole—an effect driven primarily by reductions in violations due to absconding. However, the program did not affect individual arrest rates or neighborhood-level gun violence. In Sacramento, Ariel, Englefield, et al. (2019) reported arrests decreased by 21 percent among prolific offenders who were randomized to a policing program comprising preemptive engagements by a police officer offering both carrots (desistance pathways) and sticks (increased sanction threats) relative to a control group that received no additional attention. Finally, in Chicago, Davis et al. (2025) found that a focused deterrence effort reduced violent arrests by 43 percent among treated high-risk youth relative to control high-risk youth.

Conclusion

Cities rely on their police departments to create safe public spaces. Considerable scientific evidence shows that proactive policing is effective in controlling crime and can be done in ways that minimize unintended consequences. Police abolitionists and activists that want to “defund the police” suggest a variety of nonpolicing alternatives as viable substitutes for police intervention. While a large-scale redirection of funding from police to social services has never been attempted and thus lies beyond the ability of current empirical research to test, the available evidence suggests that a great deal of caution is warranted. First, it is unclear whether the most commonly suggested alternatives to law enforcement will enhance public safety because the preventive effects of popular community violence interventions are mixed (e.g., street outreach and “violence interrupter” programs [Hureau et al. 2023]) or unknown (e.g., hospital-based violent intervention programs [Webster et al. 2022]). Second, even when effective strategies are identified, the challenges of scaling up social service–based interventions raise the enduring question of their ability to impact safety at a citywide level. Finally, it is important to note that research on community-based interventions has implicitly held police staffing, strategy, and tactics constant. As such, the available research does not speak to the ability of such interventions to be an effective substitute for law enforcement. Coupled with the recognition that the demography of crime victimization is tremendously unequal, any strategy to enhance public safety without the use of law enforcement must grapple with the reality that the people who bear the risks of such a strategy are the most vulnerable and disadvantaged in our society. In our view, while recognizing the critical importance of social investments in communities, cities must continue to support well-staffed police departments and require police executives to embrace police reforms, including the adoption of evidence-based community problem-solving strategies to manage high-risk places and high-risk people.

Police departments have myriad competing priorities that include emergency calls for service, preventive patrol, criminal investigations, traffic enforcement, and security at public events. Police are also the residual claimants on nearly every social problem that has not been adequately addressed by other city agencies, including especially challenging problems like homelessness, mental illness, and substance abuse. As Thomas Sowell (1993, 83) notes, the “first lesson of economics is scarcity,” and there is “never enough of anything to satisfy all those who want it.” Given that resources will always be scarce, the police must make tradeoffs in deciding how to allocate their efforts. To make the best use of their available resources to reduce the social costs of crime, police can apply those resources narrowly rather than broadly and focus on prevention whenever possible.

The police occupy a unique position within American society, but the management challenges within the profession are common ones that can be found elsewhere in public service and in industry. For example, consider medical interventions: While modern medicine can now treat large numbers of serious diseases, those treatments often come with considerable financial costs, and even the most effective treatments may have side effects that are as serious as the disease itself. Physicians often try to identify the least costly ways for addressing serious diseases and—critically—they capitalize on research: Government, universities, and philanthropies have poured billions of dollars into research and development that inform patient care, and clinical trials have allowed scientists to develop new drugs and therapies that can improve the practice of medicine over time. And given that medical interventions are difficult and costly, particularly for patients, good physicians urge preventive care whenever possible.

Policing faces challenges that are nearly identical to those found in medicine—only in a more difficult political environment. Most Americans will agree that it is necessary to treat a serious illness and recognize that they do not have the expertise to develop an informed opinion about the nature of the treatment. On the other hand, there is far less consensus on how to treat crime, and because most Americans have an opinion about the most effective treatments, there are far more popular debates on the issue than there are on medical interventions. Compounding the already tricky politics of public safety, criminal justice interventions have considerable side effects. Where most ineffective medical treatment harms only an individual patient, an ineffective criminal justice strategy can produce large-scale compounding side effects—especially in communities that suffer most from endemic crime and violence.

Observers on the left and the right use different heuristics to guide their thinking about why crime happens and what to do about it. While the debate remains unresolved, the key to maintaining public safety while minimizing the harms of enforcement is to invest in good science and provide evidence-based information to the people with the most skin in the game. The science of policing is not yet mature and we have much to learn, but what we do know suggests that a fruitful path forward is to focus efforts on developing problem-oriented strategies that reduce crime among active offenders in the places with the highest concentration of crime.