After more than a decade of relative calm, the past few years have seen both an uptick in violent crime and a proliferation of new approaches to combat it. In an encouraging development policymakers across the political spectrum have committed rhetorically to the position that evidence should guide policy responses to violence. Unfortunately, that stated commitment often does not drive policy choices.
The approach that has received the most attention and investment from federal, state, and local governments in recent years is “community violence intervention” (CVI). Generally, CVI refers to a broad collection of programs and practices united by the idea that community-based organizations can prevent violence by mediating disputes and targeting services toward the people who are most likely to be involved in violence–a spectrum of interventions that includes urban planning, social work, and a range of other activities.
The deployment of street outreach workers (sometimes called “violence interrupters”) is the most prominent component of CVI. By relying on “credible messengers” to navigate and ultimately disrupt disputes, CVI promises to leverage community solutions to violence that do not require additional law enforcement inputs. CVI’s popularity has grown considerably over the past several years. Many cities are launching new CVI initiatives and others are expanding existing programs, often with help from the federal government. Almost $500 billion of American Rescue Plan funds are eligible for CVI initiatives, and the Bipartisan Safer Communities Act of 2022 included $250 million for CVI.
But while the theory behind CVI is intuitively appealing, and some CVI initiatives are promising, the evidence base remains thin. This is especially true for street outreach worker programs. By contrast, “focused deterrence” — a crime response initiative closely linked to CVI that relies on partnerships between social service providers, community actors and law enforcement — has a much stronger evidence base. However, focused deterrence has not received similar attention from policymakers and advocates. The enthusiasm around street outreach work relative to focused deterrence suggests that policymakers are relying on something other than evidence to guide their response to the most recent crime wave. A well-intentioned desire to reduce violence without the harms associated with indiscriminate and unfocused law enforcement has led to the short-sighted exclusion of the police from CVI conversations.
Policymakers committed to evidence-based criminal justice policy should continue to invest in implementing and evaluating street outreach worker programs to determine whether that approach can effectively reduce crime and violence. But they should understand that CVI programs that rely exclusively on community partnerships, street outreach and the provision of social services are an emerging and as-yet unproven means of promoting public safety. If policymakers want to deliver more certain crime reduction to their constituents, focused deterrence is a policy response for which there is currently more compelling evidence.
Harms from crime, harms from bad policing
Since 2020, gun violence has once again risen in U.S. cities. Homicides, most of which are committed with guns, have risen by approximately one-third. In cities like Chicago and Philadelphia, young men living in the most violent communities face mortality risks comparable to those of U.S. soldiers serving in forward areas during times of war. And while national homicide rates remain well below their apex in 1990 and may actually fall in 2023, in several cities, including Chicago, homicide rates in the most disadvantaged communities are actually higher than they were during the peak of the crack epidemic in the early 1990s. For people living in those neighborhoods, appeals to national data are not relevant. For many truly disadvantaged, the “bad old days” have already returned.
Investments in law enforcement tend to be effective in controlling crime. When more police are hired, crime – including serious violence – drops. When police are deployed to crime hot spots, crime declines and when police are called away from their regular beats to respond to a car crash, crime rises in areas where they ordinarily would have been on patrol. However, the effects of police presence alone are fairly modest, and police departments must be very careful to ensure that such initiatives do not devolve into overly aggressive and indiscriminate enforcement efforts that generate unintended harms such as concerning racial disparities in stops, arrests, and imprisonment rates. As evidenced by the 2020 Black Lives Matter demonstrations after the murder of George Floyd, too many residents have experienced abuse and neglect at the hands of police officers. These reasonable concerns have fueled a search for alternative interventions that do not rely on law enforcement actions to reduce gun violence.
Policymakers are grappling with how to turn the corner on rising gun violence without turning back the clock to an era of aggressive order maintenance policing, characterized by the widespread use of street stops and a large volume of arrests. To those who tend to see public policy through the lens of inevitable tradeoffs, this might sound like wishful thinking. But the empirical evidence suggests such an outcome may be achievable for one simple reason: Even in the most disadvantaged and violent communities, gun violence is highly concentrated among a tiny number of risky people. These high-risk individuals tend to be well known to the criminal justice system and are often involved in violent groups and co-offending networks that use guns to settle disputes. The policy relevance of this empirical observation is clear: Gun violence can be reduced – perhaps considerably – by changing the violent norms and behaviors of individuals in this small and identifiable high-risk population. The question then becomes: How?
The promise and perils of street outreach work
Political leaders have recently turned to approaches to reduce gun violence that rely less on law enforcement and more on community partnerships. Chief among this new slate of strategies is street outreach work, commonly called “violence interruption.”
These programs generally seek to reduce violence through two complementary strategies. First, by developing and leveraging relationships with high-risk individuals to convince them to take up social services and opportunities that could positively change their life trajectories. And second, by engaging in actions intended to quash active disputes through conflict negotiation and resolution techniques. Core program activities are performed by street outreach workers who form relationships on the ground and seek to disrupt violence before it happens and social workers who link interested people to services.
Despite the new attention, street outreach work actually dates back at least 100 years. The forerunners of modern gang outreach workers were 19th-century “boys workers” and settlement house workers, whose efforts were defined by the emergence of new institutions, such as boy’s clubs, dedicated toward meeting the societal challenges generated by the rapid urbanization of the period. Guided by the Chicago School of Sociology thinking that gangs could be transformed into prosocial groups, street outreach worker programs became the default policy response to gang problems by the mid-20th century. The appeal of the approach is straightforward. Street outreach work empowers the community to resolve disputes before they lead to violence, without the potential for harm associated with increased law enforcement. As the saying goes, an ounce of prevention is worth a pound of cure.
While the theory behind street outreach programs is intuitively powerful, the quality of the evidence is not as strong as we’d like it to be. Because it is often targeted toward entire neighborhoods–the places that typically experience the most violence–street outreach work is challenging to study using a randomized controlled trial, the “gold standard” of social science and medicine evaluation. However, the available evidence, including at least one recent study on a program in Boston with a strong quasi-experimental research design, is mixed; studies report decreases, increases, and null effects on violence. The evidence from historical street outreach worker programs rolled out during the 1940s and 1950s, is likewise not particularly positive.
A recent and more promising exception is Chicago’s READI program, an initiative created by the nonprofit Heartland Alliance that identifies young people in Chicago who are at the greatest risk of becoming victims of gun violence and offers them a range of social service supports and effective cognitive behavioral therapy (which other research already suggests may offer especially large benefits by giving people the skills to more effectively resolve interpersonal disputes). The Biden administration and others have heralded READI as an effective gun violence prevention program. Indeed, the experimental results of the evaluation of READI are impressive and suggest that while READI does not reduce involvement in violence generally, it may dramatically reduce involvement in shootings and homicides, which are, by far, the costliest crimes.
The READI evaluation also has the distinction of using a randomized trial design and therefore offers a data point that can be interpreted with confidence. However, the program does not fit neatly into the street outreach category as it is much more comprehensive than most programs under this umbrella. A close review of the READI program makes it clear that street outreach workers played an important but relatively small role by making referrals for treatment. The primary intervention is the social service treatment – cognitive behavioral therapy, employment, and other services. READI provides good evidence to support the claim that providing social services and therapy makes people less likely to engage in violence. Still, it does little to validate the street outreach model.
It is also important to note that READI referrals made by street outreach workers involved people who were willing to participate in the program. Providing effective treatment to high-risk people who may be ready to change is clearly an important step toward creating neighborhood safety. Still, it doesn’t reach the people who are not inclined to participate. Moreover, even social programs that proved wildly successful in a local iteration can be challenging to scale up to more locales and participants.
Ultimately, many pressing implementation questions are linked to the uncertain impacts of now-popular street outreach worker programs. For example, should these programs embrace performance management systems; partner with criminal justice organizations; exclusively hire former gang members and convicted felons as outreach workers; embrace cognitive behavioral therapy; and provide stipends to support clients while participating in recommended prosocial activities?
Continued testing of street outreach worker programs will answer some of these questions. However, many cities need immediate relief from persistent gun violence. Given the current state of implementation knowledge, blind investments in street outreach worker programs are unlikely to be best suited to provide that relief. Street outreach worker programs at the moment are very much in an experimental phase. In the language of medical science, we might consider the approach to be undergoing something like a Phase II clinical trial. Given the appreciable benefits that would accrue to an effective street outreach worker program, investment and innovation in the model is justified, and then some. But it is premature to call street outreach work an “evidence-based solution” to gun violence.
The alternative with a track record
While the broader slate of CVI programs is still being refined and tested, a closely related approach known as focused deterrence already enjoys considerable empirical support. Street outreach workers are key participants in focused deterrence strategies that blend law enforcement, social service, and community-based actions. In essence, focused deterrence takes a “carrots and sticks” approach by combining credible threats of sanctions with offers of support in transitioning to a successful “straight” life.
In focused deterrence strategies, specific groups and individuals are identified by their involvement in repeated shootings rather than by their perceived risk or willingness to participate. The people in question are told directly by community, social service, and criminal justice leaders that they are being targeted based on their violent gun behaviors. In these conversations, leaders express concern for the safety and well-being of the community. They offer the targeted individuals a choice between help if they try to change and enforcement if they persist in criminal activity. Street outreach workers participate in adding links to social services and opportunities; law enforcement participates to send a credible threat of detection and punishment. When street outreach and social service provision fails, focused deterrence demands that law enforcement intervene through proactive investigations and other actions targeting the most violent groups.
The program evaluation evidence on focused deterrence programs suggests, on average, important and statistically significant reductions in violence when these strategies are implemented properly. Studies have shown noteworthy violence reduction impacts in challenging urban environments such as New Orleans, Oakland, and Philadelphia. Equally important, the scientific rigor of focused deterrence evaluations has steadily improved over time. Randomized experiments in St. Louis, Sacramento, and Melbourne, Australia, have shown significant reductions in violent behaviors of individual offenders. A multi-site randomized experiment in New York state showed mixed results, with offenders recently released from prison significantly less likely to violate the conditions of their parole but no effects on subsequent arrests.
Across the entire range of literature, including several quasi-experimental studies, focused deterrence appears to be an effective strategy more often than not. The evidence favoring focused deterrence is complemented by evaluations of similar approaches that are more explicitly law enforcement-oriented. These, too, find significant effects on gun violence. For instance, recent research shows that gang sweeps by police have led to sizable reductions in crime in cities as diverse as New York and Barcelona.
While effective, focused deterrence programs need strong leadership and effective management structures to maintain and ensure treatment integrity from the outset of implementation. The approach requires multifaceted intervention activities and a complex interagency structure that makes it hard to implement. When these interventions stumble, it is often due to a lack of ongoing analysis of evolving gun violence problems, inadequate governance structures that broke down after key people moved on, and political in-fighting in lead partnering agencies. Programs in Baltimore, Minneapolis, and San Francisco unraveled rapidly after some encouraging initial crime control success due to serious political problems and a lack of true interagency partnership.
The American Public Health Association (APHA) recognizes “group violence intervention” – a specific kind of focused deterrence program – as an evidence-based CVI to reduce gun violence. Unfortunately, the available APHA materials never mention the police or the mobilization of criminal justice partners to prevent outbreaks of violence; omit communications with offending populations to discourage continued gun violence; and limit discussion to program activities such as providing at-risk group members with mental health support, GED training, and employment opportunities. Providing incomplete accounts of programs to avoid necessary engagement of the criminal justice system is problematic. Focused deterrence should be viewed as a CVI program that includes law enforcement action as necessary and as a complement to street outreach worker programs that maximize their effectiveness.
Crime is a complex social phenomenon, and effective crime reduction efforts must recognize and engage with that complexity. The available social science literature suggests law enforcement and social service provision are not substitutes for producing public safety but instead complement each other in a variety of ways. In short, law enforcement versus community partnerships need not be an either/or question. Focused deterrence offers a blueprint for how law enforcement can be more precise and tailored in proving public safety to disadvantaged communities, while street outreach programs offer hope that communities may one day accomplish public safety with minimal police involvement. In the meantime, any serious discussion of the problem of community violence and our responses to it must include investment in focused deterrence.
Aaron Chalfin is Associate Professor of Criminology at the University of Pennsylvania and a senior fellow with the Niskanen Center.
Anthony A. Braga is the Jerry Lee Professor of Criminology and the Director of the Crime and Justice Policy Lab at the University of Pennsylvania.