April 9th, 2024 29 Minute Read Report by David Weisburd, Clair V. Uding, Kiseong Kuen, Beidi Dong

“It’s Not as Bad as People Think the Place Is” The Potential for Informal Social Control at Crime Hot Spots

Executive Summary

Criminologists have long known that crime is highly concentrated—in most cities, approximately 50% of crime happens on only 5% of streets. As a result, many cities have effectively reduced crime by concentrating police resources in extremely high-crime “hot-spots.”

Despite calls for defunding the police, policing is necessary for public safety on hot-spot streets, and people who live on these streets recognize that. Indeed, in one survey in Phoenix, Arizona, the majority of residents of hot-spots streets said they wanted the amount of police on their block to be increased. But ultimately, the bedrock of social control of crime is not the police, but citizens who exercise informal social control.

The practice of community policing—which emphasizes building relationships with members of the community and allowing them to have a voice in defining which problems police should focus on—attempts to leverage this fact to enhance public safety. However, it is often assumed that high-crime hot spots are simply so chaotic and disordered that the residents are incapable of playing a key role in crime prevention.

We examined this assumption through a multiyear study of more than 300 hot-spot streets in Baltimore, Maryland (all of which were in the top 3% in rates of violent or drug crime in the city), as well as almost 50 “cold” streets, which had little to no crime, and 100 “cool” streets, which fell in the middle.

We found that common assumptions about hot spots—which we shared at the outset—are not accurate. Residents of hot-spot streets in Baltimore have strong social and network ties with their neighbors, and they exercise meaningful levels of informal social control. Even on the most crime-ridden blocks, there is potential for collaboration between residents and police.

Our findings suggest that it is time to think about how both informal and formal social control can work interactively to reduce crime. Hot-spots policing is an important first step, because the level of danger in these places can simply make it too dangerous for residents to exercise the type of informal social control that they would like. But residents of these streets, contrary to stereotypes of academics, policymakers, and practitioners, form an important resource for reducing crime on these streets through informal social control.


One of the most important discoveries in criminology over the last few decades has been that crime is highly concentrated within cities. [1] Evidence has repeatedly shown that, in cities across the U.S., approximately 50% of the crime is produced by only about 5% of street segments. Indeed, this finding has been replicated with such consistency that it has been deemed the “law of crime concentration.”[2] As a result, many police departments have implemented a strategy known as “hot-spot policing,” which focuses on specific streets where crime rates are high[3] and which has been shown in a large number studies to be an effective way to reduce crime without displacing it to nearby areas.[4]

Despite the evidence of the effectiveness of hot-spot policing, many have expressed concern about the impact that it has on communities. Some scholars and pundits have argued that the benefits of policing hot-spot streets are outweighed by the negative consequences of heightened police enforcement.[5] They often portray the police as a foreign invading force on high-crime streets[6] and point to evidence of negative health and delinquency outcomes associated with aggressive enforcement strategies, such as pedestrian stops or stop-question-frisks (SQFs).[7] These arguments have been used to support efforts to “defund the police” and end hot-spot policing programs.[8]

Most communities, however, do not seem to be convinced that the drawbacks of policing outweigh its benefits.[9] In a recent study conducted by Weisburd and the National Policing Institute in violent and drug crime hot spots in Phoenix, residents were asked whether they wanted “more police,” “about the same number of police,” or “less police” on their block.[10] Only 6.6% of respondents wanted fewer police, 34.7% wanted about the same number of police, and 58.7% wanted even more policing. Individuals in these communities face problems that cannot be handled by themselves and their neighbors alone—and thus they do not want the police withdrawn from their streets.

More generally, there has been ample research dating back decades about the ways in which communities want to be partners with the police: they want to be consulted about the problems that they face and help develop programs that prevent crime.[11] This approach, known as “community policing,”[12] rejects the traditional policing model that leaves the community out of the prevention equation. Rather, it gives residents a voice by allowing the community to play a role in determining which problems the police should be prioritizing, and it encourages collaborative problem-solving efforts with the community. Like the procedural justice movement, community policing emphasizes that the community must be treated with respect and dignity.[13]

A strong partnership between police and community is particularly important because research has consistently demonstrated that informal community social controls are an essential component of crime prevention.[14] As James Q. Wilson and George Kelling explained in their seminal article on “Broken Windows” policing, over 40 years ago,[15] the bedrock of control of crime is not the police, but citizens: “The essence of the police role in maintaining order is to reinforce the informal control mechanisms of the community itself. The police cannot, without committing extraordinary resources, provide a substitute for that informal control.”[16]

Wilson and Kelling believed that reducing crime and disorder through policing was critical—partly because it is a necessary first step in a process that allows for strengthening informal social controls over crime.[17] High-crime hot spots, however, are often portrayed as chaotic, hopeless, and disordered and, as a result, any attempt to increase informal social controls is seen as a futile endeavor.

These streets are places where crimes occur on a regular basis, often with clear signs of social and physical disorder,[18] where concentrated disadvantages abound and where mental health problems are more severe.[19] In some cities, single hot-spot streets can produce hundreds of crime calls to the police in a single year.[20] There is a perception that street and gang culture is the norm in these disadvantaged places and that residents endorse violence and criminal activity to resolve conflicts.[21] And as William J. Wilson argued in his 1987 The Truly Disadvantaged, the impoverishment, structural deterioration, and isolation in these places transcends generations, which gives a sense of permanency to the problems that communities face.[22]

Although these descriptions give us insight into the hardships faced in high-crime areas, they have inadvertently hampered efforts to engage these communities in efforts to prevent crime. By reinforcing a perception of communities with little ability to advance informal social controls, and with little capability to contribute to collaborative problem-solving programs, they have often led to a paternalistic approach to crime prevention that leaves hot-spot residents out of the crime-prevention equation.

Our research contradicts this perception of hot-spot communities. They come from a study we recently completed in Baltimore—a city that has received much negative media attention for its high crime and disrepair.[23] Former president Donald Trump, for example, described Baltimore as “the worst city in the U.S.” and a “disgusting, rat and rodent infested mess” where “no one would want to live.”[24]

In our study, we analyzed about 300 hot-spot streets in Baltimore (all of which were in the top 3% in rates of violent or drug crime in the city) as well as almost 50 “cold” streets, which had little to no crime, and 100 “cool” streets, which fell in the middle.

As we have documented in previous research in other contexts, hot-spot streets differ significantly from cold spots.[25] In Baltimore, some of the hot-spot streets reported hundreds of crimes to the police in 2012 (the year our study began). On the cold streets, there were fewer than four violent or drug crime calls in 2012. There were also large and significant differences between the streets in terms of concentrated disadvantages. For example, 26% of residents on hot-spot streets had less than a high school diploma and 41% reported that they were not working; on cold streets, only 6.5% had less than a high school diploma and 15.4% were not working.

At the outset of our study, we—like many others—assumed that the residents of the streets with the most serious violent and drug crime would have little ability to exercise informal social controls or participate in community improvement or crime-prevention efforts.

But that was not what we found. As we report below, our quantitative and qualitative findings suggest that residents of hot-spot streets in Baltimore have strong social and network ties with their neighbors. They exercise meaningful levels of informal social control, albeit less than residents of non-hot-spot streets. Even on Baltimore’s most crime-ridden blocks, there is a latent potential for collaboration between residents and police, as well as other local agencies. There is much promise here that we did not expect and much hope about the ability of these communities to play a role in improving where they live.

The Study

At the outset, it is important to explain the nature of the data we collected. Until our study, there was almost no systematic knowledge about the people who live in crime hot spots.[26] There was some ethnographic work but little quantitative data, largely because the census, due to privacy issues, does not release information at that micro-geographic level.

We focused on street segments—a portion of a street between two intersections—because a series of studies have shown that street segments serve as social as well as physical geographic units.[27] We wanted to uncover the social context of such places and therefore sought to collect data on residents of hot-spot streets. We defined residential hot-spot streets as those that were in the top 3% of streets for violent and drug crime and included at least 20 occupied households. We sampled from three types of hot-spot streets: those that met only the 3% criterion for drug crime (120 segments); those that met the criterion for only violent crime (126 segments); and streets that met both criteria (55 segments). As noted earlier, we also collected data on 46 “cold streets,” which had little or no calls to the police for violent and drug crime, and 100 “cool streets,” which, while not “cold,” did not reach the threshold of hot-spot streets.[28]

For each of these streets, we sought to conduct door-to-door interviews of 7-10 respondents who were at least 21 years of age.[29] We conducted surveys in three waves, which took place in 2013–14, 2015–16, and 2017–18. The cooperation rate for all three waves of the survey was above 55%, which is generally considered very good for door-to-door surveying.[30] A total of 3,738 residential surveys were conducted in wave 1, and 3,615 surveys were completed in wave 2. In both waves 1 and 2, an average of eight surveys were completed on each street. In the final wave of data collection, 3,141 surveys were conducted, with an average of seven surveys completed on each street. As far as we know, this project—which required substantial research funding provided by the National Institutes of Health—is the largest such study on crime hot spots ever done.[31]

While our main objective was to develop quantitative portraits of attitudes and experiences of communities on these streets, we also conducted qualitative ethnographic field research in a subsample of 68 streets. Researchers visited a random sample of cold/cool streets, drug hot spots, and violent hot spots to engage with people on the streets and conduct semi-structured interviews.

The Latent Potential of Crime Hot Spot Communities: Social Ties and Social Networks

Social ties and social networks have long been linked to informal social controls in the community.[32] It is assumed that when social ties and social networks are stronger, communities will be able to marshal informal social control more effectively.

We measured social ties on a 4-point scale, based on three questions from the survey that asked how often residents: 1) “chat with neighbors”; 2) “visit with neighbors”; and 3) “help each other out.”[33] Response options included “never,” “rarely,” “sometimes,” and “often.” Scores were based on the average response to all three questions, where “never” corresponded to 1 point, and “often” to 4 points.

The differences in social ties between residents of cold, cool, and violent/drug hot-spot streets were small (see Table 1). In the first and third wave, there were no statistically significant differences, and even in the second wave—where significant differences were observed—the streets vary only between 2.82 on this scale (the violent spots) and 2.95 (the cool spots).[34]

Table 1

Analysis of Variance (ANOVA) for Social Ties

 Type of Street Segment 
  (N = 46)(N = 100)(N = 120)(N = 126)(N = 55) 
  Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)F-score
TimeWave 12.98 (0.36)2.92 (0.35)2.96 (0.36)2.84 (0.40)2.92 (0.40)2.13
Wave 22.93 (0.37)2.95 (0.35)2.93 (0.35)2.82 (0.37)2.83 (0.34)2.67*
Wave 32.88 (0.37)2.88 (0.39)2.86 (0.40)2.78 (0.40)2.72 (0.41)2.30
*p < .05; overall mean for wave 1 = 2.91, overall mean for wave 2 = 2.89, and overall mean for wave 3 = 2.83.

Among those living on hot-spot streets, 48.5%–54.7% of respondents (across the three waves) often chat with their neighbors, 20.3%–25.1% often visit their neighbors, and 37.6%–42.9% often help one another out. These social ties among neighbors of hot-spot streets suggest a latent potential to exercise informal social controls.

When it comes to the size of residents’ social networks—measured by asking residents how many of their neighbors on the street they consider to be friends—we again find similar results across hot spots, cool spots, and cold spots. Across all three waves, there were no significant differences between the different types of streets.

Table 2

ANOVAs for Social Network

 Type of Street Segment 
  (N = 46)(N = 100)(N = 120)(N = 126)(N = 55) 
  Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)F-score
TimeWave 14.60 (2.29)5.28 (2.90)5.47 (2.97)4.81 (6.18)5.58 (3.60)0.76
Wave 26.24 (13.62)5.36 (3.35)5.59 (3.82)4.79 (3.60)4.86 (2.83)0.79
Wave 34.74 (2.78)5.18 (3.18)5.54 (3.71)4.62 (3.68)5.30 (5.07)1.10
Overall mean for wave 1 = 5.16, overall mean for wave 2 = 5.29, and overall mean for wave 3 = 5.09.

These quantitative results were backed up by accounts we heard in the qualitative interviews, in which many residents of hot-spot streets discussed knowing and spending time with their neighbors. A male in his fifties who lived on a violent hot-spot street for over 20 years talked positively of the relationships between neighbors, even with changes in the area. He said that he had known his neighbors for years—most of whom were old people and families—and that he has had no issues with them: “During the summertime, we usually gather in one of the neighbors’ houses every weekend to have some drinks and socialize.” He said that most of the residents in that area were homeowners, along with a few renters, but that “it didn’t make any difference from homeowner or a renter. Everybody is good to each other, and we rarely encounter issues.”

A female in her fifties who lived on a violent crime hot-spot segment described the diversity of her neighbors and how they spent time together. She said that “it was a diverse neighborhood with a Korean family, Caucasians, a Chinese family, and black families, too. They all got along, they came out and had barbecues together, went to the playground together, and did a lot of social activities together.” Another resident of a violent segment said that he knows all the neighbors and “loves them all.” He moved to the street with his family a long time ago and has never left the area, and he said that older people get along more easily than the younger generations because the older people are more familiar with one another.

Translating Latent Potential into Informal Social Control

In addition to measuring social ties and social networks, which reflect latent potential for informal social control, we directly measured informal social control. Despite the key importance of informal social control in crime prevention, most scholars until the late 1990s measured community social control in relation to social structural indicators, such as concentrated disadvantage or residential instability.[35] However, in a seminal 1997 study, Sampson, Raudenbush, and Earls proposed a direct measure of informal social control, called “collective efficacy,” that is conceptualized as the willingness (and ability) of neighborhood residents to take action and intervene.[36] The measure, which has become widely accepted, emphasizes the capacity of a community to realize common values and regulate behavior through cohesive relationships and mutual trust among residents.[37

Collective efficacy measures informal community social control by combining two scales: one of willingness to intervene and the other of social cohesion and trust. We followed Sampson et al.’s (1997) approach in assessing levels of informal social control on street segments.[38]

Table 3 shows the percentage of residents who responded who agree or strongly agree with each specific item on the scale. For readability, results are provided only for wave 1, but the results from other waves are similar. Differences in social cohesion and trust are statistically significant across the street types, with the largest differences between the combined, violent, and drug spots and the cold spots.[39]

However, it would be a mistake to conclude that levels of social cohesion are negligible on hot-spot streets. Even on the violent and drug hot-spot streets, more than 75% of residents say that they are willing to help their neighbors, and only about 30% say that people usually do not talk to one another. These findings emphasize that trust may be less evident on hot-spot streets than on streets with little or no crime, but these are not places wholly devoid of social cohesion and trust.

Table 3

ANOVAs for Individual Measures of Collective Efficacy (wave 1)

 Type of Street Segment
Social cohesion and trust     
People on your block are willing to help their neighbors***90.984.079.077.575.6
Neighbors do not usually talk to one another on your block (reverse-coded)*78.476.276.871.373.0
In general, people on your block can be trusted***84.369.660.551.449.1
People on your block usually do not get along with one another (reverse-coded)***90.081.474.371.371.9
People on your block do not share the same values (reverse-coded)***65.752.849.344.646.5
Neighbors watch out for one another on your block***90.684.281.176.778.3
Willingness to intervene     
If some kids were skipping school and hanging out on your block?***64.262.856.353.055.7
If a group of kids were spraying graffiti on a building?***92.386.579.076.076.1
If a teenager were showing disrespect to an adult?68.871.768.366.866.9
If there were a fight in front of your home?***88.383.075.373.170.8
If a group of kids were climbing on a parked car?***
If the local fire station was going to be closed down because of budget cuts?***76.272.867.465.064.4
*p < .05, **p < .01, ***p < .001.

Measures of willingness to intervene show a similar set of relationships. Again, the combined drug and violent crime hot spots have the lowest levels of willingness to intervene, and the cold spots have the highest levels. The differences are significant across the street types in every case, except for willingness to intervene if a “teenager was showing disrespect to an adult.” On that question, about 70% of all respondents on our street segments agree or strongly agree. Again, it is important to note that the levels of willingness to agree are relatively high even for the combined drug and violent crime hot spots. More than 60% of residents agree or strongly agree with almost every item.

Even though there is more social cohesion and willingness to intervene on hot-spot streets than we might have expected, we still find strong and significant differences in overall collective efficacy across the street types (Table 4). In each wave, we observe a consistent relationship between street type and collective efficacy. The combined violent and drug crime streets evidence the lowest levels of collective efficacy or informal social control, followed by the violent crime hot spots, and then the drug crime hot spots. Cool spots have higher collective efficacy than hot spots, and cold spots have the highest levels overall.

Table 4

ANOVAs for Collective Efficacy

 Type of Street Segment 
  (N = 46)(N = 100)(N = 120)(N = 126)(N = 55) 
  Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)F-score
TimeWave 13.95 (0.26)3.77 (0.25)3.61 (0.31)3.53 (0.30)3.52 (0.32)25.50***
Wave 23.94 (0.26)3.79 (0.26)3.65 (0.31)3.56 (0.33)3.43 (0.24)27.57***
Wave 34.09 (0.21)3.87 (0.27)3.71 (0.32)3.63 (0.29)3.54 (0.30)32.45***
***p < .001; overall mean for wave 1 = 3.65, overall mean for wave 2 = 3.66, and overall mean for wave 3 = 3.74.

Active Guardianship

Another way to directly measure informal social control is to ask whether people act as guardians for others on the street. In our survey, we asked residents how common it was for neighbors on their street to watch one another’s home or property when they were out of town. Here, again, we find consistent and significant differences across the types of streets we examine, with residents on cold streets more likely to act as guardians for neighbors.

However, most residents on the hot-spot streets also said that it was common or very common to watch out for their neighbors’ home or property; in each wave, over 70% of residents on the combined drug and violence hot-spot streets answered that they watched their neighbors’ houses. Again, informal social control is less prevalent on hot-spot streets, but it is certainly not absent.

Table 5

ANOVAs for Active Guardianship

 Type of Street Segment 
  (N = 46)(N = 100)(N = 120)(N = 126)(N = 55) 
  Mean (SD)Mean (SD)Mean (SD)Mean (SD)Mean (SD)F-score
TimeWave 184.8 (19.5)83.6 (15.6)78.6 (17.5)75.8 (19.1)72.7 (19.3)5.46***
Wave 283.2 (20.9)82.2 (17.0)79.0 (18.6)73.6 (18.6)74.7 (17.6)4.54**
Wave 386.6 (14.4)84.3 (15.4)80.3 (17.8)77.5 (18.6)74.3 (16.7)5.43***
**p < .01, ***p < .001; percentage of residents who responded “common” or “very common.” Overall mean for wave 1 = 78.8, overall mean for wave 2 = 78.1, and overall mean for wave 3 = 80.3.

We saw these trends reflected in more detail in our qualitative interviews. For example, a female in her twenties living in a drug hot spot described concerns about safety for her children on the street but still emphasized that “the neighbors generally look out for each other and were close and a strong community.” She said that a lot of people had lived in the area a long time and she felt as though people were looking out for her. She also suggested that, in some ways, concentrated disadvantage can foster social controls: “people help each other because they know what it’s like . . . to be hungry and to struggle so they want to do what they can to keep someone else from having to go through that.” Another woman who lived in a violent crime hot spot said that “the area was friendly and people took care of each other, watched out for each other and each other’s kids.” She even suggested that residents were able to leverage this social capital to protect their street from criminals: “At one point, people tried to bring drugs and crime into the housing, but the neighbors got together to make sure that didn’t happen and pushed them out.”

A resident of a drug hot spot described how he and a woman down the street got together to organize neighborhood meetings and coordinated residents to do trash pickup to keep the neighborhood clean and prevent problems such as rats. He said his neighborhoods did a good job of following through and picking up the trash when they were supposed to. He hoped to do more organizing in the community in the future but said that for now, residents were just focused on trash pickup. A business manager on one of the violent segments described the amount of violence that occurred on the street and how ineffective the police were at responding. He added: “In the past, people used to be afraid when a [shooting] took place, but now, people act fast. They provide first-aid kits and try to help the victim until the ambulance arrives.”

Field researchers also directly observed other aspects of informal social control during the qualitative data collection. For instance, while being interviewed, a hot-spot resident found a package slip on the ground outside his house. He checked the house number and realized that it was for his neighbor. He kept it and said that he would give it to his neighbor later and ensure that he knew that he had a package being delivered.

At a different drug hot spot, the field researchers were interviewing a man who worked on the street and a friend of his. During the interview, another man parked his car and approached the worker and his friend to ask whether they could watch his car while he ran a couple errands in the area. They told him: “We already saw you parked your car. Don’t worry, we will watch it for you.” On another street, when the field researchers were returning to their car, a resident volunteered that she had been watching their car for them while they were gone “to make sure no one tried anything.”

Discussion and Conclusions

We think that our findings bring an important correction to the images of high-crime streets in the academic and popular literature. Despite frighteningly high levels of crime, residents of hot-spot streets have similar levels of social ties and social networks as residents of non-hot-spot streets. Residents of hot-spot streets do evidence lower levels of informal social control than streets with little crime. But just because there is not as much informal social control does not mean that there is none. These are not places of chaos and hopelessness; they possess meaningful levels of informal social control and guardianship.

The robust social networks that already exist in these places may also serve as the raw ingredients necessary to develop even greater levels of informal social control. This finding is particularly important. There is a strong relationship between social ties and informal social control during waves of our study. For example, in wave 1, the correlation between social ties and informal social control was 0.50, generally considered a meaningful effect in standard metrics in social science research.[40] However, strong social networks and social ties may be a necessary, but insufficient, condition for informal social control. Something is missing in hot spots, as contrasted with cold spots. It may be that the high levels of threat and violence on hot-spot streets impede informal social controls. Or it may be that the lack of social and political capital in these places restricts their ability to successfully exercise informal social controls.

What has struck us most in analyzing our quantitative and qualitative data is the degree to which residents of these high-crime streets have the potential to be partners in the prevention process. Fully involving them in that process, however, requires abandoning the paternalistic images that many scholars, policymakers, and practitioners bring to their efforts to reduce crime in these places.

Many residents of hot-spot streets are optimistic about the possibilities for their communities. One resident told us—in a quote that forms part of the title of this report—that he was aware of the perception that most had about neighborhoods like his but that “it’s not as bad as people think the place is.”

It is not that residents of high-crime streets do not recognize the problems on their street; the point is that, despite these problems, residents cared for their community and held out hope that things could get better. For example, one resident of a violent segment talked openly about weekly occurrences of violence and murders nearby but said that he still “loves [the street]—with concerns.” He talked about the sense of neighborhood and cooperation among the homeowners and about how he is coordinating a community organization to try to do the work that the city is not doing, such as cleaning and community policing. He was carrying papers on which he had written his observations about the area that he planned to discuss with neighbors next month, at their first meeting. He said that he did not know how successful the effort would be but that he is “crossing his fingers.” There is an opportunity to leverage “activists” like this resident, who are eager to create change, in hot-spot communities.

On a violent-crime hot spot, another resident highlighted the relationship between informal social control and community resources: “The main problem was that there were no recreation centers, nothing for the kids to do…. There was a park nearby, but it wasn’t so safe, and it was only swings and a few pieces of equipment. Kids didn’t have any role models or anyone to look up to, no programs or sports to play to keep them interested in things other than drugs.” He said that if there were anything that he would change, it would be to put in a recreation center and more community centers nearby so that there would be places for people to talk about issues and get together on doing something about them.

In our view, these requests reflect the need to invest resources on these streets. However, deficits in informal social control mean that residents of these streets are likely not as active in requesting help from the city and are not as effective in marshaling help from the city. City resources are often brought as a response to effective community activism, and residents of hot-spot streets are much less likely than those of non-hot-spot streets to have the social capital and social organization to exert such pressures. On the same street where a resident discussed a need for community centers, another resident said that he had called the city about repairing a gas leak in the vacant house next to him, and “it still wasn’t fixed.”

We think that city government should invest not just in policing these streets, but also in other city services, such as parks, recreation, and sanitation—areas in which hot-spot streets are often neglected.[41] Repairs and improvements to the environment in these places would be well received by residents and enable them to strengthen informal social controls. For example, studies show that greening/restoring vacant lots can reduce crime and improve public health outcomes.[42]

Beyond providing services, city government can help more directly by trying to reinforce the informal social controls that already exist in these places. Imagine, for example, that cities created crime-prevention units made up of social workers, psychologists, community organizer and other professionals that focused on capitalizing on the latent potential for informal social control on these streets.[43] These units would help neighbors organize and help them understand how to best engage the police and other city agencies in ways that are acceptable to them.

In Israel, Yamit Elfassi, former director of a community crime-prevention office for the government, provides insight into how such an effort might be organized. Working in Dimona, a development town in Israel, she encountered a housing project called Kovshei Eilat. The residents suffered from crime and drug problems, and the area was blighted with trash, a lack of lighting, and other problems typical of crime hot spots. After meeting with residents, Elfassi realized that simply bringing more police would not address the neighborhood’s more general problems.

Drawing on the concepts of collective efficacy and informal social control, Elfassi decided to organize residents for collective action. They had community meetings and organized projects to repair public spaces and work on green spaces. She also hired civilian crime-prevention officers, who helped the residents understand how to gain resources for their work and obtain contacts for them with city agencies. Working together with residents, they eventually were able not only to repair the blight at the hot spot but also work with police to reduce crime. Importantly, the police participated in meetings and efforts, not necessarily as crime experts but simply as people (for example, playing piano or otherwise helping at the meetings). But this allowed police to see the potential for the community to work with them—and for residents, it improved perceptions of the police and created natural networks for cooperation. This program was seen as a tremendous success[44] but was discontinued before it could be expanded across the city.

As this example shows, what is needed in crime hot spots is for the police and city government to recognize—as residents already do—that such places can be improved and that residents can play a key role in that process if they receive help in translating latent potential for informal social controls into effective prevention. One reason this is often hard to accomplish is simply the level of danger on these streets. Drug dealers and other violent people are commonplace. Our researchers often had to leave streets when there were signs of danger from guns, knives, or gangs. A few times, they were present on streets when guns were fired. For residents, intervention can be risky—which is why we were somewhat surprised by the willingness to intervene that they do express.

Going back to Wilson and Kelling’s (1982) seminal article on Broken Windows, the police must continue to be part of the crime-prevention equation at crime hot spots. This is something that residents of these areas understand, as suggested by the Phoenix hot spot survey results we reported at the outset of this report. Residents of hot-spot streets need the police to deal with serious crime problems on their streets. And if the police are successful, and it becomes safer for residents to intervene, informal social control will be strengthened. We have found that efforts to decrease crime and social disorder lead to increased collective efficacy at crime hot spots.[45] In this sense, the police can plant seeds for marshaling latent informal social control at these places. Without the police, it would be much more difficult for this to happen.

And contrary to some critics, we find that the community welcomes police on hot-spot streets when they come to improve the life of residents. In another study we conducted in Baltimore, police and social workers visited crime hot-spot streets to improve access to mental-health services and to increase collaboration between residents and the police.[46] At the outset, the police were concerned about reactions of residents when they visited the hot-spot streets, where they thought they might not be welcome. In fact, they found just the opposite. The community was happy to cooperate with them.

Our findings support neither defunding the police nor any agenda that suggests that hot-spot residents cannot help determine how to address problems on their streets. As we noted at the outset, residents of hot spots want more policing, not less. And our findings show that hot-spot residents express meaningful levels of collective efficacy and informal social control. In sum, our findings suggest that it is time to think about how both informal and formal social control can work interactively to reduce crime.

So what should mayors do to reduce crime in their cities? They should develop hot-spot policing programs, which have shown strong evidence of crime reduction.[47] They should also build in to those programs the principles of community-oriented policing and procedural justice. A large multicity randomized trial has shown that a few days of training in respectful policing can improve police–citizen interactions, citizen attitudes toward the police, and increase the crime-reduction outcomes of hot-spot policing.[48]

Police should be aware of the importance of building collective efficacy and informal social controls at hot spots and how that will strengthen crime-control efforts.[49] But mayors should also consider hot-spot efforts that more directly attempt to foster informal social control, such as those carried out at Kovshei Eilat in Israel. Such efforts may not even require additional resources, which could be prohibitive for cash-strapped city governments. In Dimona, the resources were drawn from an ongoing prevention program. Like hot spots policing programs, these efforts are likely to demand a reallotment of city resources, rather than additional new funding.

Police and city government would do well to recognize the latent ability of residents of hot-spot streets to play a key role in crime prevention. This would not only strengthen crime-control efforts but would also reduce frictions between the police and the community at crime hot spots.


We would like to express our appreciation to Danielle Rudes, who designed the qualitative data collection for this project. We would also like to thank Hannah E. Meyers and Charles Fain Lehman for their comments on earlier drafts of this work.

About the Authors

David Weisburd is distinguished professor of criminology, law and society at George Mason University, and Walter E. Meyer Professor Emeritus of Law and Criminal Justice at the Hebrew University. His main interests are in policing and the criminology of place. He is the recipient of the Stockholm Prize in Criminology, the Sutherland and Vollmer Awards from the American Society of Criminology, and the Israel Prize.

Dr. Clair V. Uding is an associate professor in the Department of Criminal Justice and Sociology at the University of Wyoming. Her research focuses on crime and place/communities, the criminal justice system’s response to mental health problems, and issues in police legitimacy. Currently, her work examines police responses to mental health crises, specifically co-responder programs, as well as health issues and other social problems at crime hot spots.

Kiseong Kuen is a doctoral candidate in the Department of Criminology, Law and Society at George Mason University. He will join the School of Criminology and Criminal Justice at Griffith University in Australia as a lecturer (equivalent to assistant professor in the U.S.) in July 2024. His research interests include crime and place, social and health issues at crime hot spots, evidence-based policing, police-citizen relations, and quantitative methods.

Beidi Dong is an associate professor of Criminology, Law and Society at George Mason University. His research focuses on identifying risk and protective factors for community violence in the U.S. and developing evidence-based strategies to prevent and mitigate the adverse effects of violence on affected individuals and communities. His work also explores health disparities and inequities related to crime, violence, and punishment.


Please see Endnotes in PDF

Photo: Genaro Molina / Los Angeles Times via Getty Images


Are you interested in supporting the Manhattan Institute’s public-interest research and journalism? As a 501(c)(3) nonprofit, donations in support of MI and its scholars’ work are fully tax-deductible as provided by law (EIN #13-2912529).