Introduction: Mary Lou Leary, Acting Assistant Attorney General, Office of Justice Programs, U.S. Department of Justice, Washington, D.C.

Presenter: David Weisburd, Ph.D., Professor, Department of Administration of Justice, George Mason University, Fairfax, Va.

• Segment 1: Introduction
• Segment 2
• Segment 3
• Segment 4
• Segment 5
• Segment 6
• Segment 7

Introduction

Mary Lou Leary: Good afternoon. And welcome to the session of what we call "Research for the Real World." I'm Mary Lou Leary and I'm the acting Assistant Attorney General at the Office of Justice Programs at the Department of Justice. Thanks to everybody for coming and I want to especially, before we begin, to thank the organizers of the event — Ron Wilson and Yolanda Curtis and others at the National Institute of Justice. And they worked very hard to bring us all together this afternoon.

We have a really interesting program today and we're going to use a little bit different format today. We're going to be talking about place-based policing. It's one of the hottest topics in police research today, and it also conforms with the Obama administration's focus on place-based programming. You may have heard that in August, the White House announced that it would be focusing on place-based policies and looking forward to the fiscal year of 2011. This focus on place-based policing will be an important part of that focus. You've probably heard the Attorney General, too, talk about the need to tie together research and practice.

You know, so many of our approaches to crime in the past decade or so have not had a sound basis in empirical research. We're finding out more and more about what really does work in policing and in reducing and preventing crime. Frankly, we're also finding out a lot more about what does not work.

Office of Justice Programs has really been working hard to improve our basis of knowledge in evidence-based practices — what does work — so that practitioners in the field can use their resources in the best way possible. The Attorney General is fond of saying, "We need to be smart on crime, not just tough on crime."

So, the research that Dr. David Weisburd has been doing is a big step in that direction. Today, rather than the usual lecture format, we are going to handle this as kind of an interview of Dr. Weisburd and that will give us all a chance to pick our guest's brain, so to speak, about some of the issues that are on our minds about hot spot policing.

Our distinguished guest is the criminologist, David Weisburd. He's the director of The Center for Evidenced-Based Crime Policy at George Mason University. He also has a joint appointment as the Walter Meyer Professor of Law and Criminal Justice and director of the Institute of Criminology at Hebrew University in Jerusalem. He is a member of the National Research Council's Committee on Law and Justice, and founding editor of the Journal of Experimental Criminology.

Recently, we are so proud to announce, he received the Stockholm Prize in Criminology. That prize was awarded for his ground-breaking work in hot spots policing. And he's here today to talk about his work. We are delighted to have you.

Dr. David Weisburd: Thank you very much Mary Lou.

Leary: Congratulations on your award. We're all very proud, especially the folks at the National Institute of Justice who have worked closely with Dr. Weisburd. So you've published a lot on this area of place-based policing, sometimes we call it hot spots policing, and in your work you argue that place is a new prominent concern for crime prevention experts. What do you mean by that?

Weisburd: Well, one way to think about this is in terms of a crime equation. Traditionally both criminologists and crime prevention experts, police and others, they put people at the center of that equation and the questions we ask are: Why do people commit crime? What leads to people committing more serious crime? What leads to desistance or stopping the commission of crime? What can we do about those offenders to prevent them from committing crimes in the future, or to rehabilitate them if they're in prisons or other institutions in the criminal justice system?

So you might say both for criminologists, but also for the police, the question of "Whodunit?" has been at the center of the crime equation. There's also been another question that crime prevention practitioners and criminologists have asked, that's about large areas and crime. They've asked why certain regions or cities, more recently in the last 30, 40 or 50 years, people have been asking why communities have crime, why some communities have more crime than others. This is a sort of, let's call it a macro-place kind of idea, and there also is an important part of crime prevention, the way the police and others think about it. We organize our police by precincts, by police boroughs, even by police beats, and these are relatively large, what I would call macro-level geographic units.

Now my work suggests that something else ought to be at the center of the crime or crime prevention equation … and Ron has a slide. A place in this case is not these large or macro-areas of place, the communities of sometimes interested crime prevention practitioners and criminologists, certainly not people in this case; the center of the crime prevention equation becomes small geographic areas.

This is from a study, the Minneapolis Hot Spots Experiment that Larry Sherman and I developed, and in this study we said that an important unit for the police to focus on is the area of a street from intersection to intersection — a street segment, as you could call it, or a street block. And in this area lots of things go on. It makes it easy for the police to target, it's a natural place for them to look at, there are lots of things that happen socially among people, etc. that make it a very important unit, much different from the units, by the way, of communities or large areas.

Just one other example, we all know that sometimes crime doesn't restrict itself to a single block, and sometimes there are groups of blocks together that have a related crime problem that's focused there. This is from the displacement and diffusion project that was actually conducted by the Police Foundation. And in this project we looked at a prostitution site and a drug and violent crime site, and in this case the hot spot wasn't just one street segment, it was a group of street segments, represented here by the orange, that had that sort of problem.

So this idea even here is much, much, much smaller than the traditional large geographic units that have been at the center of the crime equation when people have talked about places. And we noted already, most of the time people are talking about "Whodunit?" in any event.

Leary: Well, you say that there are related micro-units, is that correct? How do you know that it's not just a bad neighborhood?

Weisburd: Let's start … can we go back a little bit, which is the question of why; we were talking about this before because it'll make it easier to understand. The question is why do we think that type of place is important?

Leary: Yes, why are they?

Weisburd: If you look at this … this is a study we did in Seattle, and what we did in this case is we took 30,000 street segments — those units like the Minneapolis Hot Spots Experiment — and we said OK let's look at these 30,000 street segments. How does crime distribute? If you look here you can see less than 5 percent of the places account for 50 percent of the crime in Seattle … so less than 5 percent of these street segments, they produce 50 percent or more of crime.

Now importantly as well, they do it each year, year after year, as you can see it's quite stable — 14 years. Well this suggests a law of concentration, and this is not the only study to show it, but I should note that it's the only longitudinal study to show it. Lots of studies now show that crime is not spread throughout an urban area; it's concentrated at these places, let's call them hot spots of crime. Now that of course suggests a kind of opportunity for crime prevention. If crime is not spread everywhere, if crime is concentrated in a relatively small number of places, then there's a tremendous opportunity, perhaps, for focusing the police or other crime prevention agents at those places.

Now this leads me to the question you asked — if I can go back now, it would be easier for me — about how do I know those places aren't all in the same neighborhood? Because that's the question, right? Maybe that previous idea of community, whatever, is actually right. Maybe there's 1 percent, there's a small number of places that have lots and lots of crime; maybe those places are all in one place in the city. I think it's a good question. The data suggests it's not exactly right, it's not right.

In another study we did in Seattle, we looked at juvenile crime hot spots, and I want to use this example because it's a smaller number of places to deal with. We looked at juvenile crime incidents. We said where are they committed? We found that 86 street segments in the city produced a third of all juvenile crime incidents. Imagine, a police chief can focus in on a relatively small number of juvenile hot spots and pay attention to a relatively large amount of crime. I should note, Chief Lanier and I were talking about this and trying to take this idea to see whether we can develop a program to do something about it. But that's the natural idea. We have an opportunity.

If you look at this map that Ron just put up, these are those 86 hot spots that I chose as an example because there's not a lot of them. The first thing you see in this map is that it all over the city. This is Seattle, there are "good" neighborhoods, there are "bad" neighborhoods. These hot spots sort of exist all throughout this map. It reminds me when I did the first study I did like this in Jersey City, New Jersey, we were mapping drug markets. And I sat down with the deputy chief and we were going over it and he pointed to the map and said, "I didn't know there was a drug market two blocks from my house." Now, many times I'm sure the police do know, but sometimes they may not. It's the spread throughout the city.

Another element to this map is, if you look in the downtown area, it looks like it's just a cluster right on top of each other. Part of that is the resolution of these visuals that you're looking at. The truth is when you look at the area, like the downtown area, where there's more of those places that are problematic, certainly, there's a larger concentration, if you like, in that community than in some other communities. But even there we found tremendous street-by-street variation.

In a study that we did on crime more generally, we found that if it was a high crime hot spot on a street segment, it was more likely that the street segment next to you would not be the highest crime group, than it was that they would be. So there is a tremendous street by street variability, and that, as just a final comment, I think when we talk about neighborhoods, we're stigmatizing whole areas. You can go into neighborhoods you think are not very good and lots of the streets are completely free of crime. At least in the definition of crime, it's wrong to talk about that whole neighborhood as a "bad" neighborhood. You're not only missing the fact that you need to concentrate on certain areas within that neighborhood; you're also sort of, in my view at least, stigmatizing the whole neighborhood on the basis of specific places. Focus in on those places; don't take the whole neighborhood.

[End of video clip]

Segment 2

Mary Lou Leary: So what does this mean in relationship to our traditional concept of community policing?

Dr. Weisburd: It seems to me that the community policing about what we do when we define the units we're interested in. The truth is that community policing was never really about neighborhoods, at least in my view. One of the first community policing projects was at the Vera Institute [of Justice] in 1985 with the New York City Police Department. And I was just finishing my Ph.D. and lucked out and was put on the evaluation of this study, a new program — community policing in New York City. And what I did was I walked the beats with the community police officers. Each of them were assigned a beat, which was a fairly large area — 20 square blocks or something like that, and not as large as precincts or larger areas. But what happened was when we would walk that beat, we inevitably spent all of our time at two or three blocks. In other words, even in that beat it wasn't the whole beat, there were two or three places. And in the end, when we're talking about and evaluating the program, I said to the people at Vera, I said, "You can't evaluate this program by taking data on the whole neighborhood. Don't interview randomly people who live in these 20 square blocks because most of the time they weren't there. If you want to look at this program you need to focus in on evaluation on those specific blocks where the police spent most of their time."

Leary: I see. When you do focus in on particular areas, what about the stability? Are the crime rates stable in those units that you're talking about?

Weisburd: This is, it seems to me, a major question if you're interested in policy and not just learning something about concentration. I said before, I think this is a law of concentration — study after study, our longitudinal studies, lots of cross-sectional studies, show that most crime is concentrated in a relatively small number of places or "hot spots." But that could be, like some other statistical distributions would say, there is a law of concentration, maybe 4 or 5 percent of the street segments in the city have 50 percent of the crime. But maybe each year that changes, maybe each year a different group of segments account for that distribution. Now it turns out as well, empirically, that's not what we found. Ron beat me to the punch, but here's a figure again from the study we did in Seattle. In this study we not only wanted to get a sense of the concentration, we also wanted to get a sense of what in developmental areas — when we talk about human behavior — it's called trajectories, or developmental patterns. So we adopted a technique from developmental studies of human behavior developed by Dan Nagin, and we try to answer the question: Are there specific types of patterns that would describe crime at street segments in Seattle over a 14 year period. And this is what we got out of it. There are 18 patterns here but I want to focus your attention on the top pattern in red. That's the really hot spots — that's about 1 percent of the street segments in Seattle. They account for I believe 15 to 20 percent of the crime. Those hot spots are the — this is the place I guess, if we're talking about opportunities, look at how much they differ from many of the other trajectories — this is where we want to focus our crime prevention.

Now what this says is they're pretty stable over time. I mean look at their trajectory — it doesn't change very much at all. And if there's anyone here that knows a little bit about developmental patterns of human offending, you know that that's not what you look at. This is not the figure you would see if we we're looking at trajectory patterns over a 14-year period for individuals, let's say from the ages of 18 to 32. What you see is they'd start — well, maybe 16 to 30 — they'd start relatively lower and then they go up and up and up in their early 20s and then they would start climbing and usually by 30 most of these people start aging out of crime. Now there are some persistent offenders, but that overall trajectory of chronic offenders that you see over and over again — that suggests, for example, if we concentrate on those offenders, we catch them late, we haven't done very much for society; we catch them at age 28 and put them in jail for 10 years. Well, if you wouldn't have put them in jail, soon enough they probably would have stopped what they were doing.

These crime place trajectories — that's not what we're seeing. It's a tremendous opportunity, because if you didn't do something it would just continue that way or something close to that. And I think that's another important opportunity of this approach.

Leary: Wow, would you say then that crime is more predictable by location than by offender?

Weisburd: There are two parts to that. I think overall that's the case and if you ask the question — and we try to deal with this question a bit— we ask the question, if you're a police department, if you're in Seattle and you want to focus in on targets, how many targets would I have to focus in on to reach 50 percent of the crime? So we did that analysis on individuals, and then we did that analysis on street segments, on places. What we found was you had to focus in on four times as many offenders to give them 50 percent of the crime, as you did of these street segments, or these places or hot spots of crime. Another thing that's important: those hot spots aren't moving around the city. Anybody who's a police chief here or works in investigating individual offenders knows that just trying to find them is impossible these days. We run a survey and try to capture people a year later; we find that you capture half of them or two thirds, sometimes, if you're lucky. But the point is you lose a lot of people along the way. So not only is there greater efficiency and focusing in on places at targets, places also stay in the same place — a rather important element of efficiency.

Leary: Do you think that there are other, kind of, particular opportunities that are presented for law enforcement with this approach, and at the same time, are there particular challenges of this hot spots place-based policing?

Weisburd: Well, early on in the 1990s, we recognized that these crime concentrations provide an opportunity for law enforcement. Lawrence Sherman and I did the Minneapolis Hot Spots study and it was published in 1995, and that study was very important at the time because until that study … most people of my age group will remember that in the 1970s and '80s, the view was: police, corrections — you couldn't do much about crime. In other words, there was a sense of "nothing worked," we just have to sort of incarcerate offenders, take them out of the world, etc. — keep them away, incapacitation.

In this study we really developed it because we became frustrated with this idea that the police couldn't be effective. We thought the police could be effective, we said the problem here is they could be effective if they focus resources a bit differently. And in that study we focused resources on 55 hot spots of crime that received extra police attention and 55 hot spots that didn't. It was a randomized experiment, whether you received this extra attention or not was just by chance. And afterwards we saw that crime declined — crime and disorder — about 20 percent I believe, but significantly during the period we're looking at in the treatment as opposed to the control group. And that led to a series of five other experiments — I think all of them, supported by the way, by the National Institute of Justice — that looked at whether hot spots policing could be effective. All of those five studies found a significant effect of the prevention strategy. Some used problem-oriented policing, or other strategies to increase effectiveness; some used regular patrol. So what you've got here then is, as the National Academy of Sciences concluded, this is the best evidence we have that police can be effective.

And you asked me about what are the problems, or as I would phrase it, what do we need to know? Well it sounds like five studies is a lot, and by the way it's a lot compared to other many crime prevention programs, but five studies does not answer all the questions you want to answer. What types of hot spots are appropriate? What types of strategies are appropriate for what types of hot spots? What types of hot spots are less amenable to this type of attention? What types of hot spots are more amenable? How can we be most effective and efficient? There are many, many questions — including by the way, a very difficult one which is, what is the effect on the communities — the people who live in these places — communities in the small world idea, what is the effect on them? Do they feel that they've been taken over by an occupying force? Do they feel reinforced by police presence? There have been very few studies of this and this is a really important question before going to carry out these sorts of strategies at the same time that we increase the legitimacy in the community.

Leary: Well this question of legitimacy for police is really at the top of the agenda these days. There's a lot of thought and a lot of talk about that. What, in your view, are the implications for legitimacy?

Weisburd: The studies we have so far, suggest that the implications are not negative, but I want to be careful because we don't have many studies. There was an early study on a hot spots gun approach that found that citizens were happy with what the police were doing, etc., at least non-offending citizens. I'm doing a study now in Redlands, California, and the results that we're getting are that increasing a broken windows type of intense policing, three hours a week in selected blocks as opposed to others — again a randomized experiment. We're finding it doesn't seem to affect either for the good or for the bad, feelings about legitimacy, etc. This is a place where I think more needs to be done. It seems to me that hot spots also provide an opportunity for legitimacy. It's a place you can go to, you can talk to citizens — police shouldn't be coming in like an occupying force on the citizens. They should work on these issues with the citizens — citizens should understand why this is happening why it's good for their block and in the long run it'll turn out to be something that will make their lives safer and better. To just come in, I think, and some of the studies that have been done show crime prevention benefits — but I think we have to think more about legitimacy and there's an area I think of importance that we should be concentrating on in the future.

[End of video clip]

Segment 3

Mary Lou Leary: Dr. Weisburd, whenever we talk about particular approaches to policing especially concentrating on a place, there's always fear that you're just pushing down and it will "splurt out" and, you know, the crime will be displaced to other places in the neighborhood. Is that true in your view? Is that what you see in these studies?

Dr. Weisburd: Well this is sort of the "boogeyman" place-based crime prevention in my view. And it goes back to a view that was very prominent in the 1970s that the police could not concentrate resources, because if they did the people would just move to the places where the resources weren't located. The idea somehow was that if the police "pushed down," crime would just of pop up elsewhere — a sort of balloon model. And not only that, criminals would know where they are, etc. and so there was a view at that time — we need to spread patrol unevenly. The Kansas City Preventive Patrol Experiment, other studies at the time, suggested that just wasn't working. Now the question becomes, what happens when you focus in on a crime hot spot and you achieve some effectiveness there? Well, what happens to crime? What happens in the areas around, etc.? As a police officer in Minneapolis said to me when I was riding in his car, in a very critical way he said, "Don't you know that crime is just going to move around the corner?" So this question of crime moving around the corner is very, very important. Now the empirical evidence suggests that's just not the case. A series of studies now show that when you focus in crime prevention at hot spots, you don't just move the crime to the immediate areas around… Let me tell you about the study. It's important because it not only shows that empirical evidence, it also provides some explanations for why this happens. The empirical evidence so far is that when you push down on hot spots you don't get an increase in crime nearby; in fact, you get a decrease. And Ron Clark and I have coined the term the "diffusion of crime control benefits." If you focus in on hot spots, you get a diffusion of crime control benefits around.

Let me tell you about this study because I think it's interesting, again it's an NIJ study, and I think a rather brave one, and also a brave one for the city that participated — Jersey City. Because what I proposed to them was: let's do a crime prevention program in which we're not interested whether the crime prevention is effective. We're only going to use approaches that everybody knows are effective, that have been proved in prior studies to be effective. Let's do that, and the real interest here — imagine telling a police chief this — the real interest here is I want to see whether crime moves to the other areas nearby.

Now fortunately there was a chief I worked with … a number of years, and he looked at me and said, "Of course David, we want to know if this is having a real effect. I don't want to just keep coming here and go next door the next week." So we designed a study in which the whole study was designed to create tremendous crime prevention activities at a target area — you saw it earlier, that orange area in the Jersey City study of a group of blocks together. And then we wanted to see what happened — did the crime move to the areas nearby? We chose those areas nearby so we didn't choose an area where there was a river so you'd have to cross the river to have displacement, we were very careful about how we chose it. Now this is just one slide from a whole series of slides that tells a story. The story is what the other empirical evidence shows. The blue line is the target area. As you can see there's lots of prostitution — this is based on systematic social observations, thousands of them I should note — in the target area the intervention comes, the line there on your screen, first line, and then things go down. With the other studies, it's not a surprise when you bring in 10 new cops, all kinds of new crime prevention, community, I mean everything went into this target area. And that went down and it stayed down throughout the intervention, but towards the end where it looks like it might be going up a little bit after the intervention period.

Now what's interesting though is the red line is the first catch of an area; that's one block out from that target area. And you can see, what you'd expect if the displacement hypothesis was correct, would be that crime would go up in this thing — it didn't do that. It followed the same pattern almost exactly. And if you look at the green line as well, although there were less prostitution activities as well, that's two blocks out — quite a large way, two blocks out from the target area like the one we looked at earlier. Now in this study we were fortunate in that we had both qualitative data collection in terms of interviews with the offenders and also an ethnographer that worked completely independently of the rest of the project. Whereas I hired someone, I said, "Go out there do what you do. Talk to me in a year and half." The only communication we had was if there was going to be a major police crackdown, we didn't want the ethnographers being picked up, so we warned them a bit. It was a safety issue that we had to follow.

Now just to start out with a little bit of theory if you don't mind. In this kind of idea, if crime is concentrated in certain places, the next question is, why? The reason why is there are special opportunities at those places. In our juvenile crime study for example we found that those 86 places, where were they? You know where they were. They were malls, they were restaurants, they were not bars — this was underage people. They were not bars. In other crimes studies, bars show up prominently, because those are the activity spaces that juveniles hang out at. If you want to rob, or so to speak, they are going to look for good opportunities for that, maybe around a subway station or a bus stop.

One of the fallacies …there are two fallacies about the previous way of thinking about crime in my view. One, I think, is the Dracula model — offenders are like after blood and they have to it — True Blood on television or whatever — and they're going after them. Nothing you can do about it; they're totally involved. A lot of people that commit crimes are not that involved. Most people that are arrested once are never arrested again. Much of the crime problem is committed by people who aren't that committed. And even people committed still have to make food every day, Sometimes they have children they have to take the daycare. There's a study in Israel that shows that women offenders take their kids to daycare then they go out and offend, so lots are going on for these people. They're not as motivated as you want.

The second is that crime opportunities exist everywhere; crime could just move anywhere. A prostitution market is not going to move onto a street with 500 door people with 10,000 people living on it, right next to the police precinct, OK? That's not going to be a good place. They're going to look for certain sorts of opportunities. Quite often prostitution markets, for example, are in places where there aren't many people who can call the police. They're in places where there is an abandoned lot or an abandoned building where you can carry out your activity. The idea that these crime opportunities are unlimited is wrong; there may be more than a few, but unlimited is certainly the wrong way to view it. A drug market needs a certain type of place where people can get in and out. Quite often there are places that are safe to get to because people from the suburbs coming in don't want to get killed every time they happen to buy drugs. So there are all kinds of components that make a specific place a good place to carry out a certain kind of criminal activity.

Now in this study we asked the question, "So why is it that these offenders did not just displace? What's going on here? What happened?" OK, so one of the things we did is we asked people through these interviews, like, "Why? What's going on? Why aren't you moving? Etc." We tried not to ask leading questions and we got answers like this; one of the answers was "Well look, if I have to move, I have to re-establish my knowledge of the area, I have to re-establish my clients, there's a lot of work to be done." This sounds like regular work and I think, by the way, that one of the mistakes we make are thinking of offenders as very different from ourselves. Offenders make decisions often just like we do, maybe not with the same level of rationality or knowledge, etc. but they are making decisions in many ways the way all of us make decisions.

So one problem is, if I move I have to re-establish things, I have to tell people — customers, others. The other thing is if I move and I go to an area, I don't know it. I don't know who's going to call the police and who won't call the police — what a tremendously important idea, right? If you know who's not going to call the police you just carry on the activity while they're there. The opposite of this, by the way, is in the 1980s when the New York City Police Department was moving cops from precinct to precinct to keep them from getting corrupt. They forgot that the cops went into an area that they didn't know, so they'd go and beat someone, stop someone, etc., who's been there, an old guy who just acts a little unusual, because they just didn't know the area.

In the same way offenders, familiarity of the area — knowledge — is very important for issues of safety, issues for carrying out the activities they want to carry out.

There's also the question of danger. If you're in the market especially, and if you go to another market they may not want you there. One person said, "Why don't you just move over four blocks away? There's another drug market, plenty of customers." He said, "If I go over there they're going to kill me." I mean you just can't move; it's not that simple. There are other sorts of dangers besides the police calling.

Now finally in a very sort of human way one of the prostitutes when asked by an interviewer, "Why didn't you move to another spot?" did not mention those issues, she simply said, "I don't feel comfortable there. Those are not my type of girls." Now that really says to me the nature of which offenders are making these sorts of decisions. We think it's easy to move where you offend. Is it easy to move where you live? Familiarity, a place, is a very important component of our lives. It's not easy to change.

[End of video clip]

Segment 4

Mary Lou Leary: So these kinds of things really do present opportunities, I guess, for law enforcement, all of these, you know we have at least one…

Dr. Weisburd: Mary Lou, let me just comment on that because one of the questions you might ask, too, is what happens to these offenders?

Leary: Yeah, and what happens to the ones that don't get arrested?

Weisburd: Exactly. One reason for diffusion of crime control benefits could be if you arrest a lot of people, they may be the people offending not only in that target area but also in areas surrounding and that's probably one reason. But we found in our interviews some interesting things going on. One was, in the prostitution site, remember, we used what we thought was knowledge of crime prevention that had been developed over the last 10 years. The NIJ had produced a report, What Works, What Doesn't, What's Promising, I believe was the title. And we built on that and also some experts, and we also included community policing kinds of approaches. So when the cops went to the prostitution site they also identified organizations nearby that worked with prostitutes, tried to help them get out of prostitution, etc. It turned out that it had some traction to it. About a third of the prostitutes - according to the ethnographer, not according to our data - a third of the prostitutes desisted and started getting involved in these sorts of groups. Now whether they desist in the long term I don't know, but it suggests also the salience - you might actually get a benefit.

Now another element if you think about prostitution, not about from our study, is again we tend to think of people as absolutely committed to doing this, and prostitutes or drug deals would seem to be pretty involved. A lot of offenders are not as committed to crime as we think.

In a study, in a housing project in the UK, they cracked down at a specific place, at a prostitution market, and they found out a lot of the women did not continue offending. And they interviewed them and they found out some were housewives or people that had other jobs that were doing this to make extra money. Now I know that might sound strange but that was in fact what was happening. So there are, on the one hand, models or opportunities for desistance, for actually stopping these people from getting involved in crime. I think it's also very important to note that a lot of people get involved in crime in a serendipitous way. If we take away that opportunity, they may not get involved. We know this with our children sometimes, we know this with ourselves - removal of opportunities can sometimes stop us. A lot of crimes are committed without much pre-thought: on the way to something, with a group of guys, with a group of gals, etc. And if you can block the opportunities that make that crime easy or make it attractive then you also can also have a long-term impact.

Now there was one type of displacement I should note in our study, and it was an area displacement that we noticed as much, it was that sometimes the prostitutes or the drug dealers in this study would change the way they worked. In other words, they started working with beepers. The prostitutes started making dates. Now in this sense the strategy of the police did not end the crime, but they removed them from the immediate area. And at least the police chief in Jersey City said to me, "You know, if I can get people off the streets, if people aren't being bothered, etc., well, that's a benefit for me." And I look at that as still a crime prevention benefit.

Leary: Interesting. So you're a police chief in these economic times, they slashed your budget, they cut your staff, and you want to use your resources efficiently. But how do you go about doing that under the current circumstances?

Weisburd: Look, I think efficiency is a critical issue here. Now the truth is that my own view would be, one should be careful about reducing the strength of urban police too much, because there's a point at which you can't do anything beyond answering 9-1-1 calls. A lot of what the police do is that they simply respond to citizens who call them after the fact. Now if you want your police to do something about crime prevention, there has to be resources for that. Now those resources may be more limited in a period of economic difficulty, but if that's the case, why wouldn't you draw on the efficiency of hot spots policing?

I mentioned before that you need about a quarter as many targets to get at about the same amount of crime. Crime is stable in those areas, meaning that you're not wasting resources on a place that just would have gotten better anywhere - what statisticians call regression to the mean. It's an approach that's very easy for the police to carry out in some sense because it doesn't involve tracking people across the world for the most part, though that might be a part of it in some sense. The places themselves are right there and you can go after them. So it seems to me that choosing a few hundred places in the city that are particularly crime prone is a lot more efficient strategy than trying to cover the whole entire city in some sort of blanket, which is what many police departments try to do - the idea of omnipresence. Omnipresence is a gigantic mistake. The police do not have enough resources to be omnipresent. If you did, if you could put a police officer in every single street segment in the city, well that might be good. You'd be wasting a lot of money in my view. But nonetheless it's not realistic in the present circumstance. And therefore my view, at least, hot spots policing provides a tremendous opportunity for the police to focus in resources, to save money if you like, to be efficient about policing, rather than to spread those resources across the urban landscape.

Leary: So could you just tell us a little bit more about how law enforcement needs to change in order to implement hot spots policing?

Weisburd: Well the key to this all is what I said at the very outset about the crime equation, or for police, the crime prevention or control equation. The police need to change the center of that equation, in my view. They need to focus their interests and efforts much more on places. Well how do you do that? What would they need to do?

Well here are just some ideas. One is, that police know an awful lot about the perps. They know "whodunit", what they look like, where they came from, sometimes their education. Some databases and police departments have all sorts of socio-demographics on these people - lots of information; the modus operandi, etc. But we know very little about the places where crime is occurring. There are places in the city with hundreds of calls, some places even perhaps with 1,000 - I don't know if that occurs in DC. Very large numbers of calls to the police. Is there a database readily accessible to police like the database on the individual offending that tell them: here is the history of crime at this place, here's the kind of thing that is going on, this is where it's happening, it's in the bar, it's in the building, there are lots of kids in the street.

In a study that we have now, it turns out that if you have juvenile delinquents on the street - I should put that differently - if you have kids that are truant in school and have all sorts of behavior problems, you're going to have more traditional crime on that street than on other streets. So you should know about the characteristics of this street. Now the truth is we don't. The only reason we know where crime occurs is because the police had to send cars somewhere and they needed the addresses. Indeed, until 10 years ago those addresses were always, when I tried to geo-code, to put in a geographic information system, data in Minneapolis in the early 1990s, we found that we could only use the program for about 20 percent of the addresses. We had to recode everything. Now most police departments have improved on that - on that they have improved - but they still haven't collected that other data.

Now the other thing that the police need to do is they need to change their focus a bit in terms of what crime prevention is about. They need to start thinking about "wheredunit" rather than "whodunit." They need to start thinking about the place; that says a group of different things. If I'm concentrated on the person, then what I'm usually thinking about is identifying that person, arresting them, bringing them to justice. Now I'm not saying that's not an important part of police work, but when you focus on the place, a new set of questions begins to emerge. What can I do to decrease the opportunities for crime at that place? Should I increase surveillance of the police? Should I work with the community so they'll increase surveillance? Should I change certain opportunities for crime on that street? Should I identify the times of day and maybe increase police presence? Should I develop problem oriented policing at that place? It changes the framework. The question becomes, how can I make that place safer? How can I make the people at that place safer?

Now there's an added benefit if we take this approach, I want to note. The approach person-centered prevention has a downside to it that we all know about today which is, in the end we put a lot of people in prison, more than any other place in the world, more than we ever put before, many times more. We have a gigantic prison population that costs us a fortune - $50,000 to $100,000 a year per cell. It is a gigantic social problem in a sense, the fact that we have to do something with all of these people.

Now I think place-based prevention offers something different. We're still going to have prisons and jails, that's not going to go away, and there are going to be a lot of people in those places. But if we change that focus to doing something about the place, then maybe we can prevent some crime, maybe in the case of juveniles we can stop some of those juveniles from getting involved in the criminal justice system. So in that case, the evidence is we can get more crime prevention by focusing on places. Well if that's true, we can get more crime prevention, and we get less people in the criminal justice system, we've got more safety, and we have less cost from prisons. And it seems to me that's an experiment we ought to think about.

Leary: So you think we could actually decrease incarceration through this approach?

Weisburd: I think so, let me put a caution on that. Throughout the United States today police departments know about hot spots. Many police departments throughout the U.S., I think, in the police foundation survey a few years ago, something like half or more of the larger police agencies were doing hot spot analysis. They want to identify where the hot spots are. Now the problem is if you just throw police to those hot spots and you just arrest people, then the result will probably be you'll arrest a lot of people. More than the traditional approaches? I don't know. But that suggests that when we think about hot spot policing, it has to be more than just arresting, more than just enforcement, and why not? Because there's something about opportunities at those places. Do something about it. In the case of juvenile crime, wouldn't it be nice - there's a whole series of studies on juveniles that suggests that they get into trouble when they're unsupervised - unsupervised socializing. So here we find we have all this crime committed at restaurants, malls, activity spaces for kids. Well what would happen if we sent out people, not police officers even, to put some more supervision in those places? Maybe then we could have a crime prevention effect without arresting those kids and bringing them into the criminal justice system. So you can just increase arrests if you focus in on hot spots. I don't know if you can increase it beyond where we are now, but certainly there is an opportunity to arrest less, if you like, and get more public safety.

Leary: That's fascinating.

[End of video clip]

Segment 5

Mary Lou Leary: I'm going to let folks in the audience ask some of their questions. We'll open it up and if you could please come up to the microphone and identify yourself and ask your question.

Dr. Weisburd: While you're thinking, I was going to draw attention to this New Yorker cartoon that I love. I think these guys have it right. It says, "Location. Location. Location."

Leary: [Laughs] First rule in real estate and policing.

Weisburd: Soon, maybe in crime prevention.

Robin Demon: Hi my name is Robin Demon. I'm with the Office of Legal Policy here with Justice, and I was interested because you were mentioning the people-based policing programs and how often we will arrest someone who might be in their early 30s who is ultimately going to age out of crime in any event, so what have we really done when we're arresting and prosecuting folks that might be close to aging out of crime. I'm wondering how that theory intersects with the location theory. So if we do go in and we're looking at taking enforcement action, should we be taking enforcement action in a particular place against a particular group of people who are maybe on the younger end as opposed to on the older end? So how do those two theories intersect?

Weisburd: Well, that's a little scary I guess.

[Laughter.]

I guess I think that that's not how we start. The way we should start is we've got a place, and that place had 100 crime calls in a year. Where is that occurring? What's generating those crime calls? Has this been going on for 20 years? Do we just keep going in and doing something on a continuous? What can we do about it? If its juveniles and it's a mall, can we increase police presence or get more security guards in the mall to do something about it? If a robbery is occurring around certain times of day, can I put something out there; can I put a car out there around the time that these things tend to be happening?

So you start off with a different set of questions. Sometimes you need to arrest people and sometimes that's a good strategy, but the problem is that it's a very costly strategy and sometimes it won't achieve what you want. I think his name was Goldstone, or Goldstein from New York State, was a prosecutor. If anybody remembers him he was a very innovative guy. Did I get that right - his name? Goldstone, I think. Anyway, so he had this idea that prosecutors ought to stop just trying to arrest offenders and think more about problem-solving. And he gave me the example of the Fulton Fish Market which he said, everybody thinks it's like a mafia thing. You look at it before the mafia there was corruption and crime at the Fulton Fish Market. It just went on forever. He said every time you arrest a group of people, a couple of weeks later, a couple of months later, a new group comes in. He said, "That doesn't work. I've got to change the way this place operates." That's going to be a long-term solution. So even in a way, in an organized crime kind of way, there can be very important elements of opportunity at place, or in the systems of place, that will provide an opportunity for crime prevention.

The idea somehow that we're going to create public safety without arrests or we're not going to have a criminal justice system with people in prison is not realistic, and I'm not sure it's a good thing. Remember that one of the things we try to do in the criminal justice system is express our norms, our values, and sometimes you want to express a value that someone did something really bad and that person needs to be punished. In the criminal law, that's the punishment that we view as most serious. So certainly that's part of what we should be doing. But you know, in the rest of the world, America is looked at quite negatively these days for the very large prison population. It can't be that we need to put all of those people in prison. Now, maybe it's necessary, I don't know. I think this approach may focus us a bit. We may get just as much crime prevention - I think more and we'll have a little bit less. Because if you focus in on "whodunit" and it becomes getting him and prosecuting, then that's what you're going to get in the long run. This provides an opportunity, I think, for something else.

Leary: Hmm, interesting. Next?

Phelan Wyrick: Hi I'm Phelan Wyrick with the Office of Justice Programs. My question is, how well equipped are law enforcement and policing agencies to take this approach? Is it really just a change in emphasis or are there new technologies, new types of analysis required?

Weisburd: Well on one level the technology has been developing irrespective of the police - GIS and other analysis systems for location of crime, etc. And many police departments, most larger police departments, have those technologies there. There's lots more that needs to be developed. Just to give you an example of something now, the Police Foundation are doing a study of ... automobile vehicle locator systems or AVL ... or automated vehicle locator systems. And in Dallas, Texas, it turns out, every police car has a GPS so you know where they are all the time. Well, up to now it's only been a safety issue. What they are trying to say is, where are the police? - it's a study I'm involved with as well. - and, where is crime? We have found, so far, is that there is a law of concentration per police patrol just like there is for crime. About 16 percent of the patrol areas - they're larger than my hot spots, but not that much larger - account for about 50 percent of police patrol - so there's that law of concentration again. But the funny thing is - well not funny - but the problematic element is it doesn't always fit with high crime and the high patrol areas right on top of each other. So it raises some questions about using. But Dallas is the first place I know, I think there might be a few other places that had begun to think about using this technology where the police are to do something about crime.

There is just one other addition. I don't think it's a technology issue for the most part. I think the technology may be even ahead of us. I think it's more of a perspective, a mentality. It's the idea of moving from "whodunit" to "wheredunit." Data systems need to be developed by the police for this part but they also need to think about the reallocation of the way they work. Police operate now in large geographic areas. There are police precincts or boroughs, and that's how they allocate their patrol and officers. This work might suggest, well, wait a minute, should I be having special patrol groups for hot spots? Is that what I should be doing? Are there other ways to organize the police? The way we organize the police, this happens every time. If you look at precincts or police reporting areas in Dallas or other places, they were created ages ago quite often and they have this life to them that may not have much to do with how we can develop crime prevention. So I think it's more of a reorganization of the way police work and think than it is in the technologies.

Leary: Hmm, other questions? Yes, sir.

Third Audience Member: Hi, I'm Samir from the Office of Civil Rights and I would like to know - since people of certain races tend to congregate in certain areas and the focus of the strategy is "location," how would you counter the claim that this is just an organized form of racial profiling?

Weisburd: Well, first of all, our data suggests that there are also hot spots in areas that are not only of a certain ethnic group, etc. or a socioeconomic group. So the hot spots are not only - don't have in your mind that every one of these hot spots is all minority, all poor, etc. That's not what it is. Those factors may be correlated with where these hot spots are. But first of all you should know that you have that right now anyway, but in a much less effective way. Police chiefs I speak to, and here's the irony, think of your question and I'll tell you the way a police chief thinks about this. Police chiefs I know, really good police chiefs say the following. I say, "Chief, where do you want to put the resources of your police department?" He says, "I don't want to put it in the rich neighborhoods. They don't need me. They have private police. I want to give the resources to the people who really need me." The poor and disadvantaged, often minorities, maybe.

So what are the police doing? They're focusing in their resources to poor neighborhoods because, in this case, it's a very forward-thinking police chief who wants to provide service, because he looks at what he does as protecting a public and providing a service. Now the problem with that is, what happens if arrest is your major idea? You throw all of those police in a neighborhood with high minorities and all of a sudden we've got a high-minority arrest rate irrespective of whether minorities are committing crime at that much higher rate in the city. So it's a problem, this geographic issue more generally. But you see my view is, if you change to a place-based focus, it should be less of a problem. You shouldn't be going there and arresting everybody. You should be going there and saying, "There are a lot of problems on this block. We'd like to help you. These are the strategies we're thinking of using," or involving the community a bit to make it safer.

There is a kind of difficult problem in urban areas. On the one hand, everybody wants more police. The police in America are very popular, more popular than police in most other places. Americans love their police and everybody wants more than - or most people, not teenagers for example, they probably don't want more of the police, but people want more. But the police coming to an area, especially a person-centered crime prevention, means that in the end it's just going to mean more arrests of the people in that area, inevitably leading to higher arrest rates for those sorts of people. That's not what I'm suggesting. I'm suggesting a place-focused approach that hopefully won't do that and will lead to greater public safety. And by the way you have to be thinking about these issues. Police chiefs today have to think about what it means when they put large numbers of police in a particular minority area, in terms of the young kids there, who are now going to be arrested for things that in other neighborhoods they won't be arrested for because there aren't enough police around to arrest people for those issues.

Leary: Thanks.

[End of video clip]

Segment 6

Mary Lou Leary: Another question?

Brecht Donoghue: Hello my name is Brecht Donoghue. I'm from the Office of Justice Programs. My question was - considering this place-based approach, do you find that it's more effective or only effective with crimes that are sort of place-based? And what I mean by that is, I'm thinking drug markets, I'm thinking prostitution, which seems to be sort of a place-based crime versus perhaps like property crimes. I'm wondering - or muggings, you know those kinds of things. Is it really going to focus on specific types of crimes or do you find that it sort of bleeds into other areas?

Dr. Weisburd: Well, I want to answer the question, but another question for the person in the back. I think one of the things we really need to know is - what is the demographic makeup of these hot spots? It's a very hard question to answer. We have some data in Seattle. I'm actually going to try to look at it. You can't get census data at this low in aggregation, but it's a very good question, I'm going to think about it, so maybe we should be in touch. It's an interesting kind of issue you've raised.

But anyway, this issue of what about types of crime. You know, I chose the prostitution hot spot and the drugs and violence hot spot, because I wanted to pick a type of crime that everybody thought there would be displacement. In other words, I chose those types of crimes because I thought everybody would tell me - what are you kidding me, prostitutes? They're just going to walk around the corner, right? So I purposely in that study didn't choose types of crime that people actually think less that they just move around easily. You mentioned some ideas. If you take things like robberies actually, they're quite placed-focused. Robberies tend to congregate around bus stops, subway stations, in areas that are a little bit dark. In other words, what do you need for a robbery to occur? You need someone that's worthwhile to rob. You need an offender who can get there. You need a place that offers some opportunities, maybe a dark spot or a place that can't be seen. Those three things converge and they have a robbery. So robberies are not spread throughout the city.

The other example you gave, one was robbery and one property crime. You know one of the most important findings, separate from this, was a finding in the UK by a group of researchers in which they found - you know cops used to come to people's houses and say, "You've been robbed, or burglarized, don't worry, you won't be again." You know what? It turns out, if you've been burglarized once, you're much, much more likely to be burglarized again. So, place is important. We find that also with burglary offenders tend to work in areas. In other words, not just one house gets burglarized, a whole series of them, they tend to like - it's like fishing. You know they fish in a certain area, and then they've cleaned out the area, and this is interesting. In this case they tend to move on to another area to fish in, but interestingly enough, they tend to come back when people have managed to buy some stuff. So they can come back again because the opportunities that made a good place still make it a good place because sometimes the citizens don't increase crime prevention, etc.

Now, one should be careful about this. Does this mean every crime is geographically focused? Does this mean every type of prevention should be based on place as I've said? No. Those are things we need to learn more about. In some cities, car thefts are highly focused. Now one thing that is highly focused is where you find the cars, usually near someplace where they're selling parts or something like that. But in Jersey City once I did an analysis, I found that he city was terribly urban, you know, intense. And it was pretty spread out; it wasn't as focused. So I'm not saying there aren't crimes that are focused, but crime overall is incredibly focused by place. And just because something is obvious does not mean it's not an important insight. Sigmund Freud wrote a book, The Interpretation of Dreams. Its 600 pages or something. It says that dreams have something to do with the way you feel. That would not be a new idea to the people in this audience. In the 19th century, it was a pretty new idea. I think it's like displacement. The fact that you don't get displacement, you get diffusion of crime control benefits, it's a startling idea but it makes sense once you start thinking about it.

Crime is attached to places because places are part of that equation, and I think a very central part of that equation. So it's not surprising. Burglaries are businesses. Do they occur in a residential area? No. So when you start thinking about it, without any empirical work whatsoever, you can start isolating what you're developing, what you're knowing. So what I would say to you is we need to learn much more about this and if I can just give one - I don't know who is in this audience, etc. But you know I've written a paper for the the Harvard Executive Session on Science and Policing, and one of the things I point out is the National Institute of Justice had $7 million last year for all social science research - policing and everything else. The National Institutes of Health has over $1 billion a year. The Dental Institute has a few hundred millions a year. The National Institute of Education has $167 million a year for cutting edge research projects. So what I say here is you are getting an amazing amount of information on a very small amount of investment. So I suspect this is preaching to the choir, but if someone is here that's not from the choir, then I'd say that it's really time to invest in crime prevention because there's a lot we can know and a lot we can know that will help us be more effective.

I should note that tomorrow is a Congressional briefing that our center is doing, the Center for Evidence-Based Crime Policy - I hope you don't mind. Cynthia, where is it?

Cynthia: [Unintelligible.]

Weisburd: So at the Capital at 9:00 in the morning? 10:30 in the morning at the visitor's center. There are these panels we run on evidence-based crime prevention where people are talking from different perspectives about what we learn. And what I've got to say is it's amazing how much science has been produced on so little research support.

Leary: It would be a great thing to hear those presentations and to hear what we've learned about what actually works in crime prevention and control. It's really quite amazing. This was fascinating, fascinating today. Are there any other questions?

Fifth Audience Member: David, first let me compliment you on your insights in looking at these issues. I'm wondering whether or not the existing model that we have for providing the police services will be able to focus efficiently on place-based policing to have a significant impact. We want to prevent crime, and if it's robbery we want to make sure there is enough light in the street. But the police have no power over the streetlights - that's a governmental function at the municipal level, but it's not anything the police have power over. And I wonder whether or not if there's more of an interdisciplinary approach needed in terms of the organic model that the police will use in the future to leverage place-based policing to have a significant impact.

Weisburd: I think you're right on target. There are two things that need to be done. Let me say at the outset, the one thing that needs to be done is that police chiefs and others need to start thinking about this. How could you reorganize the police? This could be a type of problem that should not be handled purely by researchers but really with the intersection of researchers and practitioners. So I'd say at the Police Foundation you should start thinking about these issues and bringing the police in to start thinking organizationally. But in some sense the police have been struggling with this issue at least since the early 1980s. When Herman Goldstein wrote his paper on problem-oriented policing, at that time he thought the police needed to focus in. He wasn't only talking about places, but he thought that they needed to focus in on problems. And that article, if you read it, is really a cookbook, I always say. What Herman did was to go into how the police could do it and he talked about the police partnering with other agencies, like housing or liquor law enforcement or whatever the other agencies are, to do something about crime prevention. The police have been struggling for years - third-party policing, involving prosecutors especially - using civil ordinances to do something about crime. These partnerships, these ideas have been developing for years. So I don't think it's a really big step. I think it's really just using some of the ideas we've had so far and developing them. Place puts a slightly different complexion to it but in some sense many of these ideas have been developing anyway. The third-party policing idea, for example, I believe developed in Oakland, California with a program called Beat Health, developed by a sergeant in the Oakland Police Department. And what they did was pick out houses that had people using drugs and selling drugs at the house and the Beat Health was making those places healthy. And they used the courts, they worked with the prosecutors, to use the courts to say to the landlords that if you don't change this, this is going to be closed out. The city is not going to allow you to rent this to people, etc. So we've been thinking about this for a bit. I think we need to think about it more.

Leary: Yes, indeed.

[End of video clip]

Segment 7

Mary Lou Leary: Any other questions? Robin.

Robin Demon: Pardon me for monopolizing your time but it's just so wonderful to have you here and to have this opportunity to have a conversation. I was wondering if you could just speak a little bit more about the displacement that you didn't see in your study? You had mentioned that you had talked to the police chief and he said, "Hey you know the prostitution moved off the street and that's a success and that's something that we should feel good about." And I think this is sort of an intersection with the question that was asked earlier. That might be OK, but I don't know if that would have necessarily been his response if it was robbery that had moved off the street instead and was now happening on the inside of a corridor in an apartment building. So I'm wondering if you could talk a little bit more about that displacement and what, if anything, you think this place-based approach might be able to do to even address that kind of displacement.

Dr. Weisburd: Well look, there's only been one direct study of displacement, the one you just saw. So my first view would be that we ought to be doing more of these to learn more about this. If place-based crime prevention becomes very central, we need to focus in on the issue of displacement. We need to understand what's going on. The knowledge we have now says immediate spatial displacement — not a really serious problem. And I think the knowledge is good. There have been a series of randomized experiments, the study I showed you before, or talked about before. So answering the question, is this going to move around the corner? I'm not saying the police can't move it around the corner if they don't want to. They certainly can; police are good at that. The idea of moving crime to make it less troublesome is something the police have thought about that for a long time. But overall crime prevention that's focused in on hot spots, immediate spatial displacement, unless you purposely develop an approach that's going to start to lead to that, is not a major issue. Now what's important about that is that most people, including the cop in Minneapolis 20 years ago that spoke to me, or 15 years ago, that was the assumption — they're just going to move, that's the easiest thing to do. The first thing we need to say is moving, it's not that easy. By the way, in some of the studies we've also checked whether there were places that grew up in other parts of the city. It's not that a drug market won't eventually grow up in other parts of the city, but there's not a lot of evidence that it just easily moves along, you get the same amount of crime etc., immediately around the areas, as I noted before, diffusion of crime control benefits.

Now the question becomes, what about the other issues? Do you get method displacement? I talked about method displacement before — the prostitutes were making dates. Is that a problem? Now part of that is chasing them around. The method displacement in this case, the police chief said, "I don't care about that. I don't care." In a way that police chief made any normative legal decision. Prostitution may be illegal but I'm not going to really go after it unless it hangs out in my streets. Now that's a particular decision. In certain types of crimes, you're right, we're not going to be very happy if burglaries were initially using certain types of, I don't know, ways of getting in, and people put in alarms because the police asked them, and now they're using something else. I think we need to learn much more about method displacement. There may also be crime displacement. Will offenders of one type displace to another type? Right? Is there a type of specialization of crime? It's a very big issue in criminology. There doesn't seem to be that much specialization, but quite often that's because we're looking at sort of groups of crime. A knife, gun, etc., I mean these are all related forms of specialization.

One of the things I think we learned from our study is that it wasn't so easy to change your job, so to speak. Some of the prostitutes started trying to sell drugs — it didn't work. As one prostitute said, he said, "I started selling drugs and then I started using them and then I didn't sell enough and the dealers were trying to kill me." It does seem to take some time to change your job, so to speak, for offenders as for other people. But let me just say, to some degree, some of it's not speculation but a lot of it is speculation. We just don't know enough. We need, I don't mean to harp on this too much, but you can't develop, you can't solve, or reduce a public health problem, a public safety problem with $7 million a year or something. $1.5 billion for medicine I think it is — it's very large — lots of money during this stimulus and they're still struggling to solve many public health problems including H1N1.

Policing is a gigantic problem. The police invest an enormous amount of money. Take all of that money together—it's multiple, multiple billions of dollars a year; we invest in that. $45 billion a year, I don't remember exactly what it was, but it's a very large sum, and $7 million for research, and of that $7 million most of it is not for policing anyway. That is a very sad story, and I think if we really want public safety we're going to have to do something about it. And by the way, we have at least one ranking police officer in our audience, but police have to take some responsibility for this like public health professionals do. The police have to advocate for research. There has to be a sense that it's important to what you do, and can provide you with benefits. And I think this is a very key issue because we can't, as I said before, it's amazing how many questions we can answer, but it's also very, very critical. But we can't answer many, many key questions because we just don't have the support to do it.

Leary: Any other questions from the audience? David, I wanted to follow up on one thing. We just talked quite a bit about displacement. Could you tell us a little bit more about what we know about the displacement of the positive effects of this place-based policing?

Weisburd: I guess the way to think about that is with this idea of diffusion because one of the questions is, "I understand why crime didn't go up in those areas, but why did it go down?" I think there are group of explanations for that type of issue. One is that the, simply put, when we do take off certain really problematic people, they are also responsible for crime in the area surrounding. So there are certain people that need to be removed. They need to be put in places where they can't harm the public, that's for sure. And I think that accounts for some of the diffusion. Some of the diffusion ... you gain by the fact that some people are desisting not only in the target area but also in the areas that we looked at around. But there's something else as well, you could call it a form of deterrence that occurs. Remember that offenders — when a police department carries out strategies in hot spots it doesn't go to inform all of the offenders in the community because that's exactly where it's working. It doesn't say, "Hey guys, I'm going to be at Ninth Street between F and J. No other place is getting this intervention."

So the truth is from an offender perspective we found, when the cops start doing their job aggressively and trying to solve the problem, they get nervous all over the place. There is a residual deterrence if you like. "My god, the cops are out like crazy, we better hide underground." You know they don't say, "Let's go three blocks away; they are not going to go there." Now first of all they know it would be irrational if the police didn't follow them at some point, but nonetheless you get a certain amount of deterrence, and that's one of the elements. The final element may be that when you solve crime problems you empower the community. In other words, a community ... quite often you have a spot with lots of problems and the people around that area are frightened. They don't want to go they don't want to walk there. They're troubled. There's not much they can do. Now you take away that problem, and the problem they're dealing with is much more manageable. All of a sudden you can hang out in the street. All of a sudden you do call the cops because you can see that they're doing something. So one of the elements may also be an element of empowerment. A lot of this, by the way, is speculation and we need more research to learn about it.

Leary: Wow, OK. Any other questions? Well, I hope you agree with me that we are taking away so much to think about from this discussion today. It really has very important implications for what we do in policing and in crime prevention, especially if we want to do it in an evidence-based way. So thank you so much, David, and we've learned a lot and look forward to continuing the conversation.

Weisburd: Thank you, Mary Lou.

[Applause.]

[End of video clip]