A Solution to America’s Gun Problem

At the center I run at the University of Chicago, there’s a standing joke. Whenever we’re discussing a new idea related to criminal justice, I either say, “I like that idea! Let’s bounce it off of Jens.” Or alternatively I say, “I don’t really like that idea, but let’s bounce it off of Jens.” Who is Jens? Jens Ludwig is my guest today, an economist and the director of the University of Chicago Crime Lab. In my opinion, he is the leading thinker on crime today. And he believes that both the left and the right are getting it wrong when it comes to fighting crime.

LUDWIG: The data winds up suggesting that most shootings in America are not premeditated. Most shootings in America are arguments that go sideways and end in a tragedy because someone’s got a gun.

Welcome to People I (Mostly) Admire, with Steve Levitt.

*      *      *

Before we get into solutions to the American crime problem, I think it makes sense to talk about some of the facts, trends, and causes of crime. We’ve experienced incredible declines in crime over the last 25 years, but the homicide rate in the United States is still three to four times higher than a typical wealthy nation. I started a conversation by asking Jens whether he thinks the huge numbers of guns in the U.S. is largely responsible for that difference.

LUDWIG: Yeah, I totally do. The thing that really makes the United States exceptional compared to other rich countries around the world is lots more murders with guns here. In America, the crime problem basically is a gun violence problem. and that’s the thing that is most worrisome for the public. I can tell you from first-hand experience — I’ve lived in cities for 30 years. I’ve had my car broken into, my bike stolen. A few years ago, I was held up at gunpoint picking my daughter up from her piano lesson here in Hyde Park, and that was a whole other thing. It’s just a completely life changing experience.

LEVITT: So when talking about guns, people who study crime use the word instrumentality to explain why guns are so bad. What do criminologists mean when they say instrumentality?

LUDWIG: There are two ways in which having 400 million guns in a country of 330 million people could in principle matter for the crime problem. One possibility is that guns might make it easier for people or entice people to commit more crimes than they would otherwise. The data seem to suggest that’s not such an important thing.

LEVITT: You can rob people at knife point, or you can rob people with a gun.

LUDWIG: Yeah, exactly. The number of guns in a country doesn’t seem to matter so much for the total number of crimes that you have. But the thing that guns do that matters a lot — and this is what criminologists mean when they refer to an instrumentality effect — is that they’re way more lethal than the weapons that people would use to commit violent crimes otherwise. Fists, knives, sticks, bicycle spokes — which is a thing they’re increasingly worried about in the U.K. And so, the main effect that guns have on the crime problem is to make violence much, much more deadly for people than it would be otherwise. 

LEVITT: What you’re saying is obviously completely true, Jens. But It’s a little bit of a puzzle. Because in economics, we have this idea of deterrence — let’s say nuclear deterrence. And the idea is if you have nuclear weapons, then you don’t have to use them very much because they are so lethal. One might think that because guns are so lethal, people would maybe threaten with them but not use them very much. But, I think there’s actually something really subtle about guns which is also critical to why they lead to so much violence. The thing about guns is it makes it hard to predict who’s going to win a fight. So let’s say we’re going to fight with our fists. A fist fight between me and almost anyone else, well, I know who’s going to win that fight, and it’s not going to be me.

LUDWIG: If it’s a fight with me, it’s a coin flip. 

LEVITT: I was going to use that example that I realized, yeah, that actually we’re probably both in the same boat. You and I are not generally going to fight with people because you just don’t need to fight so much when the outcome of a fight is predetermined, say, by one person being much stronger than the other. But guns are a great randomizing device. If I’m carrying a gun, the chances of me winning a fight against another person go up a lot. What does it even mean to win a fight when it’s a gunfight? I’m not even sure what that means. But I do think that is an underappreciated factor of why guns lead to violence.

LUDWIG: Your point makes me think of one of the most widely publicized shootings that Chicago has had. The case of Yummy Sandifer in the early 1990s, who was, I think, like a 10- or 11-year-old kid, who was like five-feet tall and 80 pounds or something like that. It was the availability of the gun that turned him into a net-lethal instrument. I think you look at this and you say, “Guns make crime more lethal when criminals use guns, but they also in principle have a deterrent effect against crime by other people,” and then your point that they add some randomness to fights as well. An economist would look at that and say, “What we’ve got to do is look at the data and see how all of this nets out.” And so what we wind up doing is we look across states over time that have higher or lower levels of gun ownership or different trends in how gun ownership rates are changing. And we see what happens to the crime rate. And it looks like those different forces that are pushing in different directions suggest that when you have more guns in a place on net, the murder rate winds up going up.

LEVITT: It would seem that If guns are at the heart of the elevated murder rate in the United States, then the obvious solution would be to clamp down on guns. But I think it’s fair to say that the kinds of gun control that we’ve enacted, policies that are politically feasible, they haven’t made such a big impact. Do you agree with that statement?

LUDWIG: Yeah, I do for a reason that’s analogous to, you know, an issue that comes up in other policy areas like in environmental policy. So, if you look at the types of gun laws that have been passed in the United States, they’re fairly modest national gun laws. Modest compared to countries like the United Kingdom, that set a floor, not a ceiling on what cities and states can do. And so I think an honest assessment of what the pretty modest national gun laws have done would say, they’ve not had really big impacts on public safety here.

LEVITT: Just to make this more specific, there was something called the Brady Act passed after Ronald Reagan got shot, which I think we’d call the most prominent federal law on guns that we’ve seen in a long time, maybe ever, and you studied that law very carefully. What did you find about the Brady Act?

LUDWIG: Yeah. Phil Cook and I wrote a paper that looked at the effects of the Brady Act. And what the Brady Act did is it said that if you wanted to buy a gun from a licensed gun dealer. You had to go through a background check. Because before that, the dealer didn’t really have to check that you didn’t have a disqualifying felony record. And so that led to a lot of what people called lie and buy. And then it also imposed a waiting period. And, as you say, that was one of the biggest changes in gun laws that we’ve seen in the United States, certainly in my lifetime.

LEVITT: And if I remember, the way you tried to study the Brady Act was there were certain states that already had laws that were as rigorous or more rigorous than the Brady Act. And so they weren’t affected by the imposition of the Brady Act federally. And you could compare what happened to violence in those states to states that didn’t have laws so that the Brady Act was binding. Is that a rough summary of how you approached it?

LUDWIG: Yeah, exactly. You can compare what happened over time to murder rates across the two types of states, and we don’t really see any detectable differences in the trends in what’s going on with murder.

LEVITT: Now let me ask you about it. Because I’ve known you for a long time, and I think you’ve never really liked guns. This is an interesting paper for you because I’m sure you wanted to find that the Brady Act worked, but it didn’t. And as an academic, you had to go and publish this result.

LUDWIG: Well, maybe I would clarify a part of the premise of the question, you know, my dad was in the army. I grew up with three guns in the home. The strongest predictor of whether you’re going to own a gun is whether you grow up in a home with a gun. So, statistically, I’m at high risk of being a gun owner. I’m a data person, like an empirical economist at heart. If the data showed us that like having lots of guns around dramatically improved public safety, I’d say, “Great.” It turns out that the data doesn’t seem to suggest that. Having more guns around leads to more dead people. And so I’ve been then trying to think about what we can learn about how to solve that problem. You ask, how could the Brady Act possibly not have made a difference? And, boy, did Phil Cook and I get a lot of angry emails in response to that paper. But I think, you know, the data say what the data say. And I think the explanation is that the Brady Act requirements of like background check and waiting period apply only to sales by federally licensed gun dealers, and it doesn’t touch the 30 to 40 percent of gun transactions that happen every year outside of a gun store. Phil Cook coined the term “secondary gun market” for those sorts of transactions. If you’re selling me a gun and you’re not a licensed gun dealer, you don’t have to do anything under federal law. And that’s a huge loophole that undermines things like the Brady Act, and I think is the most likely explanation for why Brady wound up not having more detectable impacts on murder.

LEVITT: And related to what you’re saying that makes our approaches to gun control really difficult is that guns last for a really long time. So if you start in a world in which you have 400 million guns. And you change laws that might even dramatically affect the flow of new guns into the system, well, you still have this problem of the other 400 million guns that are going to be around essentially forever. It really seems to me almost hopeless to think that any gun control rules that only work on newly produced guns really can’t have a very big impact.

LUDWIG: I think that is probably true. The only small, maybe friendly amendment I would make to that, maybe like an open question. So most shootings in America are disproportionately committed by young people. And there’s some hint in the data that young people have a preference for newer guns. Because they’re less likely to have been used in a past crime. You don’t want to get caught with a gun that somebody else has put a body on, is the thing that we hear from people on the street sometime. And the degree to which that preference would persist in a world in which you really slowed down the production of new guns, I don’t think we know, but that would be the open question. What you’re saying is totally true. At the federal level, it’s hard to imagine what you could do. At the city and state level, it’s really, really hard to regulate their way out of the gun violence problem in a world in which you have open city and state borders. When you look at the crime guns that get confiscated in Chicago, almost none of them were bought in a gun store in Chicago. They’re all coming from either the Chicago suburbs or places like Illinois and Mississippi and so on. We tend to think, oh, the solution to that would be national laws. But as you point out, it’s really hard to imagine the sort of national law that you could get having a big impact in the foreseeable future. 

LEVITT: So, you’ve written a book, it’s called Unforgiving Places, and one piece of your book that I really love is how you capture the assumptions that people on the right and the left make about the causes of crime, essentially the two worldviews that dominate how we approach public policy. Could you talk about that?

LUDWIG: Almost everybody in America, when you ask them what causes crime and violence, tends to point to one of two types of conventional wisdom. One sort of conventional wisdom is that crime and violence are due to bad people who are unafraid of the criminal justice system. That leads you to think that the solution’s just to hammer people with bigger sticks — more prisons, longer prison terms, that sort of stuff.

LEVITT: The worldview of the right—— 

LUDWIG: Of the right. 

LEVITT: Law and order. And the public policy implication of that view is that we should lock people up, we should have a lot of police, and also, it’s possible that some people are just rotten and no matter what we do, we have to protect society by separating those people out from society.

LUDWIG: Exactly. And that’s obviously been the conventional wisdom that’s driven our public policy or a public safety policy for the last 50 years, since the 1970s. The other segment of America looks at the crime and violence problem and asks, Why are people doing this? And their answer is that it’s bad economic conditions that lead people to act out of desperation. People are going to do whatever they have to do to put food on the table. And under that perspective, the solution is we just need bigger carrots. We need to make the alternatives more appealing to people so they don’t have to resort to crime.

LEVITT: So that’s the worldview of the left. And I think it’s a little broader than that, which is, it’s not just economic. It’s this sense of, there’s a whole segment of a society that has no opportunity that they’re stuck. It’s a natural reaction for those folks to be angry and to want to lash out against society. So, aside from pure economic issues, there’s a frustration — what would be called a justified frustration about the unfairness of our very unequal society.

LUDWIG: Friendly amendment totally accepted. That’s right. And I think the solutions that it leads to in practice, you know, you hear lots of people saying things like, “We’re never going to solve the gun violence problem unless we end concentrated poverty, and segregation, and social isolation,” and on and on. And I think a lot of people hear that and they think, oh, well, one implication of that is like maybe this is a problem that’s just too big to fix. That would make you think that American cities are just resigned to this problem for decades off into the future. So I think there are maybe two interesting things about those conventional wisdom. One thing that’s interesting is we’ve been stuck in this rut for about a hundred years. All the way back to the 1930s, the Republicans were running on law and order. And all the way back to the 1930s, the Democrats were running on addressing root causes. It’s not clear that going back and forth between left and right on these conventional wisdoms has necessarily moved the ball forward. If you look at most leading causes of death in America over the last 100, 150 years, they’ve plummeted. There are two exceptions: One is cancer. If something else doesn’t kill you first, cancer will kill us all in the end. And the second is homicide. The murder rate today in America is almost there’s been ups and downs, as you noted, but the murder rate today is almost exactly what it was like in 1900. The murder rate today in America is off the charts compared to any other rich country in the world and we have the highest incarceration rate. Following this sort of conventional wisdom has led us to have the worst of all worlds in many ways. And the other thing that I would add to that is like we spend a lot of time in economics and public policy more generally worrying about inequality in income, inequality in wealth for totally understandable reasons. We don’t spend nearly as much time as we should talking about inequality and something even more basic and that looms much larger in the lives of so many low-income people in America, unfortunately, which is concerns about public safety. In Chicago, people like to say, “Oh, well, at least it’s not as dangerous as it was at the peak of the crack-cocaine epidemic in the early 90s.” and if you’re white in Chicago, that’s definitely true. If you’re Hispanic in Chicago, that’s largely true. If you’re Black in Chicago, it’s not at all clear that’s true. The murder rate for Black residents in Chicago today is not at all that different than it was in 1991 at the peak of the crack-cocaine epidemic. So, the implication is you’re seeing a widening of disparities in public safety that I think is hugely underappreciated in thinking about the inequality problem in America and what you need to focus on to make it better. And then the one other thing that I just wanted to say about the conventional wisdom is like we’ve been fighting about this between left and right for 100 years, bigger sticks, bigger carrots. And it’s easy to lose sight of the fact that the left and the right actually do agree on something. Both sides implicitly view crime and violence as being the result of some sort of premeditated, rational, weighing of pros and cons before you pull the trigger, commit whatever the crime is. But the data winds up suggesting that most shootings in America are not premeditated or motivated by economic whatever. Most shootings in America are arguments that go sideways and end in a tragedy because someone’s got a gun. That shared assumption is really the launching point for this new way of seeing the problem of gun violence, specifically, that I talk about in the book. 

LEVITT: I definitely can see some logic, some truth in both the worldviews of the left and the right, the conventional wisdom about causes of crime. But at some level, when you think about them, they’re the kind of arguments that You and I would make over a glass of wine at a dinner party. And I think even a simple example can make it clear that there’s a lot more to the story. And the story I’m thinking of actually involves you personally. Can you tell me about the time that a neighbor’s dog tried to attack your dog?

LUDWIG: Yeah, I’m very happy to sacrifice my dignity. it was just such an eye opening insight for me about the problem that we’re trying to solve. I live in Hyde Park on the South Side of Chicago. Everybody there is connected with the university. And I’ve got a 60, 70 pound German shepherd, hound dog mix named Aiko. She is truly a lover, not a fighter. And every Wednesday morning, I put on my running shoes, I leash up Aiko, and I do a call with the leadership of my research center here at the University of Chicago — the Crime Lab. And so I’m walking around Hyde Park, and I round the corner and this unleashed dog comes racing down the driveway, teeth bared, growling, barking like a maniac like it’s going to attack and seriously injure Aiko. And the owner is very leisurely strolling down the driveway after it, as if he’s on a golf course and he’s like eyeing up the chip to the green or whatever. And so, like, if you were to imagine what would the deliberate weighing of pros and cons Jens do in that moment? You would say, I would have had the presence of mind to maybe pick Aiko up, or to try and defuse it, or whatever. None of those reasonable ideas is what I actually did. This dog’s racing down the driveway, teeth bared, growling. I decided it would be a good idea to turn to the owner and say, some version of, “What the f**k are you doing letting your dog off its f*****g leash? This is the third f*****g time this has happened. If it ever f*****g happens again, I’m going f*****g absolutely go crazy, you stupid m**********r.

LEVITT: It’s worst thing in the world, would be your dog being attacked.

LUDWIG: Yeah, it was really like the worst thing in the world is that my dog is going to be attacked. And I responded accordingly. And so the whole leadership team of my research center is hearing me say, “You dumb m**********r. I’m going to f*****g kill you!” And the owner, turns out to be a guy is actually a doctor in the medical center here at the University of Chicago. And worse than that, his kids were actually in the same grade as my kids.

LEVITT: And so did the doctor react really calmly and diffuse the situation? 

LUDWIG: It went off the rails in ways that I would not have imagined possible, were I not actually participating in this. The doctor reacted exactly the same way that I did, which is like, how f*****g dare you yell at me right in front of my own house. So it’s like, he is m*********ing me and I’m m*********ing him out in front of all the other neighbors. And then I’m finally like, f**k you. And I start walking around the corner. I’m walking down another street in Hyde Park, apologizing profusely to the leadership team of my research center. The owner had gotten on his bike and chased me down, hops off his bike and starts yelling at me again. And the only thing that sort of saved me in this was that the University of Chicago worried about the safety of faculty, students, and staff. They hire a bazillion unarmed private security guards to patrol all over Hyde Park. So he’s yelling at me again. I’m yelling at him again. The security guard like turns the corner, rolls down the window of their little patrol car and says, “Is everything all right?” And I’m like, “This a*****e just hopped off his bike and got in my face.” And the security guard says, “All right, I’ll call University of Chicago Police, they’ll be here in a minute.” The guy gets the message, he hops on his bike, he bikes off, giving me the double finger as he goes, and luckily it didn’t wind up something worse than that.

LEVITT: What I love about that story is that it is, I think, to everyone so recognizable what happened. It’s where behavioral economics possibly starts to provide some answers about where violence comes from. It’s what Danny Kahneman calls “System 1” versus “System 2” thinking. Can you explain to someone why those two modes of thinking matter?

LUDWIG: Reading Danny Kahneman’s book, Thinking, Fast and Slow, really was just a life changing experience for me and so many other people. Because the key point of the book is that your mind actually has two ways of thinking of which you are aware of only one. The voice in your head that’s deliberate and rational and weighs pros and cons that is what Kahneman calls System 2. That’s like your slow thinking That’s how we solve complicated problems and, slow thinking is amazing, but it takes a huge amount of mental effort to do. And so our minds have evolved to basically do as little of that as possible. Most of what our minds are doing is instead, operating below the level of consciousness. It’s like this automatic, fast thinking that Kahneman calls System 1 that consists of a series of responses that we’ve developed to routine low stakes things that we see over and over again. And our responses usually work okay for us, but not always. They can sometimes get us into trouble when we deploy them in the wrong situations, like my argument with the neighbor over his dog attacking my dog. Luckily neither the neighbor or I had a gun in that moment. But sometimes someone does and a tragedy results. And so that’s to say, like, I think a lot more of the gun violence problem in America than we’ve appreciated is driven by System 1 rather than System 2. So, fast thinking rather than slow thinking.

LEVITT: And just to make it clear, when you were arguing with that doctor, you were caught up in a moment reacting in a way that was very animalistic, very visceral, and that’s the feature of System 1. And I got to say, my conversations with inmates at the Cook County Jail, that’s the big Chicago jail, are so consistent with what you’re saying. A couple years back, I was there and I was out in the courtyard when the inmates were having their free time. And I was just talking to a few of the people and I asked them for examples that had led to shootings, including shootings they’ve done or they’ve been victims of. And it was stunning how trivial the circumstances were that were instigating these things. One kid had made a social media post where he falsely claimed he had slept with another kid’s mom. And it was totally obvious he had made it up, but that had led someone to get shot or someone bumps into another person walking down the street. It’s really shocking.

LUDWIG: I was visiting the Cook County Juvenile Temporary Detention Center, and one of the staff members there said something that really crystallized all of this for me. He said, “Twenty percent of the kids in here they’re dangerous. You just let them out, they’re going to hurt other people. We have no choice but to keep them in here.” He says, “But the other 80 percent, I always tell them, if I could just give you back 10 minutes of your lives, you wouldn’t be here.” I think that really just has crystallized a lot of this for me. A lot of this is normal people acting like normal people in really difficult 10-minute situations, and unfortunately a gun’s there. 

We’ll be right back with more of my conversation with economist and criminologist Jens Ludwig after this short break.

*      *      *

LEVITT: Okay, and this has all led you to embrace radical new approaches to violence reduction. These are policies that focus on the psychology or the behavioral economics that underpin violence. And I first learned about these approaches from you, actually, many years ago. And I think they’re embodied in some sense best by a guy you told me about named Chico Tillmon. What’s Chico Tillmon’s story?

LUDWIG: Yeah. Chico has — like he’s one of the smartest guys I’ve ever met. I’ve just learned so much about the gun violence problem in America from talking to him. So he grew up in the Austin neighborhood of Chicago on the West Side.

LEVITT: Maybe the toughest neighborhood in Chicago.

LUDWIG: Yeah. Austin is definitely one of the leading contenders for that title of toughest neighborhood in Chicago. A lot of people wonder, why do people join gangs? Chico explained, like, it’s not really as much of a choice as you might imagine, right? You grow up on this block in Austin, all the other gangs in Austin assume that you’re with the gang on that block, and they will victimize you as if you’re in the gang. If you’re not in the gang, you get all of the costs of being in the gang without any of the benefits. So, it’s just a very tough situation lots of young people are in. Chico winds up joining a gang, gets caught up in drug distribution, winds up in federal prison.

LEVITT: He was in there for a long time, right? 

LUDWIG: Long time, 20 years, I mean, just an insane amount of time. So he gets out, he winds up working as a violence interrupter, which I’ll explain in a second, at the University of Illinois Chicago, winds up getting his Ph.D. from there. And so, the theory of this sort of violence interruption, it was developed by this guy named Gary Slutkin. He was on the faculty at U.I.C. for a long time. The idea is really motivated by the insight that a lot of these shootings really are driven by 10-minute windows that people regret afterwards. And I think one of the things that was most amazing for me in writing the book is you realize the country is full of insanely smart practitioners who’ve been working on this year after year after year. And from seeing the problem up close like this, and intuitively understanding it, they improvise and come up with solutions that look very different from what the conventional wisdom would suggest, and that wind up being super helpful. And then, we academics come in later and, do the data work and measure it and show that it helps and then try and figure out why it works. But I think, like, all the credit for all of these, policies go to the people working on the ground who’ve just been so clever and committed to figuring it out. 

LEVITT: So Chico Tillmon, he starts out in a gang, he sells drugs, he gets sent to prison for 15, 20 years. He comes out and one of the few jobs available to people like Chico when he comes out of prison after so long is being on the streets and he has credibility on the streets because he’s been in prison so long, he’s a tough guy. And so they get hired in this job of a violence interrupter where essentially the point is they’re supposed to be there and figure out how to convince tough guys who are about to make big mistakes not to make them. Could you give me some examples of cases Chico’s told you about how he’s intervened? 

LUDWIG: Yeah, and you know, it’s not just having credibility to talk to people and knowing how to do that. It’s also being able to figure out the social dynamics in a neighborhood. And understand when those 10-minute windows are sort of brewing, like knowing who’s got beef with who. Where does someone need to step in and mediate to prevent a shooting from happening? And so, one of the examples that Chico described to me — it’s a little bit like your story of somebody posting something insanely stupid on Facebook. It was like a gang stole some valuable item, like a piece of jewelry from some other gang and then started to talk s**t about it on social media. And the leader of the gang that lost the thing, they’re like, “Well, look, we don’t really care about the thing. That’s not the point. But now that they’re going around and they’re talking about it, it makes us look weak. And so we got to do something, right?” And, you can sort of see that guy’s kind of fast thinking System 1 is doing a version of what I was doing with my neighbor. It’s like subconsciously, your System 1 is telling you, there is nothing worse than letting my dog get attacked. There is nothing worse than letting these guys brag about stealing this thing from us and getting away with it. And once your System 1 has led you to that point, all of the downstream thinking that you do from there is polluted by that upstream mistake. It’s like, well, going out and shooting up the other gang, by definition, can’t be as worse as letting this stand. And so, they’re killing guy after guy after guy in this other gang and Chico finally gets the call to go in and talk to the gang leader. He says to the guy, “What do you want?” Because Chico says, you know, he finds the most powerful way to change people’s behavior is not to tell people things, but to ask them questions. Like, part of what Chico’s doing in these moments is trying to give people’s System 2s, their slow thinking selves, more of a chance to intervene and not leave everything to their fast thinking System 1 selves. And so, Chico’s asking the question, What do you want? Leads the guy to pause and say, “What do we want? What is the endgame here, actually?” And Chico says, “If I found out that you guys are doing this, don’t you think it’s just a matter of time before the detective bureau finds out that you guys are behind all of this as well?” Again, question asking not telling. All designed to get this gang leader to be a little bit more deliberative, and that wound up defusing it. The guy’s like, “You’re right. We’ve killed more than enough guys in this other gang to make our point and we’re done.”

LEVITT: Now, there’s this amazing story and I think you’re right. These street smart, thoughtful, interveners have just incredible insights for people like you and me to learn from. But that’s not really a policy — a public policy towards fighting crime. How have these kind of approaches been transformed into public policies?

LUDWIG: In the last federal administration, they started the first Office of Gun Violence Prevention. There was a lot of attention to community violence intervention. Lots of people looked at that and probably said, “I have no idea what that is.” Cause it — if you’re on the left, you’re like, “That doesn’t sound like solving root causes to me.” If you’re on the right, you would say, “That doesn’t sound like throwing lots more people in prison for a long time to me.” And what community violence intervention basically is, is recognizing that a lot of the shootings are stemming from making a mistake in these really difficult 10-minute windows. And trying to fund organizations to put people like Chico in between two people who are about to like escalate something and turn it into a tragedy.

LEVITT: So one of the organizations is called Becoming a Man. B.A.M. You told me about this organization and I found it intriguing back in the day, but I was extremely skeptical because one thing I’ve learned is it is so hard to change anyone’s behavior. But you actually went out and did randomized studies and wow, the results were incredible, right?

LUDWIG: I started off as skeptical as you, I’m an economist as well. I was raised thinking incentives, incentives, incentives. And so, we partnered with the Chicago public schools and this amazing nonprofit here called Youth Guidance that had developed Becoming a Man. So, what Chico Tillmon is doing is basically like trying to find out when conflict is happening and then stepping in between the two people. And Becoming a Man and programs like that You can think of as very complimentary to Chico and that it’s changing what people bring into those 10-minute situations. It’s helping people anticipate moments like that, and be a little bit more skeptical of their own minds. Let me give you a sense for what these programs are like by quickly talking you through the very first exercise these kids do. So it’s middle school, high school, male students on the South and West Sides of Chicago they’re doing this in school once a week in a group of 12 kids, there’s a B.A.M. counselor there, they’re getting out of an academic class to do it. That’s part of the incentive. The very first exercise you can imagine the kids walking down the hallway, they’re like, “Oh, this is going to be corny. I don’t know what this is.” Very first exercise they pair the kids up. So you and I would be a pair, say. And the B.A.M. counselor would give me a rubber ball and they turned to you and they’d say, “Steve, you have 30 seconds to get the ball out of Jens’s hand. The only rule is, there are no rules. Go.” And you can imagine what these kids do. You try and pry my hand open, you break my pinky. You put me in a headlock. You start punching me in the stomach. Try and bite my ear. Then the counselor calls time. “All right, all right, that’s it.” Now they give you the ball. They say, “Jens, you’ve got 30 seconds.” I do the same thing to you. Then the counselor calls time. They reconvene, and then they debrief. The first thing that the B.A.M. counselor will do then is say, “Steve, tell me what strategies you tried to get the ball from Jens.” And you’ll say, “Well, I tried to pry his hand open, he’s stronger than he looks, and then you called time.” And then the B.A.M. counselor will say, “Steve, why didn’t you ask him for the ball?” We’ve been talking to Youth Guidance for 10 years now about this, and they’ve served, I think, something like at least 10,000 kids in Chicago over that time, and they say, literally not more than a handful of kids ever think to ask. And most of the kids, when asked, like, why didn’t they ask, they’ll say some version of, “Oh, if I would have asked, Jens would have thought I was a wuss, weak.” And then the B.A.M. counselor turns to the other kid and says, “If your partner had asked you for the ball, what would you have done?” And most of the time, the kids say, “I would have been more than happy to give it to him. It’s just a stupid rubber ball.” I think there are a couple things that are insanely clever and powerful about that exercise. And the one is that it’s so engaging to young people, right? It’s like show, not tell. And the second thing is, it is holding up a mirror to the kids about their own minds, showing them how their minds can jump to conclusions about the situation, and that they basically misconstrued the situation. And then more of that over the course of the academic year to help kids just become more skeptical of their mind, consider other possibilities of what you might do in response as they’re going around navigating these 10-minute windows that could lead to conflict out in the world.

LEVITT: How many hours of treatment would be part of this B.A.M. program? 

LUDWIG: Maybe, like the average kid gets something like 15-20 hours.

LEVITT: Okay, so not much at all. Okay?

LUDWIG: Yeah, this is not like a super intensive thing.

LEVITT: Which is why when you described this to me over lunch many years ago, I probably didn’t say anything. But I’m sure inside my head, I was thinking, ‘I cannot believe that Jens is wasting so much time and so much money evaluating a program that is giving kids 15 hours of treatment, when Jens and I are in front of college kids all the time, trying to teach them stuff, and they don’t learn anything in 15 hours.’ Okay? But then, of course, you actually had the data. So tell me, what kind of data did you collect on this randomization, and what did you find?

LUDWIG: We got data from the Chicago Police Department on arrest records for these kids and you can see that the kids who went through B.A.M. had a 50-percent reduction in violent crime arrests compared to kids who were similar on average and didn’t get the program. 

LEVITT: Okay, and 50 percent — let’s just put these magnitudes in perspective—— 

LUDWIG: Huge. Insanely huge. 

LEVITT: Like nothing we’ve ever seen. It’s as big as the incredible crime declines we saw over 20 years in the U.S. Nobody would have predicted this. And these results stuck for a while, too. It wasn’t like it was only in the very short term that these results were happening. They persisted, right?

LUDWIG: Yeah. We track the kids for a bit and you could see 20-percent increases in high school graduation rates as well. And so you can see some persistent change in the behavior of the kids.

LEVITT: Which, again, sounds so absurd. It is just amazing.

LUDWIG: Rahm Emanuel was the mayor at this point. We go to city hall, we present the results to him. He’s like, “Oh my God, this is amazing.” And if you’re the mayor of Chicago, you’re always looking for something to say about public safety other than, we’re going to do the latest, like get tough, whatever. So he announces that the city is going to put a bunch more money into this. And so we said, “Could we randomize that as well to make sure that the first finding wasn’t a fluke?” And we see basically 50-percent reductions in violent crime again.

LEVITT: Yeah, that’s awesome. 

LUDWIG: Scaling this has proven to be a challenge, as is true with every social program. And so I think that’s sort of the next frontier challenge.

LEVITT: You’re saying after this happened, then attempts were made to make it much more broadly available?

LUDWIG: Yeah, so as the program went from serving one or two thousand kids to five, six, seven thousand kids around the city, the data are a little noisier than we would like, but it looks like there’s some sort of decline in the program impact.

LEVITT: Do you think that’s because you run out of people like Chico who are really good at doing this, or because you start applying the program to a set of kids who aren’t going to get as much benefit from it.

LUDWIG: I — uh, I don’t know. I think that’s really sort of the open frontier science question. When you look at social programs, in general, I think the difficulty of maintaining program sort of delivery quality as scale goes up seems to be a consistent pattern in every aspect of social policy. So, I’d be surprised if that wasn’t at least part of the challenge.

LEVITT: So it hasn’t been seamless as we’ve tried to generalize this, but it has been applied in all sorts of places now. I think there was a program in the Cook County Juvenile Center, right? 

LUDWIG: Yeah, we did a version of this in collaboration with the Cook County Juvenile Detention Center. That partnership really got me to see what the at scale version of this could look like. Because as much as Becoming a Man was expanding in Chicago, it still cost a couple thousand dollars per kid. And there’s always the challenge for a city with budget problems like Chicago, is where do you get that money? But, the Cook County Juvenile Temporary Detention Center, a federal judge had said it was so bad that it had to be taken over by a temporary administrator. That guy’s like looking at what’s going on. The kids go to school in the morning and the afternoons, they’re just like sitting there in a common room watching TV with a guard standing there looking at them, making sure nothing happens. And so the administrator says, “Surely we can do something that’s better for the kids than having them watch TV. I’ve got the kids. I’ve got the space. I’ve got the guard. Why don’t we train the guard to do something B.A.M. like?” This is not like gold-plated social policy or silver-plated. This is like tinfoil plated. This is not a criticism, right? He’s just like, “I don’t have any money, but I want to do something better for these kids. What can I do?” And we found that, again, in a randomized experiment, we found that reduces recidivism rates for these kids by 20 percent.

LEVITT: Awesome. It comes for free.

LUDWIG: Yeah, exactly. And some people look at this and they say, “Oh, 20 percent. I don’t know how to think about.” It’s like, look, find me five things that are free that reduce recidivism by 20 percent. And I think I’ve made a big dent in the problem. We’ve been desperately trying to find a public school system or a charter network that might work with us on this. The dream next step on this to think about impact at scale is I think there is, like, similarly low opportunity cost time in the $800 billion a year public school system. I went to high school at Lenape High School in Medford, New Jersey, and the state of New Jersey said I had to take four years of health class. When you look at the data, like, what are the things that are the most important health risks to kids? It is homicide, suicide, and driving accidents in cars. Why in the world are we not trying to claim 15 to 20 hours of that four year health curriculum to do this with kids. If you were serious about changing the health of American teenagers, I think that would be like the most important thing that you could do.

LEVITT: And just to be clear, these are not a set of tools that are only good for preventing homicides. These are a set of tools for whenever anybody gets into an uncomfortable situation. You want a person to slow down, assess the situation in a way that’s actually thoughtful, and then begin to weigh options, and then to make a choice. It’s really very fundamental. It’s a skill that everybody should have, and I think that our education system should do a whole lot more of teaching kids life skills — with this being front and center in life skills — and a whole lot less of how to do abstract math or things that are just honestly much less useful to kids in their life than figuring out how to navigate social situations.

LUDWIG: Yeah, amen. 

You’re listening to People I (Mostly) Admire, and I’m Steve Levitt. After this short break, Jens Ludwig and I will return to talk about one of my more radical crime-fighting ideas.

*      *      *

I have some pretty wacky ideas for reducing crimes. I’d like to share one of them with Jens to get his reaction. Rahm Emanuel hated the idea when he was mayor of Chicago, but of all people, I think Jens might be sympathetic.

LUDWIG: Let me hear it.

LEVITT: So the first one is a campaign that I call Real Men Fight. So the idea is to enlist well known MMA fighters and boxers and professional athletes who would lead a media blitz where they say things like, “Cowards use guns and real men fight with their fists.” So you can imagine billboards and stuff around Chicago. And you would set up boxing rings around the city, and try to create a norm where the socially approved — the honorable way to resolve a beef among these young kids would be via highly public boxing matches. So you’d actually have these boxing rings surrounded by seats so a lot of people could watch the fights. And so you’d offer boxing lessons to the people who are going to be part of these public fights. You try to create honor and respect around the process so that if you step in the ring, even if you lose, you still won. Okay, so what do you think of that idea? And you can say it’s terrible ’cause I know it’s terrible.

LUDWIG: No, no, no, no, no. I think that the objection that people would have to it, reflects a challenge in thinking about public policies in this whole area, which is people are going to say, “Well, we don’t want kids fighting at all.” And I think what is getting missed is, like, it is really, really hard to use public policy to make the world better. And we’re just choosing among a bunch of second best options. I’ve been running this research center at the University of Chicago for, oh my God, 17 or 18 years, the University of Chicago Crime Lab. And we spend a lot of time talking to people in government. And I think it’s just really, really hard to get people to accept the idea that there are trade offs in life. That is sort of the core of your idea, right? In a world in which there’s going to be violence, let’s turn the violence from shootings to boxing. 

LEVITT: Exactly, yeah. It’s accepting there are disagreements. And there’s a culture around manhood. So you learn how to box, you’re in front of everybody, you settle it. Of course, as a public policy, it’s a total non-starter. The immediate, like, oh, well, think about the legal implications if one of these kids gets a brain aneurysm. I think the other thing is, people feel like if kids are getting shot in the streets, that’s not really our fault. But if we organize a boxing match and then the kid gets hurt, then it’s our fault. Which is what you’re talking about, which is the set of policies we’re allowed to do is so constrained that there’s really not that much room for creativity and taking risks. 

LUDWIG: Yeah, harm by commission versus harm by omission, or acting versus failure to act. The other part that gets lost in the conversation is, look, it is totally true that if you had kids fighting each other, there would be kids who would get hurt. But guess what? Look at today’s issue or yesterday’s issue of the Chicago Tribune there are kids who are getting killed every day due to gun violence. It’s hard for people to really wrap their heads around the idea that incremental progress could be progress. 

LEVITT: So you talked about how initially when the Biden administration put in these innovative violence reduction approaches that center around the psychology ideas that we’ve been talking about today, that neither the left or the right really knew how to think about them. With experience now, have you seen an increasing openness from either the left or the right or both towards these new psychology-based approaches?

LUDWIG: It was a big step for the Biden administration to put real federal money into this. And I think that you’re seeing lots of willingness at the city level where people are really on the front lines of the gun violence problem. My hope for this is that it could get us out of the rut of these left-right debates that we’ve been in for 100 years. Because a lot of the fights between the left and the right in the U.S. are around what to do after a crime has happened. Because people have different implicit views about how much of the responsibility is due to the person versus to the situation. But, prevention is the place where people in principle should be in agreement. I think starting around the 1970s, social scientists basically just gave up on the idea that crime prevention was possible. And this new sort of behavioral economics way of seeing the world finally, is starting to give us a glimpse of what really effective crime prevention and violence prevention could look like. My hope would be that there might be a door open to some sort of surprising left right coalition around actual evidence-based gun violence prevention. Because I do think prevention is the place where the left and the right should be able to meet and agree.

LEVITT: So we’ve talked about two kinds of public policy related to these interventions. So the first is really community-based organizations that are trying to do these interventions with kids who are at risk. The second one was, let’s just get this into every school and expose every kid to 15 hours of training about this. Are there even bigger and more ambitious public policies?

LUDWIG: In 1961, Jane Jacobs published a book called The Death and Life of Great American Cities, where she observed, in the same way that I note in my book, like, isn’t it wild that when you look around similarly poor neighborhoods, there are huge differences in violence rates across those communities. What’s going on? And Jane Jacobs put her finger on one particular thing, which she called famously “eyes on the street.” The gun violence problem in America is very disproportionately concentrated in American cities. Something like half of all the murders in America occur in just a hundred cities. And within those cities, it’s concentrated in just a subset of all neighborhoods. So it’s very geographically concentrated. The insight that Jane Jacobs gave us was like urban design, and city planning winds up mattering in surprisingly important ways for this. Because the way that we design our cities shapes the degree to which you have people out and about in public spaces. The University of Pennsylvania did this amazing randomized experiment a few years ago where they worked with the city of Philadelphia to randomly pick some vacant lots and clean them up. So this is a randomized control trial. You can see what happens to gun violence and what cleaning up these vacant lots, and turning them into little pocket parks does, it gets people out of their homes, gets more people out in the public. It’s basically creating more eyes on the street. And what you see in the data is something like 10 percent reductions in shootings as a result. Another example is these researchers in L.A. looked at what happens when restaurants open and close or when marijuana dispensaries open or close. And that winds up having really big impacts on crime, including violent crime by changing the amount of foot traffic out and about. And so, the great thing about that is cities like Chicago and Philadelphia and Baltimore and Detroit, they’re sitting on tons and tons of vacant land that they could do something about. And we have all of these crazy zoning laws that I think the economics research suggests really winds up having a surprisingly important impact on the geographic distribution of things like commercial activity. That we should be thinking about as part of the gun violence policy approach too.

LEVITT: I think many people’s intuition is that well, if you’ve got a lot of people out on the street, that’s going to create violence because it’s these social interactions that lead to fights that lead to people killing each other. It’s so interesting that in fact, empirically, it’s the opposite. If you have enough people on the street, fights will happen. Probably more fights will happen. They just won’t turn out to be so lethal.

LUDWIG: Yeah, exactly. What it also highlights is that the left is right that there is a hugely important component of the social environment in explaining gun violence in America. But the good news is, we’re starting to learn that there are lots of other aspects of the social environment that we can change that matters a ton for gun violence and is easier to change than these really big causes like concentrated poverty and segregation, so on. Some people hear that and they say, “Oh, why are you ignoring root causes?” And I have come to believe that all of these other social problems that cities face, sit downstream of gun violence. There’s a lot of really good evidence that shows that gun violence is a huge headwind for community development and economic development. And the flip side is that if you get it under control, that can be a huge tailwind for turning neighborhoods around. And so I would say, let’s get the gun violence problem under control as one of the most important things that we can do to solve the root causes that we’re seeing in so many neighborhoods around the country.

I expected this conversation with Jens Ludwig to be insightful, data-based, and thought-provoking. Which it was. What I did not expect was that Jens would shatter the all-time record for most expletives in an episode of PIMA. If you want to dive deeper into Jens’s ideas, check out his new book. It’s called Unforgiving Places, The Unexpected Origins of American Gun Violence. Or give a listen to Freakonomics Radio episode 567, called “Do the Police Have a Management Problem?,” where Jens talks about work he’s been doing around training police officers.

LEVITT: So this is the moment in the show where my producer, Morgan, joins me and we tackle a listener question.

LEVEY: Hi, Steve. So instead of a listener question today, I have a follow up question from the interview with Jens Ludwig. 

LEVITT: Great, fantastic.

LEVEY: In the interview, Jens shares a few different personal stories about things that happened to him that sort of gave him this “aha” moment. He’s an economist, he’s studied crime for a really long time, but it took these things that happened in his personal life for him to see the issues in a really different way. As someone else who’s studied crime for a long time, was there ever a moment or a thing that happened to you in the world that made you look at crime and study crime in a different way?

LEVITT: Oh, absolutely. The way you learn about crime when you read academic papers has so little to do with what is actually happening in the world. And so these real world experiences like Jens talked about, they affected me in the same way. And I have so many, but we were working with a homeless shelter, a faith-based homeless shelter. And they had this problem, which is that they had a long, long waiting list. So many people who wanted to use their services, but many people they tried to get out of homelessness through their program, remained homeless at the end. And so the idea we had with them is how could we maybe set up a pre-screen that instead of randomly taking people in whatever order they show up, we would give them a pretest where we would look at various personality characteristics, and we would study what characteristics led people to succeed. So if there was a certain kind of person who could benefit more from their program than a different kind of person, then they should be targeting those people in their program. Does that make sense to you — what I’m saying, Morgan?

LEVEY: Yes, absolutely.

LEVITT: So we set up a bunch of little tasks and we worked with this homeless shelter to have hundreds of people come and do these tasks. These were mazes or things to measure I.Q. There’s a whole bunch of different things we did to try to figure out if any of those correlated with success in the program. And the short answer is that nothing worked very well. But that’s not actually the interesting part of the story. We had a little twist too. So at the end of the session, our worker would say, “Okay, I just need to go get a sign off from my manager to get you paid.” Because they were being paid according to their performance on various tests, and so we had to have a bunch of $1, $5 and $10 bills in there to be able to pay them. And so, intentionally what we did was we had our worker leave the envelope with all of the money right on the table, right in front of the homeless person. We wanted to see how often people stole money and how much they stole, and whether that kind of stealing was correlated with success in this program. Let me ask you, Morgan. Let’s say there were a hundred people who participated in our experiment. How many of them do you think took some money out of the envelope?

LEVEY: Oh my God. I really have no idea.

LEVITT: Yeah, we didn’t know either. But if you had asked me, I would’ve guessed 25, 30 percent, something like that. And partly people wouldn’t want to get caught, you know — but we so tempted them. And the incredible thing was that exactly one person out of over a hundred took the money. And what was even more interesting is that one of the people when our worker came back in the room, said, “Hey, you really need to be more careful because people around here are horrible people. They’re all going to steal your money.” Now, this person who said it happened to be the homeless person who got, by far, the highest scores on all the I.Q. tests — this person was certifiably a genius. And that was the one person who had stolen the money.

LEVEY: Oh my God.

LEVITT: We really struck by it. I have to say, that day spent interacting with a hundred homeless people. It really changed the way I think about things. Not only the lack of stealing, which was amazing, but just the chance to talk with folks and it was, I think honestly the first time I’d really conversed at length with anyone who was — are we supposed to say unhoused? I should use the word unhoused, but this happened so long ago that I don’t think the word unhoused even existed at the time. But I was left with the view that they were no different than me. As I’ve gotten older, I think my views on fighting crime have been more deeply shaped by my interactions with criminals, with victims than they have been by academic economics for sure.

LEVEY: Listeners, we hope you enjoyed the interview with Jens Ludwig. If you have a question for Jens, we can get that to him and possibly answer it in a future listener question segment. Our email address is PIMA@Freakonomics.com. That’s P-I-M-A@Freakonomics.com. Our inbox is always open for questions for Steve Levitt as well. We read every email that’s sent and we look forward to reading yours. 

Next week, we’ve got an encore presentation of my conversation with physician and bestselling author, Abraham Verghese. And in two weeks, I’m so excited to bring author John Green back on the show for a second visit to talk about his amazing new book on tuberculosis.

GREEN: Tuberculosis exists because we allow for it to exist. And to me, that makes it the exemplary disease of injustice.

As always, thanks for listening and we’ll see you back soon.

*      *      *

People I (Mostly) Admire is part of the Freakonomics Radio Network, which also includes Freakonomics Radio and The Economics of Everyday Things. All our shows are produced by Stitcher and Renbud Radio. This episode was produced by Morgan Levey, and mixed by Greg Rippin. We had research assistance from Daniel Moritz-Rabson. Our theme music was composed by Luis Guerra. We can be reached at PIMA@Freakonomics.com, that’s P-I-M-A@Freakonomics.com. Thanks for listening.

LUDWIG: He was like, “Oh, dogs will be dogs.” I was like, “Dogs will not be dogs, m**********r.”