Jens Ludwig on American Gun Violence
Let’s cut to the chase: “The overwhelming majority of murders in the United States involve guns,” says economist Jens Ludwig. “And in fact, most of the difference in overall murder rates between the United States and other countries are due to murders with guns.”
This may seem intuitively obvious to outside observers, but studying guns within the United States has long been a fraught endeavor, and the amount of research isn’t commensurate with the impact on U.S. society. That said, Ludwig has taken on exploring the roots of American gun violence, work that serves as grist for the Crime Lab he directs at the University of Chicago and for many of his books, including his latest, Unforgiving Places: The Unexpected Origins of American Gun Violence. What’s he’s found is that the folk wisdom around gun violence doesn’t rally hold up to the evidence.
In this Social Science Bites episode, he explains to interviewer David Edmonds how – using insights about ‘system one’ and system two’ thinking developed by Daniel Kahneman – cognition in individuals has more explanatory power than traditional variables like poverty, education and environment.
“I think system one plays an underappreciated role in all interpersonal violence, all of the issues, and this way of seeing what is driving violent behavior among people is equally true for knife violence in the UK and on and on,” Ludwig says. “So I think this is really a universal thing about people’s behavior. This sort of frame on the problem helps make sense of a bunch of patterns in the data.”
Ludwig is the Edwin A. and Betty L. Bergman Distinguished Service Professor at the University of Chicago, Pritzker Director of the Crime Lab and codirector of the Education Lab at that campus, and codirector of the National Bureau of Economic Research’s working group on the economics of crime.
He and his labs are routinely recognized for their work. The Crime Lab in 2014, for example, received a MacArthur Award for Creative and Effective Institutions, while eight years earlier Ludwig himself was awarded the Association for Public Policy Analysis and Management’s David N. Kershaw Prize for Contributions to Public Policy by Age 40. Some of the books he’s co-authored or co-edited include 2000’s Gun Violence: The Real Costs, 2003’s Evaluating Gun Policy, and 2012’s Controlling Crime: Strategies and Tradeoffs.
To download an MP3 of this podcast, right-click this link and save. The transcript of the conversation appears below.
David Edmonds: The United States is a gun crime superpower, In the Midwest City of Chicago, the murder rate was down considerably in 2024 — fewer than 600, most of them from guns. Still, the number was about the same as for the whole of the United Kingdom. What explains American gun violence? Jens Ludwig runs the University of Chicago’s Crime Lab and is the author of Unforgiving Places: The Unexpected Origins of American Gun Violence. Jens Ludwig, welcome to Social Science Bites.
Jens Ludwig: Thank you so very much for having me on. This is terrific.
David Edmonds: We’re talking today about killing, and in particular, killing with guns. Violence with guns. You’ve been researching this for years. Tell us about the scale of the problem in the United States.
Jens Ludwig: The United States is a massive outlier with respect to murder rates per capita compared to any other rich country in the world. So our murder rate has fluctuated over the last several decades between five and 10 per 100,000 that’s tens of thousands of people killed every year. On a holiday weekend in Chicago, long holiday summer weekend, there might be 50 or 60 people shot. Our murder rate is multiple, multiple times what you would see in the United Kingdom or in a country like Japan, substantially higher even than what you would see in a country that you don’t think of as being particularly well run, places like Turkey. We’re just unimaginably off the charts.
David Edmonds: And what percentage of homicides involve guns?
Jens Ludwig: The overwhelming majority of murders in the United States involve guns. And in fact, most of the difference in overall murder rates between the United States and other countries are due to murders with guns.
David Edmonds: OK, so you’ve talked about the US being an outlier when it comes to gun violence. Is that true of other types of crime as well, for example, like burglary?
Jens Ludwig: If you compare, for example, the United States to the United Kingdom, the US is much less of an outlier with respect to almost every other sort of non-gun crime — burglary, robbery, assaults, that sort of stuff. And I think another way to sort of see that it’s not something that’s fundamentally different about people in America is that the rate at which people kill each other through means other than firearms in the United States, on a per capita basis, isn’t really all that different from the UK. The real American exceptionalism here is murders with guns.
David Edmonds: OK, so let’s get to the root cause of this, or the root causes, perhaps. The obvious cause is that guns are just a lot more readily available. The US is not like the UK, where it’s very difficult to get hold of a gun here. Now I’m guessing that this is not the only answer, otherwise there wouldn’t be much for you to research. But tell me why the answer isn’t just as simple as I’ve made out.
Jens Ludwig: Yeah, let me start off by acknowledging the first part of the question, which the data suggests, is very true, that the 400 million guns that the United States has for a country of 330 million people is without question part of the story. You know, we have state level data on a proxy for household gun ownership. And we can see that over a decades-long period, the household gun ownership rate between the northern part of the United States the southern part of the United States have been converging over time, and we can see over that same time period that murder rates have been converging across these regions as well. So it’s a nice sort of natural experiment that points to something that’s suggestive of a causal relationship between overall gun availability and murders. And so, if you had a wand that one could wave, that would get rid of the 400 million guns in the United States, I think it is very much true that the United States would become much, much safer. But as you say, that’s not the whole story.
David Edmonds: America is not the only country awash with guns. My elder brother lives in Switzerland, where almost everybody has a gun, but they don’t have the murder rate that you have. So it seems like guns alone can’t be the answer.
Jens Ludwig: Yeah, gun violence is really the product of two things, not just one thing. My little cartoon equation for this in the book is gun violence equals guns plus violence. What you can see in the data is that, for instance, that Switzerland and Canada have almost identical levels of gun ownership, and yet, the murder rate in Canada is multiple times what you see in Switzerland. And I think the explanation there is that the rate of violent crime is substantially different between Switzerland and Canada. And so I think what the data seemed to suggest is that guns don’t cause violent behavior, cause violent crime. Guns make the violence that happens much more deadly. So you can have lots of guns and not many murders, lots of violent crime without guns and not many murders, but if you have lots of guns and lots of violence together, that’s the thing that leads to lots of murders.
David Edmonds: So it’s guns plus something, and the key question is, what is the something? And let me propose an answer, and then you can tell me whether this is right. Again, it seems like there’s an obvious answer. It’s guns plus poverty, or guns plus single-parent families.
Jens Ludwig: You know, the high-level idea that sits over all of this is it’s guns plus violence, equals gun violence. And so then, as you’re driving at the natural next question is, so what is it that is driving violent behavior across countries, across different places, within countries, across people? For hundreds of years, social scientists have thought that poverty was a very central feature of the entire crime problem. What you see when you really look closely at the data is that’s a little too simple. The best sort of natural experiment, research in economics, suggests that changes in people’s incomes change in people’s employment status, social programs that put more money into people’s pockets, do indeed reduce people’s involvement in property crime, but don’t change people’s rates of involvement in violent crime. We have mayors in the United States who say things like, “Violence is the expression of poverty. We’re never getting rid of the violence problem until we get rid of poverty.” And I think that is not what the data seem to suggest.
David Edmonds: So it’s not poverty. Another obvious proposal is that American inner cities are just awash with gangs, and that this is the product of gang warfare.
Jens Ludwig: I think gang warfare really sort of touches on two of the key conventional wisdoms in the United States. So a lot of people look at gang warfare and they think gang warfare reflects poverty. These are people who are economically desperate. They’re selling drugs. They’re in gangs to put food on the table, and that’s not what the data seem to suggest. There are other people who look at gang warfare and say, “I know what I’m seeing. I am seeing fundamentally characterologically broken human beings. These are people without morals who are not afraid of the criminal justice system.” And I don’t think that is what we’re seeing, either. Because when you look at the data, the overwhelming majority of the murders that happen the United States do not seem like they are committed by people who are psychopaths. We have research that tries to understand what fraction of people in prison for murder are psychopaths. You can look at the fraction of killings that are committed by people who are doing serial killings. It doesn’t seem to be that.
David Edmonds: So at this stage, I give up. It’s not poverty. It’s not a product of gang warfare in particular. What’s the answer? Why are there so many homicides as a result of gun violence in the United States?
Jens Ludwig: Yeah. So the light bulb moment for me: I am a social scientist, but I also live in the real world, and one of the ways social scientists learn about the world is by being in the world. I live in Hyde Park here on the south side of Chicago. Every Wednesday morning, I walk my dog around Hyde Park talking to the senior leadership team of the research center I run here at the University of Chicago, called the Crime Lab. So one Wednesday morning, I put on my headphones and my running shoes, and I leash up my 65-pound German Shepherd/hound dog mix. I’m walking down the street in Hyde Park, and I hear ferocious barking and snarling, and I turn and I see a neighbor’s dog off leash racing down the driveway to attack my dog. It turns out that the owner was another University of Chicago faculty member. In fact, their kids go to the same school that my kids do. I had no plans in moving from the neighborhood anytime soon. The rational, premeditated, smart thing to do, had I been weighing pros and cons, would have been, pick my dog up, put myself between me and the other dog. Worst case scenario, I’ve got to take my dog to the vet, then de-escalate. And instead, I took it in exactly the opposite direction, and I lost my mind – I don’t know if this is a podcast where one swears, but I can assure you, every four-letter, seven-letter and 12-letter word I know was deployed at maximum value. And luckily for me, the entire leadership team of my research center is on the phone with me, listening to this whole thing as I am like saying all of these like unspeakably horrible things to a neighbor.
After the fact, I felt enormous embarrassment, for starters, and enormous regret, like, “What in the world was I thinking?” And so that turns out to be what most shootings in the United States are. In fact, most violent crimes in America and in most countries, are not robberies or murders, they’re assaults. They’re arguments that escalate into violence, and in the United States, those tend to end in tragedy because someone’s got a gun.
David Edmonds: So you lost your temper. You’re not saying that, had there been a gun available, you’d have used it. But this is what happens. Generally, people lose their rag, and for 10 minutes they’ve got no control over their emotions. Arguments get out of hand, and because guns are readily available, somebody gets shot.
Jens Ludwig: Exactly. Like I literally lost my mind. I honestly have no idea what I would have done had there been a gun, and I have no idea what the neighbor would have done had there been a gun. I sent a copy of the book to my University of Chicago colleague, Richard Thaler, who won the Nobel Prize for behavioral economics, and his first comment to me after reading it was, “Oh, my God, you’re a hot head.” So there’s a way to look at it where it’s like, “oh, this is very specific to me as a person,” but it is not me. It’s not specific to the character of the individual. All of us are capable of that in these sort of difficult situations.
David Edmonds: If arguments are at the root cause of this, does that mean that young men, who I imagine are more hot-headed than older men say, does that mean they’re more susceptible to this? Does that mean that it’s more like to take place when it’s very hot weather and people get more short-tempered. How is this linked to other data like that?
Jens Ludwig: Yeah, the thing that’s universal here, I’m sure you’ve read the wonderful book by Danny Kahneman, Thinking Fast and Slow.
David Edmonds: He’s been on the podcast.
Jens Ludwig: I feel very honored to be in such distinguished company. So if you read Thinking Fast and Slow, one of the most amazing books I’ve ever read, Danny Kahneman talks about there being two types of cognition that our brains engage in, of which we are aware of only one: the sort of voice in our head weighing of pros and cons, type of cognition that we recognize as cognition is sort of slow thinking, what Danny Kahneman calls system two. That’s what most of our public policies imagine interpersonal violence is driven by, and so our policies have all focused on appealing to that power of our minds through incentives. We’re going to change the carrots and sticks of violence and solve the problem that way.
And I think the data show us the limitations of that way of thinking. But Kahneman’s book shows that there’s another type of thinking of which we’re not aware at all, which happens all below the level of consciousness, automatically. This is fast thinking, what Kahneman calls system one. It is a series of sort of automatic responses that happen quickly below the level of consciousness. We’re not even aware that it happens. They’re designed to usually work well for situations that we see over and over again in daily life. And I think one of the key insights to connect that to my experience yelling at the neighbor is that all of us have system one, system two, running around in our heads and in experiences of really fraught interpersonal conflict, you get a massive adrenaline dump. You know, my friends, Sendhil Mullainathan and Eldar Shafir have a wonderful book called Scarcity that says anything that stresses you out that much depletes mental bandwidth. You are relying on system one. You are relying on system one in those arguments. And so I think that’s the part of it that is universal,
David Edmonds: Yes. And so to get back to my question, is system one likely to come into play more often when the weather’s very hot, for example.
Jens Ludwig: So the baseline is, I think system one plays an underappreciated role in all interpersonal violence, all of the issues, and this way of seeing what is driving violent behavior among people is equally true for knife violence in the UK and on and on. So I think this is really a universal thing about people’s behavior. This sort of frame on the problem helps make sense of a bunch of patterns in the data. For instance, the pattern that we see violent behavior very disproportionately concentrated among late adolescents and 20-somethings. We have a bunch of research from neuroscience that suggests like your brain is still developing over time, and so your executive function control, like your ability to have system two regulate system one in heated moments gets better over time. Men are much less likely to walk away from conflict than women. So I think that explains when you look at the data in every society that I know of, men are overwhelmingly represented in violent behavior. It explains why there is more violent behavior in the summer than winter months. Partly, as you point out, it’s hot, people are more aggravated.
And it’s also the case that it has to do with the nature of the social interactions that are happening out in public places. Most of the stabbings, most of the shootings, are out in public places. And so when you look in the data in the United States, for instance, you see that shootings are disproportionately likely to happen very, very late at night, on the weekends. What’s going on, lots and lots of young people interacting. So there’s lots of opportunity for conflict, and not a lot of, let’s say, middle-aged people like you and me, who are around, who could serve as sort of eyes on the on the street. The conventional wisdom that interpersonal violence, gun violence, is due to poverty or characterologically bad people give you no leverage over explaining those sorts of patterns,
David Edmonds: But it does suggest, then that a standard deterrent, like you’ll be imprisoned for 15 years if you attempt to kill somebody, is not going to work, because these are emotion-driven actions, where people are not reflecting on the consequences of their actions.
Jens Ludwig: Yeah, exactly. I think one way to think about interpersonal violence is it is much more a crime of passion than of profit, then I think we’ve realized and recognizing that rage is one of the most powerful of all human passions.
David Edmonds: I lived where you currently live in Hyde Park for a year, and it really was one of the strangest places I’ve ever been to. When I was there, you were told not to go north of 47th Street, not to go west of Martin Luther King Drive. You were told not to go south of 60th Street, and to the east was Lake Michigan. One of the weird things about Chicago is that the murder rate varies so dramatically from place to place.
Jens Ludwig: Yeah, one of the key questions that I look at in the book is a puzzle that conventional wisdom just helps us make no sense of. So I’m here in my office at the University of Chicago right now in Hyde Park, and two neighborhoods south of us here. It’s South Shore and Greater Grand Crossing that are literally right across the street from one another. Dorchester Avenue, it runs one block from my office. It heads down, and then it just divides South Shore to the east and Greater Grand Crossing to the west. And if you look at the gun laws in those two places, they’re exactly the same. If you look at the criminal justice system that serves the two neighborhoods, they’re in the same city, right across the street, there’s no difference in the criminal justice system. It turns out the poverty rates are the same for the two neighborhoods. Levels of racial segregation are the same. And so one of the key mysteries that got me really interested in understanding this problem more is, like, how can it be that these two neighborhoods are so different? And I think that the behavioral economics perspective on gun violence starts to give us a way to understand what’s going on.
David Edmonds: So tell me the answer. What is the explanation for why one area has such a higher level of gun violence than the other.
Jens Ludwig: It is accelerants to violence differ across neighborhoods, and breaks or impediments to violence differ across neighborhoods. So Greater Grand Crossing and South Shore, same poverty rate and so on, but there’s way more stressors in place in Greater Grand Crossing. For instance, the data we have suggests there’s more disorder and trash and graffiti and all of that stuff in Greater Grand Crossing. There’s more premature mortality in Greater Grand Crossing. So people’s mental bandwidth is much more depleted in Greater Grand Crossing than in South Shore. And so as Sendhil Mullainathan and Eldar Shafir would argue in their book, Scarcity, that means that people in Greater Grand Crossing are navigating daily life, including interpersonal conflict, much more reliant on system one, and so things are more likely to take off and escalate in Greater Grand Crossing than South Shore, not because the people are fundamentally different, but because the social environment is very different, and the mental bandwidth because of the differences in stress across places are so different.
David Edmonds: What then are the policy implications? It sounds like the answer then is to design urban infrastructure in a particular way that arguments are quickly diffused.
Jens Ludwig: Yeah, I think that there are really two sorts of policy responses that come out of this. And again, if you sort of think about South Shore and Greater Grand Crossing, one of the other key differences across the two neighborhoods is the data suggests that there are lots more what Jane Jacobs would have called 60 years ago, “eyes on the street,” in her wonderful book, Death and Life of Great American Cities. So there’s much more commercial interspersed with residential in South Shore than in Greater Grand Crossing. You know you can see it when you’re in South Shore, there’s just lots more people walking around to the corner store for milk or going out to dinner. And once you recognize that most violence is arguments that happen in the moment, you realize that huge wisdom of Jane Jacob.If you read her book, you can see examples of neighbors stepping in and intervening, of interrupting this violence.
And so one thing that we want to be thinking about through our public policy is like recognizing it cannot just be incentives and deterrence and ending poverty. It also has to be things that intervene and step in between people, system two and system one, to de-escalate. The effects are really big, right? There’s a wonderful study from Los Angeles. We had a couple different regulatory or economic natural experiments that cause things like marijuana dispensaries to open or close, or restaurants to open and close. And when those retail establishments open or close, like when those close, foot traffic goes down, and violent crime in that like 1/8-mile radius around there goes up by a lot, on the order of like 10 or 20 percent. So these are not like small effects.
Let me give you one other example of city planning being so important. There was a wonderful randomized experiment in Philadelphia. This amazing research team at the University of Pennsylvania – I have no idea how they figured out how to do this. They got the city of Philadelphia to identify all the vacant lots — trash strewn like dirty, scary places that nobody would want to be around — and randomly picked a subset of those to fix up and turned into what you might call like a little pocket park. And then they compare the vacant lots that got fixed up into pocket parks with the ones that don’t. And you can see that people are way more likely then to come out in public and hang out in or near the pocket park. And, in low income areas, when they fix up a pocket park, shootings go down by something on the order of like 10, 20, 30 percent. So you might look at this and think, “Really? Zoning, cleaning up vacant lots like this feels like such a second-order distraction from the core of the problem.” But adding these together, these are like 40, 50 percent reductions in in shootings, right? It’s amazing. So like, the good news about this is the aspects of the environment that matter so much for interpersonal violence are much easier to change than the huge root causes that the left typically focuses on, things like poverty and segregation, as obviously important as it is to change those, the good news is that there are easier things to change that sort of sit downstream for gun violence.
David Edmonds: Something that intrigues me about your work is that you describe spending a lot of time with police, both on foot and in police cars, and your research is very much enlivened by that. But I wondered whether there was a single thing that you learned from observing the problem at first hand that you couldn’t have got from the data.
Jens Ludwig: Yeah, one of the biggest things that I learned from going around with the police here in Chicago and other cities, that was really one of the early eye-opening moments for me to realize that the gun violence problem is not what I thought. If you watch the news, if you watch The Wire, you really do think of the problem as being driven by like gang wars over drug selling turf, and I remember being in the back of a Chicago police car late one night, there’s a call out over the radio for two giant groups of people in the middle of the street fighting. And so they put on the lights and sirens and they like, race 80 miles an hour, like it’s their mom on the other end of the 9-1-1 call. So they race over there. They get out of the car, and they’re like, “Everybody, shut the fuck up! What’s going on?” They’re trying to get to the bottom of this. And it turns out, what happened is there was a teenage girl on one side of the street who had been dating a teenage boy, literally living right across the street. Somebody broke up with somebody else, someone decided that it was a good idea to storm into the other person’s house and punch them in the face, and then it was just sheer chaos. And the cops are like, “Everybody, shut the fuck up! Everybody go in their house — if we have to come back here, all y’all are going back to prison. Everyone’s getting arrested if we have to come back.”
And I’m doing my best to be disguised as a cop. I’ve got my baseball cap and the Kevlar vest and jeans and sneakers. I left my glasses and pocket protector at home, and I asked the sergeant who was leading the team, I said, “Do you think that’s going to help?” And he was like, “Absolutely not. Like, that’s going to blow up and we’re going to have to come back with way more people later in the night,” And I realized, like, “Oh, I see that is really what the violence problem is.”
And then you can also see there’s a whole part of violence prevention that I just did not understand before, partly through going out with the police. I was like, “Oh, I see. There’s all of this interpersonal conflict, sometimes the police are able to defuse it like that, and sometimes they can’t, and we need other institutions to do that.” And that helped me realize, like in the United States now, there’s a lot of interest in things like community violence intervention organizations. These are people who have had experience with the criminal justice system. They understand, sort of the social scene out on the street. They’re plugged in. They understand when there’s violence brewing, they sort of step in and diffuse that. I would really not have gotten what that whole thing is about had I not been going around with the police, racing out to these scenes and just seeing it.
David Edmonds: You’ve been researching gun violence for many years now. What’s next for you?
Jens Ludwig: This book made me realize that all of economics, all of social science, is really about human decision-making and the importance of mistakes that people make that they regret afterwards, the huge importance of system one. And so my friend Sendhil Mullainathan and I have been trying to think of scalable ways of making progress on helping improve human decision-making. And so the next big thing that Sendhil and I are going to try and make a big push on is, how do we use developments in big data and AI using algorithms — not to replace humans, but to think about algorithms as a way to augment humans, to make people better, particularly at some of these truly life or death decisions that are so important in people’s lives. So if I can just put in a plug, Sendhil gave a wonderful lecture on this at the most recent American Economic Association meetings. It’s just super interesting and wonderful, and I’d recommend anybody interested in that sort of general topic to take a look.
David Edmonds: We’ll see if we can put a link on the website to that. Jens Ludwig, thank you very much indeed.
Jens Ludwig: Thank you so much. It’s really been such an honor and a treat to be on thank you so much so fun.
