ISIH S3 E12 / Gun Violence in America: More Guns = More or Less Crime?

iTunes Google  SoundCloud  Spotify RadioPublic  Stitcher

John Lott: …the states with the biggest increase in gun laws have the biggest relative increases in murder rates.

John Donohue: …that factor seems to be overwhelmed by the ways in which right-to-carry laws facilitate or increase violent crime…

John Lott: …when people throw out data, they better have a good reason for doing that…

John Lott: I don't want to get paid by people who are involved in the gun debate.

John Donohue: …the states that managed to avoid adopting right-to-carry laws, and now it's a relatively small number of about nine jurisdictions, those states experienced a very substantial drop in crime in the neighborhood of about 43%…

Celine Gounder: Welcome back to In Sickness and in Health, a podcast about health and social justice. I’m Dr. Celine Gounder. This season we’re looking at gun violence in America.

Celine Gounder: Guns are an emotional issue in the U.S. They’re tied up in culture, sport, and, as we’ve seen in previous episodes, with identity and what it means to be an American. And there’s a deeply held belief that guns make us safer. To a lot of people it’s just common sense.

John Lott: …simply brandishing a gun is sufficient to cause the criminal break off an attack…

Celine Gounder: But common sense doesn’t always play out when we look at the numbers… that is, when there are good numbers on guns and crime.

John Donohue: That would be a wonderful number to have…

John Lott: I wasn't really able to get much data after that.

John Donohue: …it's unclear whether it happens hundreds of times a year, thousands of times a year, or somewhat more…

John Lott: There's no reason those mistakes should be there.

Celine Gounder: For a number of reasons there is a lack of good data on guns in America. And that leads to a lot of debate about what can really be said on the topic at a macro scale.

Celine Gounder: In today’s episode we’re going to take a good hard look at some of the studies on guns and crime. What do the numbers say? What are the gaps in our knowledge? Can we can come to an agreement on whether guns… can really... make us safer?

Celine Gounder: There’s no way to have a conversation about guns and public safety in America without this guy.

John Lott: If an attack does occur, it turns out the safest course of action for a victim to take when they're confronted by a criminal is to have a gun.

Celine Gounder: This is John Lott.

John Lott: Guns, obviously, make it easier for bad things to happen, but they also make it easier for people to protect themselves and prevent bad things from happening.

Celine Gounder: Lott is an economist and the president of the Crime Prevention Research Center. But before that, he was a professor at the University of Chicago, who was asked by his students to talk to them about gun regulation.

John Lott: I'd read some papers in the area, and they were really bad, but I just assumed— I hadn't really looked much at the literature. It turns out pretty much everything that was done at that time was pretty poorly done.

Celine Gounder: This was back in the 1990s. Lott saw that the studies at the time looked at guns in one of two ways. Either they were comparing two countries experiences with guns… say the US and the UK. Or they were focusing on just one place over a period of time.

John Lott: All the studies up until that point had been one of those two types of studies.

Celine Gounder: But this isn’t a great way to get at the impact of different policies in different places and contexts over time. For one, it’s difficult to know what’s really going on if you can’t isolate all the factors that could be influencing outcomes.

John Lott: It's only by having enough experiments in enough different years that you can hopefully disentangle these things that are going on. That's called panel data.

Celine Gounder: OK, bear with me here. Panel data is when you look at multiple dimensions of information over a period of time. Maybe you’ve heard of a longitudinal study? That’s basically the same thing. For example, we might look at the impact of smoking on lung cancer... not at one point in time… but over a lifetime. As part of such a study, we’d also want details on other factors that might increase someone’s risk of lung cancer… like exposure to secondhand smoke… or having worked as a miner.

John Lott: The way to solve these problems is essentially to combine time series and cross-sectional data, where you're following lots of jurisdictions over time to see how the murder rate or violent crime rate is changing before and after there's a change in the law, and to see how it's changing relative to other places that aren't changing their laws. If you have enough places that change their laws in enough different years, then you can begin to disentangle things hopefully a little bit.

Celine Gounder: So... kind of like you’d study many people who smoke, who used to smoke, and who never smoked; how much they smoked; and if they got lung cancer… Similarly, Lott set out to study gun regulations. He looked at the years between 1977 and 1992. Specifically, he looked at crime in the eight states that passed right-to-carry laws compared with states that didn’t. The results of that study set off a bombshell in the gun debate.

John Lott: The reason why I end up focusing on right-to-carry laws was because it’s the only one that seemed to make a difference in what was going on in terms of crime rates. It wasn't just that violent crime rates fell after you pass right-to-carry laws, but what you found is that the longer the law is in existence, you saw a bigger drop. In particular, it was related to the changing share of the population with concealed carry permits. If you believe that the greater the risk to the criminal, and that greater risk is going to be related to the likelihood that they'll run into somebody who's able to go and defend themselves. As that increases, you should see greater deterrence

Celine Gounder: Lott’s hypothesis boiled down to “more guns, less crime.” That’s the name of the book he went on to publish in 1998, while he was working at the University of Chicago. Right-to-carry laws—also known as shall issue laws—mean you don’t need to demonstrate “good cause” to get a license to carry a concealed firearm. Lott found that the states that adopted right-to-carry laws saw a drop in murder rates compared to states without right-to-carry. His research also saw a reduction in violent crime—basically, crimes between two people, like a robbery or an assault.

John Lott: Question is why should that be the case? What I would argue is that, again, it's related to the probability that somebody's going to be able to stop a crime from occurring.

Celine Gounder: Think of Lott’s work as the statistical basis for the idea of “the good guy with a gun.”

John Donohue: The Lott thesis was if more law abiding citizens were carrying guns, they could thwart crime when they were accosted by a criminal, and there certainly are examples of that happening, but that factor seems to be overwhelmed by the ways in which right-to-carry laws facilitate or increase violent crime.

Celine Gounder: This is John Donohue.

John Donohue: Yes. I am a professor of Law at Stanford Law School, and also a research associate of the National Bureau of Economic Research.

Celine Gounder: If John Lott made a name for himself saying more guns equals less crime… John Donohue made his... by saying the opposite.

John Donohue: He did econometric evaluation of the impact on crime of the adoption of that law in those eight states and came to the conclusion that right-to-carry laws actually reduced crime. This created quite a stir, and I've been looking at that question almost ever since.

Celine Gounder: John Donohue’s critique of Lott’s research focuses on methodology.

John Donohue: Yes. I mean, it turns out that it's a trickier task than many people realize to try to figure out what the impact of a legal change like the adoption of a right-to-carry law is.

Celine Gounder: One of the biggest issues Donohue has with Lott’s work is the impact of crack cocaine. Between the early 1980s and the early 1990s, the U. S. was wracked by the crack cocaine epidemic… at precisely the same time that violent crime was shooting up. Study after study has shown that crack and the homicide rate were closely intertwined.

John Donohue: The problem that it created was that the original eight states that Lott looked at were largely rural states…  

Celine Gounder: States like Montana, West Virginia, Mississippi and Maine—

John Donohue: …and their experience with crack was very different from states like Illinois, or New York, or California that did have a big crack problem. These crack states were states that did not have right-to-carry laws. So when Lott did his analysis, he found that crime rose more in the states that did not adopt right-to-carry laws than it did in the eight states that did, but it was really a crack phenomenon that was explaining that differential rather than the adoption of a right-to-carry law restraining a crime increase.

Celine Gounder: John Lott again.

John Lott: He can go and argue whatever he wants to about cocaine, but there's like five different ways I dealt with it before he even claimed that I hadn't dealt with it. We had gotten data on cocaine prices by county in the United States over time. We didn't have it over the full period, but we had over most of the time period that we looked at.

Celine Gounder: The way Lott tried to control for the impact of crack on crime was through prices. Donohue says this would be like doing a study on car accidents, and using the price of cars as a surrogate for the number of cars in a particular state. You might have that data, but Donohue argues the price doesn’t really tell you anything.

John Lott: …it's one thing if Donohue had gone and argued that, "Look, Lott and Mustard tried to deal with cocaine, I don't think they dealt with it right. I would have done it differently." That would have been one thing to go and argue, but that's not what he argued. He kept on arguing, and I'm sure you've seen this if you've looked through his papers. He has argued over and over again that we haven't dealt with it. My '98 paper with Bronars, one of the reasons why we set it up, long before Donohue brought up the objections about cocaine.

Celine Gounder: John Lott is talking about a 1998 paper he did after his book More Guns Less Crime came out. In this study, he and Stephen Bronars looked at counties on either side of state borders, where one state had right-to-carry and another didn’t.

John Lott: …that was the paper that had looked at these adjacent counties on opposite sides of state borders. Well, if you have two urban counties right next to each other, adjacent, touching each other, and we had also done it in terms of how far the centers of the counties were from each other, whether they were within 5 miles or 10 miles of each other. You have two urban counties where the centers of the cities are within 5 miles or 10 miles. If you have a crack cocaine epidemic in one urban county, you're going to see it in the other one,

John Donohue: That is actually another terrible Lott paper, the Lott and Bronars paper

Celine Gounder: John Donohue.

John Donohue: …where he tried to make it seemed like he was comparing what was happening  on one side of the line versus another... The way he did the analysis, he was not comparing one side of the line to the other… it was really a crazy analysis. … It aggregated all of the other areas outside the particular right-to-carry state rather than looked directly across the river or whatever the state line was. … Lott makes it seem like if we have a right-to-carry law, the bad guys are going to run away, and maybe they'll cross the border and start doing their crime there. … What I have found is that the bad guys arm themselves more extensively, so rather than run away, they stay where they are, but they just make sure they're carrying guns. As a result, you see the percentage of robberies committed with weapons rises in the aftermath of right-to-carry adoption. We see no decrease in the number of robberies.

Celine Gounder: Lott also cites shootings in gun-free zones… places like schools, movie theaters, and places of worship… as further evidence that guns deter crime.

John Lott: You look at the Aurora Batman Movie Theaters Shooter. Seven movie theatres within a 20-minute drive to the tiller department, only one of them that were showing the premiere of the Batman movie. Only one of them had posted sign banning all forms of concealed handgun. … Last year, we had the synagogue shooting in Pittsburgh. Another gun-free zone, by the way.

Celine Gounder: Donohue thinks this line of thinking gets it backwards. By analogy: Does carrying an umbrella increase the chance rain? Or do you grab an umbrella when you leave the house because you heard it's supposed to rain later?

John Donohue: Louis Klarevas has researched the mass shootings... the factors that usually explain mass shootings are a place that somebody has a particular gripe against, so either a school that they feel they were mistreated at or a workplace. The mass shootings that are done in public are usually focused on some particular grievance or target that seems to be unrelated to whether it's a gun-free zone or not.

Celine Gounder: Does making a building a gun-free zone increase the probability of a shooting? Or are certain places designated gun-free because they’re known to higher risk?

John Donohue: …the NRA position we’re better off if we arm more people and allow the populous to be walking around with guns. The extension of that position is that there shouldn't be any areas where you're not allowed to carry guns because we may invite more criminal activity in the so-called gun-free zones.

Celine Gounder: Take for instance the ultimate gun-free zone: airports and airplanes. Has making them gun-free zones increased the likelihood of violent crime?

Celine Gounder: In 2005, the National Academy of Science’s National Research Council looked at Lott’s research on more guns, less crime. The Council’s report arrived at a very unsatisfying answer: that there was “no credible evidence that the passage of right-to-carry laws decreases or increases violent crime.” The Council basically said, Lott’s argument that more guns equals less crime might or might not be true. But until we have better data, we won’t know. So, in 1999… and in 2002, 2003, 2011, 2014… and again in 2018... John Donohue tried to do just that. He replicated Lott’s analysis and conducted his own... and with each study, he had more and better data at his disposal. So what did Donohue find?

John Donohue: …right-to-carry laws were associated with increases in crime. Generally in the neighborhood of 1 to 1.5% per year after adoptions… so that by the end of a 10-year period after adopting a right-to-carry law, your average state seemed to have about a 13 to 15% increase in violent crime over what it would have had, had there been no adoption of the right-to-carry law.

Celine Gounder: As much as a 15% increase in violent crime in states with right-to-carry laws. That’s a 180 degrees from Lott’s findings. So what would explain such a big gap in the findings? We already mentioned the roll of crack cocaine. How did Donohue get such different results?

John Donohue: I simply tried to follow the same protocol that he had engaged in, but now with a richer data set.

Celine Gounder: Lott looked at data from eight right-to-carry states. Donohue’s sample size was much thirty-three states. He also looked at data from those states over a longer period of time. Lott’s study followed fifteen years. Donohue’s covered thirty-five.

John Donohue: Just having a lot more data made it a more effective study.

Celine Gounder: So, if crack cocaine was such a big issue in Lott’s work, how did Donohue deal with it?

John Donohue: I just thought, ‘‘why not take crack out of the equation, entirely, and just look at 2000 on?'' And, of course, that has benefits and cost, which is, literally, always the case in these empirical studies. The benefit is that we were taking out a confounding variable that no one had a good quantitative control for… Only eight states adopted right-to-carry in the post-2000 period that I looked at and so, that wasn't ideal, but, of course, when Lott wrote his initial paper, only eight states had adopted right-to-carry.

Celine Gounder: There was something else Donohue’s research pointed to—states with right-to-carry laws had higher rates of incarceration.

John Donohue: Yes, yes. Probably one of the most amazing socio-economic phenomenons in the United States in the last 50 years was the enormous drop in crime that we got in the 1990s.

Celine Gounder: Why this happened is still being debated. But one thing people point to is mass incarceration. The War on Drugs drove a fourfold increase in incarceration rates starting in the early 1970s. Research predicts that if you double the prison population, you get about a 15% drop in crime.

John Donohue: In this case, when my study suggested that right-to-carry laws would increase violent crime by about 15%, it shows you that to address a factor that elevated crime by that magnitude you'd need to essentially double your prison population to get us back to where you were had you not adopted a right-to-carry law.

Celine Gounder: In a sense, are people less free in states where you have a right-to-carry law given what happens to the prison population?

John Donohue: Yes, and essentially what has happened is that the states that have adopted right-to-carry laws have experienced an increase in violent crime because of it, but they've tried to moderate that increase by growing their police force and locking up more people.

Celine Gounder: It feels like Lott and Donohue don’t have much in common. But there’s a high price to pay if you want to study guns in America, regardless of where on the political spectrum you find yourself. One of the reasons why there’s so much debate about this kind of research is the lack of good data. Seemingly basic information—like how many concealed carry permit holders are in a given state, or if a concealed carry holder was involved in a crime—aren’t available.

John Donohue: …the NRA has made the job of researchers much tougher by getting all sorts of laws passed to restrict access to data that would be helpful, or to prevent federal funding of research in this area. As you look at the big picture, you assume that the forces that are trying to keep people in the dark have something to hide. I take that as a sign that the NRA crowd knows that if the truth comes out, this will not be helpful to their goals and ambitions.

Celine Gounder: Even Lott has a hard time getting data for his research:

John Lott: I know some mid-level people in the NRA that I've tried to get data from and stuff like that over time. Things like NRA membership by state and things like that. …they initially shared data with me... I wasn't really able to get much data after that...

Celine Gounder: Remember the Dickey Amendment we talked about in the first episode of this season? That 1996 legislation has prevented the CDC from funding research into gun violence.

Celine Gounder: Donohue says he’s gotten grants from the National Science Foundation. And his research has been supported by the universities where he’s worked. Lott, too, struggles with funding.

John Lott: I've not received any funding from the NRA. I have not received any funding from any gun maker or a manufacturer of ammunition or anything like that. The reason why I haven't done it is because I want to be able to say that my work hasn't been biased by any of that stuff.

Celine Gounder: Lott told me his non-profit, the Crime Prevention Research Center, accepts donations from individuals. It also accepts grants from organizations like the Bradley Foundation, which is definitely a hard right organization. The Foundation is dedicated to the “struggle against the forces of tyranny”—a common refrain in pro-gun circles.

Celine Gounder: Then there’s a certain level of harassment gun researchers face. Donohue:

John Donohue: It is a challenging area to work because the intensity and near-religious fervor on the gun side often means that if you say anything that the gun crowd doesn't like, they're very angry, hostile, and of course, armed. When you're dealing with angry, hostile, and armed individuals, there's always a certain level of discomfort.

Celine Gounder: Lott says he was basically run out of academia because of his research.

John Lott: …he's extremely sorry, but that Chicago would continue paying my salary through the end of the year, but that I had to leave the University at that time…  

John Lott: …there was basically no way that I could stay at Yale….

John Lott: I couldn't get another academic job basically because people knew that I would get attacked by politicians for doing the research that I was doing.

John Lott: Anyway, I'm not going to go into it more, but those are just some of the things that had happened.

Celine Gounder: Donohue and other critics of Lott draw parallels between Lott and climate change skeptics. They say… even though the research we have now shows that at best right-to-carry laws may not increase crime... there’s an audience out there for contrarian views.

John Donohue: …at least in the climate change world, you're dealing with a simpler concept because the physical world actually is easier to model than the human behavioral world. … we have perfect data on temperatures and all of these factors, CO2 emissions that go into the climate science, so it is easier. You also have hundreds or thousands of top scientists working on that question. The gun area is a little different because the data is weaker and the number of people who work in this area is more limited... a lot of people just says, "I don't need the headache and the implicit danger that attaches to saying anything that the NRA and the gun crowd don't like." So what you'll get is some really second-rate work along the lines of some of the challengers to climate science… The climate change analogy here works I think as well, that as long as you can cast doubt on the good work, you achieve an important goal for the people who are trying to move the country or policies in a way that's consistent with the bad research rather than the good research.

Celine Gounder: Just as the majority of researchers say the science shows climate change to be real, the majority of gun violence experts, including Donohue, say their research shows more guns lead to more crime.

John Donohue: I think that my current paper is probably the best study available on what the impact of right-to-carry laws is on crime, but that certainly won’t keep the gun crowd from trying to poke holes in it. We'll just see, in the fullness of time, who prevails in the academic world versus who prevails in the political world.

Celine Gounder: Lott’s views on guns weren’t well received in academia. But regardless of what his colleagues think of his work, an increasing number of politicians, especially on the right, leverage Lott’s research in support of their views. The number of right-to-carry states continues to grow.

John Lott: I think it's easier to overestimate the impact that anybody has. It's probably true for my work. I don't think it's had as much impact as some people claim.

Celine Gounder: In fact, the latest trend is permitless carry, which means someone of any age can legally carry a loaded gun without a license or permit. Vermont was always a permitless carry state. Alaska made the switch in 2003, and since 2010, over a dozen states have gone to permitless carry, and most of those also allow open carry.

John Donohue: …we're going to have more information on this, shortly. … If you look at violent crime after Alaska went to permitless carry, which was quite some time ago, violent crime rose very substantially after that. I suspect we're going to see this…

Celine Gounder: These changing laws are natural experiments… further tests of Lott’s hypothesis, “more guns, less crime.”

Celine Gounder: We could very well see some even bigger changes in gun regulations in the years to come. The Supreme Court recently decided to hear its first major gun rights case in almost a decade. This case, New York State Rifle & Pistol Association v. New York, could potentially allow permitless carry nationwide. And one of Donohue’s former students will soon get to weigh in:

John Donohue: …now, Kavanaugh is on the Supreme Court… he was in an economics class I taught at Yale when he was an undergraduate.

Celine Gounder: But not in the direction Donohue’s research would point…

John Donohue: We may be in a situation where the Supreme Court strikes down every gun regulation in America on Second Amendment grounds, which is what my former student, Brett Kavanaugh, seems to want to do based on his earlier decision as a DC Court of Appeals judge in which he would have voted to strike down the assault weapon ban and probably the high-capacity magazine ban, and the registration of guns in the District of Columbia.

John Donohue: in this business, what you ultimately come to realize is, for many politicians and legislators and even judges, what matters is not what is the best science; what matters is, is there something that looks like science that I can use to support the position that I want to reach?

Celine Gounder: Donohue… and others… have discredited Lott’s research… yet Lott continues to have an outsized influence in the American gun debates.

Celine Gounder: In our next episode, we’ll keep delving into the research on whether guns make us safer… or not. We’ll delve specifically into self-defense and guns… when and how people use guns in self-defense… and how that plays out.

Celine Gounder: Today’s episode of “In Sickness and in Health” was produced by Zach Dyer and me. Our theme music is by Allan Vest. Additional music by The Blue Dot Sessions. You can learn more about this podcast and how to engage with us on social media at That’s If you like what you hear, please leave us a review on Apple Podcasts. It helps more people find out about the show! I’m Dr. Celine Gounder. This is “In Sickness and in Health.”