A (4:28)

Okay, so I'm going to start out with some big picture questions and then we're going to get into most most of the topics you deal with in the book the natural Experiments, your prescriptions for, for dealing with crime in a smart way. But big picture, one of the most important questions that shapes people's outlook on crime in general and has a political dimension in that conservatives and liberals are quite often split on this question, have different instincts is the so called root causes theory of crime, namely the idea that basically crime is something that has social causes, that people generally don't commit crime because of human nature, but because of something that in principle could be changed about society. Like their upbringing, they had a tough upbringing in a bad neighborhood, or they're poor and so forth. As opposed to the view that what we call crime is just a behavior that for some portion of the population inheres in their nature, it doesn't mean it's the only thing they want to do all day. But humans are aggressive, so some people are going to assault regardless of what society does. How do you think about the question of why human beings commit crime at the most abstract level?

B (5:53)

I think it's Got to be both. Um, I mean it's, it's surely a combination of, you know, our genes and, and who we are as people and then how society is interacting with that. So I would not put myself firmly in one camp or the other. Honestly. I don't spend a ton of time thinking about the root causes of crime. I spend much more time thinking about, like, what's the impact of an intervention on, on someone's behavior going forward. I think there are a lot of scenarios where we would love to know how we got to where we are, but that's actually not that necessary for figuring out how to go from A to B and how to change course. So yeah, so I'm not going to plant a flag either camp there.

A (6:40)

You make a good point because later I'm going to ask you to explain how, for instance, the notion of having a long time horizon versus having a short time horizon influences is like a critical key inflection point for what criminal justice policies actually work. And that's a question that isn't really answered by your position. It doesn't fall out of your philosophy on root causes. And there's a lot of sub questions that are really where, when we're deciding to pull the lever on this policy or that policy, knowing what you think about the root causes theory is not really going to answer that question for you and for most, if not all the questions you're addressing in your book. So I think it is possible to focus too much on those big philosophical questions at the expense of the real questions that we can actually, that actually have consequences.

B (7:37)

I do, yeah. I think just to riff on that a little bit, I think a lot of these issues of like what do you think about the root causes of crime or what's the theoretical lens you bring to different issues there, those are all really helpful for generating hypotheses about what we can do to change what our outcomes. Right. So we've got some trajectory, we have some status quo. We want to know how to get somewhere better. And all of those ideas and conversations are hypothesis generating. But I'm in the business of hypothesis testing. And so really ultimately, and we do this now a lot at Arnold Ventures where I work now, we're funding lots of studies on that are testing lots of things out in the real world. And we try to be really agnostic about what we think might work. So if someone has an idea, even if I really skeptical that it's going to work, if they think it'll work, you know, through their theoretical lens or their Life experience or whatever. Who am I to say they're wrong? Let's take it to the data and let's just go test it out in the real world. So we do need just lots and lots of ideas swirling about what could be effective. But there's no way we're going to know. We're not going to solve all those problems with, we're not going to answer those questions with those root cause conversations.

A (8:50)

Okay, One more big picture question about how criminal justice is discussed in America, in particular in elite spaces, in colleges, at think tanks and so forth. The most popular framework has been mass incarceration. This is the phenomenon that America compared to the countries we compare ourselves to, like Canada and the United Kingdom and Western Europe, Australia, East Asian countries, South Korea, Japan, we incarcerate a far higher proportion of our population than these countries do. Now that's a fact. Nobody disputes that. But to frame that as mass incarceration, I've always felt there was a. I, I disagree in the point of emphasis, which is to say there are countries in the world that have a lot of crime and there are countries in the world that have a lot of state capacity to imprison. And there usually aren't the same countries. Right? So like Brazil has a ton of crime, Mexico has a ton of crime, but they don't really have the state capacity or the, the political ability at the moment to actually put most criminals in prison. And then there's countries that have the reverse phenomenon. They really don't have that much crime to begin with. Japan for instance. Japan just has very little crime. But if they did have crime, you know, they would be able to put, if they wanted every criminal in prison. They have great state capacity. America's in one of the unique situations in the world in that we have a pretty effective government in terms of fighting crime. We have good state capacity, we have courts, we have police officers. It's not super corrupt, but we also have a lot of crime. And so it seems to me the comparisons to other countries which don't take that into account, the unique situation of America, fundamentally frame the problem in a way that's confused. What do you think of that?

B (10:59)

Yeah, I think there are a lot of differences between the United States and other countries just in general, like pick your favorite country. I, yeah, I, I am generally reluctant to make comparisons between US Policy and other countries policies because the baselines can be so different, the populations and social norms, cultural norms can be so different. And so, you know, the idea that we could just take another country's policies and implement them in the US and get the exact same outcome is generally, you know, wishful thinking is in my mind because there's so much other stuff going on. Now again, you know, what other countries are doing and seem to be finding benefits from can generate some hypotheses about what we might want to try in the US But I would want to test whatever it is we're doing. And, and a great example here is kind of Norwegian style or Scandinavian style prison units emphasizing more rehabilitation and freedom and responsibility. You get, you can walk around, maybe you can leave during the day to go, go to work and you have to come back and you have more collegial interactions with the guards. They're not just locking you in your cell 23 hours a day. That is totally possible. It seems to work really well in Norway. Maybe it would work in the US but there are a lot of reasons it might not work in the U.S. maybe we're just, maybe in the U.S. we're just more violent than in other countries. Like, I, I don't know, I'm not going to stake, I'm not going to claim to know that. But we now have a scenario where enough people are interested in figuring out a better way to do things, really forced by major workforce shortages in corrections facilities across the United States, that we are doing randomized trials of these, these Scandinavian prison units in a bunch of states, not just blue states, you know, we've got South Carolina in there, Missouri, you know, so there's sort of a good assortment of states that are interested in just trying anything else that might get us to a different equilibrium. And why not try what Norway is trying? But yeah, I think in general there are a lot of differences across countries. And I think the evidence just within the United States would suggest that we are overusing incarceration on the margin. Sentences are too long from a public safety perspective, we're incarcerating too many people again, from a public safety perspective. But I would base that on studies from within the United States, not cross cultural or cross country comparisons.

A (13:38)

Okay, so the 1960s crime starts going up. It goes up and up and up and up for three decades, peaks in the early 90s, and then goes down and down and down and down pretty much until 2020. This is one of the greatest massive trends in American history, one of the most important trends of the past hundred years. And I've never gotten a clear answer from the people who study it why it happened. Tell me why this happened.

B (14:11)

Yeah, well, it's good that no one you've talked to has given you a clear answer because if they did, they would be, they'd be lying or BSing you. I, I think the reality is we, we probably will never know. It's probably a whole bunch of stuff that changed. You know, during the, the early to mid-90s, people were worried about crime, so we hired more police, we, we built more prisons. So there's, there's some stuff we intentionally did to, to fight crime that surely had some impact. There's also a whole bunch of stuff going on in the background that had been going on over previous generations or at least previous decades that could have an impact on who's committing crime today. We're one is that we removed lead from gasoline. Again, I don't know if we'll ever fully nail down if that is one of the major drivers of this, but we have lots of other evidence now that lead exposure in young childhood, in early childhood, leads to violent behavior and violent arrests in teen and young adult years. So it's totally possible that that was a contributor. We also have, you know, the war on poverty that was, was going on in, in the 60s and that, you know, by the time we have the next generation or two in the early, early 1990s, that might be paying off. Right. The kids of the kids that were affected there are doing better than previous generations had. And so that could have had dividend. So I think the reality is there's just a bunch of stuff going on. I compare it often to these sort of macro level trends, sort of like what knowing what, what's driving crime trends to questions about like what's driving the stock market day to day. It's a whole bunch of stuff. But just like I can tell you how to make a business profitable, even if I can't explain what's going on with the stock market, I can tell you how to intervene to make communities safer, even if I can't tell you exactly why crime rates are what they are or why they're going up or down.

A (16:09)

Okay, so it seems to me one of the most counterintuitive, interesting discoveries in the field of criminology. And correct me if I'm wrong, I think economists, not just you, but others, have had a hand in discovering this and popularizing this, is that when criminals, when people who commit crimes are considering how to behave, they have a much shorter time horizon than the average person does, which means they're not thinking about what could happen to them in five or 10 years. They're thinking very short term. And so it's not that they're irrational, they're rational, but just over a very short time horizon. Why is this such an important observation when it comes to criminal justice policy?

B (17:03)

Yeah. So, you know, we all, we all discount the future in econ terms. Discount the future to some extent. The future is more uncertain than today is. And so we all care a little bit less about our future happiness than present happiness. But it varies across people how much you discount the future. So economists would call this your discount rate. Right. Like how much you, you discount. The future is just sort of a fundamental thing about you that determines a lot of decisions that you make. How much you study for tests, how much you, how much effort you put into your entry level job, whatever. Turns out it also really affects how you respond to criminal justice policies. So if someone is not thinking past today or maybe next week, if we're lucky, then trying to reduce crime, or trying to deter crime in particular by adding years to already long sentences. So if the sentence for a robbery is two years now, we're going to make it five or 10 years. That's not going to register. That's not going to affect your behavior today if you're not thinking past next week. Right. So what does matter is basically any consequence during the time horizon that you're paying attention to. And so the swiftness and certainty of punishment, knowing that with a very high probability you are going to be arrested and put in jail for whatever it is for robbing this convenience store, that's what deters crime. It's not ramping up the potential sentence that you might receive if you get unlucky and are caught. And so what we want to do in that case is shift our emphasis from making punishment more severe or sentences longer to increasing a probability that you get caught quickly, which really is a big shift from how we're doing in the United States today right now. And this is partly just based on what state legislators feel they have control over. Right. If state lawmakers are worried about crime, they can, with the stroke of a pen, change what the sentences are for different offenses. Right. They can change the sentencing guidelines. It's much harder to think of a policy you can implement that will increase the probability of getting caught. That's something that, you know, it's really about police practices and investigations and that's really decentralized. And so it's just harder to directly control if you're a lawmaker. But it's really important that we figure that out because we're basically waste. Again, from a public safety perspective, what we want to do is reduce crime. We're wasting a lot of money on these really long sentences that we should be spending on improving investigations and getting our, our clearance rates up.

A (19:51)

In your book, you write, a simple saliva swab deters crime in a way that long prison sentences do not explain.

B (20:02)

So one of the first studies I did as an academic was looking at the impacts of DNA databases in the United States. So the US and many other countries around the world now have law enforcement DNA databases that contain DNA profiles from people who've been convicted or arrested of certain offenses, as well as DNA profiles from crime scenes that. From crimes that aren't solved yet. And it basically just looks for matches between these two things. And a lot of people freak out when they hear about, you know, DNA databases. They think we're just uploading stuff to 23andMe. It's your whole genome. That's not what it is. It's really just a string of numbers. It's kind of just. You could think of it as a better fingerprint. Um, and, and this being in the database then means that you're more likely to be caught if you reoffend. And so because we. It enables law enforcement from across the country to be constantly comparing new crime scene evidence to your DNA profile. It means that they'll identify you as a suspect in a lot more cases than they might have otherwise if you were the actual perpetrator. And so this is exactly the sort of tool that makes that shift I was talking about. It doesn't do anything to what potential sentence you might receive, what the punishment is for committing robbery, say, but increases the likelihood that you'll be caught if you do commit robbery. And so then the question is, okay, this is a hypothesis, right? What's the real world impact of this? And it turns out we can use the way that these databases are implemented and expanded over time as natural experiments where we can compare people who committed the exact same crime, but one committed just before the database was expanded and the other one committed just committed the same crime just after the database is expanded, that second person goes in the database and the first person doesn't. So we can follow those two people over time and say, what happens to their behavior. It turns out that the second person who's in the DNA database and so is much more likely to get caught. If they reoffend, their recidivism drops dramatically. So I have a study in using Danish data, Denmark has amazing data. So we're able to just get much more, a much better information about what Everybody's doing. And we're seeing drops in recidivism on the order of 40% for this group. And so that's just, it's just a huge impact on behavior when you basically tell people you're now much more likely to get caught if you reoffend. That's the incentive a lot of people need, especially first time defendants or people who are early in their criminal careers. That's the incentive they really need to like clean up their act, make some different choices. And so, so yeah, it's just a much more effective tool, a much more effective policy move than. And making sentences longer.

A (22:55)

Yeah, it's a, I mean, it occurs to me a few ways of understanding this psychological difference between most criminals and most non criminals is like, if you imagine gamblers, right, Gambling is enormously addictive. Lots of people do it, lots of people lose a lot because they do it. If it's because like you could lose your house, right, if you go on a gambling spree, you could also win hundreds of thousands of dollars. But if you were just like certain that you were going to lose a very small amount, that same person probably wouldn't gamble, right? So even though the penalty, if the penalty was only $1, I know I'm going to lose $1 playing poker. That's the game I'm playing with 100% certainty that person wouldn't gamble, but they will gamble if there is a 10% chance they'll lose their entire house, right? And so there is a certain psychology which is insensitive to massive penalties as long as they're lower probability and sort of abstract or far in the future. And then the second touch point that occurs to me is a lot of the way that normal people encounter crime. I guess if you're lucky enough to avoid it on the streets, you don't live in a high crime city is through television and movies. And the type of crime that we, I think are most interested in movies is like the Wire or the Sopranos or Ozark or Breaking Bad. It's like these criminals that are extremely deliberate, not impulsive at all, have very long time horizons. And in fact, if you're in the New Jersey Italian mob as portrayed in the Sopranos, a short prison sentence is actually the cost of doing business for you. And because you're looking at the longer time horizon of the money you're going to make over the next 30 years, as long as you keep your mouth shut and do your eight month stint. And so if that's the portrait of the criminal underworld that you have in your mind, it's going to misfire pretty spectacularly when you try to apply it to the real world where most criminals are doing something really impulsive that they think they can get away with totally.

B (25:12)

And there's actually a recent paper that came out looking at the impacts of longer sentences for white collar criminals and they find that they're very effective deterrence. And so it's really, it's really is these like two sets of, of potential offenders, the super hyper rational ones who are committing, you know, financial crimes. And they've got it all calculated out. They are forward looking often or at least much more forward looking than your typical street offender, I guess. But, but yeah, there's just a really large share of people who are committing violent crimes, property crimes, burglary, theft, those sorts of offenses, who either just naturally are just more impulsive than, than your average person or they're struggling with addiction or a mental health issue. That's, that's just leaning them even more in that direction. So they just don't have the same, that same time horizon. And so it's smart to adapt our policies depending on who we're trying to, whose behavior we're trying to change.

A (26:12)

Okay. About 10 years ago when I was in college, I was reading all the articles about ban the box and ban the box was all the rage. And what ban the box was was if you have a criminal record, you're out of prison, now you're looking for a job. There's a box on the job application which says, have you ever been convicted of a crime? And you know, we've all checked these boxes on various forms, like, you know, visas quite often to other countries require you to check that box. And the criminal justice reform movement on, on the left at the time was very focused on getting rid of these boxes. You shouldn't have to tell your employer. They would say that you have a criminal record. And then I read your, I don't know if it was an article or if it was a paper. And you can explain it saying, actually this hurts the very people that you're trying to help. How is that possible? How could it be possible that getting rid of the requirement to tell your employer that you've been convicted actually harms the people it was intended to help.

B (27:15)

Yeah, this is a classic case of unintended consequences. Right, so, so, so as you said, this policy is, is aiming to help people with criminal records find jobs. We know we have lots of evidence that employers are reluctant to hire people with criminal records. And so One potential solution to that is to just tell them they can't ask if you have a criminal record, at least not until late in the hiring process. So you can do a background check at the very end, but in the meantime, that person's gotten their foot in the door. They maybe get an interview, they have a chance to build some rapport with you. You get to know them as a person, and then when you find out they have a criminal record, maybe you don't care as much. Right. That's the theory. The potential problem is that if employers are worried about the criminal record for some reason, and we're not, you know, we're not. We're not interrogating in any way why they might be worried. We're just saying, well, we'll all accept employers are reluctant to hire someone with a criminal record. They're worried about that record for some reason. We're not changing any of the incentives that they're facing here. We're not changing that underlying concern. We're just telling them they can't ask. Then what I, as an economist would expect is that they would take the remaining information they can see and try to guess who's likely to have a criminal record. And in the United, in. In the United States, because of big racial disparities in. In the criminal justice system, young black men are more likely to have a recent conviction that might worry an employer than other groups are. And so an employer, or an employer who's trying to avoid wasting their time interviewing someone that they're not going to want to hire on the back end might then just be much less likely to interview young black men, especially those with more limited education. And so, you know, that's the theory. That was my theory as an economist, based on everything I knew and the way I thought about the world. But we wanted to see how it would play out in practice. And so I have a paper with Ben Hanson at the University of Oregon looking at the staggered rollout of ban the box policies across the country, across counties and states. And we're able to look at what happens to actual employment across different race and age groups. And we find that employment fell by about 5% for young black men with less education. Exactly the group that we would have hoped that ban in the box would help if it were helpful. Now, at the same time, there was a really awesome field experiment run by Amanda Agun and Estonia Starr, where they sent out job applications from fictional job applicants, randomizing race and randomizing whether the person had a criminal record before and after ban the box went into effect in New Jersey and New York City. And they found along the same lines that when ban the box went into effect, what had been no racial gap in, in. In callbacks widened to a substantial racial gap which implied that so basically employers did discriminate based on the criminal record, but treated white and black applicants the same if they had a record or didn't have a record. And then after ban the box went into effect, the employers seemed to be basically assuming that the black applicants were likely to have a criminal record and they called them back at much lower rates than the white applicants. And so these two papers side by side painted a really compelling picture that this policy was, was really has, has big unintended consequences for a group that's really vulnerable in the labor market. And then, and then subsequently there were, there were studies of looking at like the actual group, the people with criminal records and seeing, well, are they helped at all? Right. So there might be big costs for young black men without a record who now can't signal that to employers. And so they don't get callbacks anymore and they're not getting jobs. But maybe there's, we're helping people with records a little bit and some people might think that that's worth it. And the research has just shown there's no benefit to people with records. And so, so what seems to be happening is employers just, you know, they check the record on the back end and they just don't hire them. Right. They reject them at that point. So anyway, so yeah, this is just an example of a place where it's a really well intentioned policy focused on an important goal like how do we reintegrate people back into society successfully? And it is just unfortunately completely divorced from the evidence at this point.

A (31:47)

Tyler Cowen has a phrase he always writes on his blog, marginal revolution, solve for the equilibrium. Right? And I think this gets at how economists think very differently from people in other fields. When you change one variable in an equation, the other variables don't just stay the same. And when those other variables are human beings with rational interests, they quite often will adjust their behavior to meet the same goals they always have in the new environment. And I think a lot of the most interesting and counterintuitive results in economics come from applying that mantra, solve for the equilibrium to various especially politically charged issues quite often. And I think, you know, I, I don't know if this is true, but in my life experience is true that people first come to the political issues and public policy issues from the point of view of intentions. In other words, this policy is intended to help people with a criminal record. Therefore, I'm going to assume it helps people with a criminal record. But economists have been trained to just basically disseparate intentions from results fully, which I think is probably second nature to someone like you, but it really isn't second nature to most people that are thinking about politics for the first time.

B (33:15)

Yeah, yeah. The other way I put it, I like Tyler's phrase, but I often feel like my main contribution as an economist in conversations is like, when someone suggests doing something to say, and then what happens? Right? Like what then? How, how will people respond? And then, then what happens after that? And then what happens after that? They're sort of, it's almost like we think we change the policy and it's like, great, well, everyone will do what we say and our work here is done. And it's like, okay, but people are going to iterative, iteratively just respond to whatever the change is until we get to, to some new equilibrium in Tyler's, just Tyler's phrase. So, yeah, it doesn't, doesn't necessarily feel like a uniquely economic intuition, but it seems to be economists who bring it to the table most often.

A (34:06)

One of the books I sign in the class I teach at University of Austin is, well, I actually signed two books, the New Jim Crow by Michelle Alexander and John Pfaff's book Locked In. These books have in some ways opposite theses. In some ways, John Pfaff's is like a refutation or an attempt at a refutation of Michelle Alexander's book and the New Jim Crow. You know, it's probably the most, the single most influential book in particular among Democrats for how to think about criminal justice reform. It would surprise me if there was another book on that issue which had been bought by more people and so forth. How does the thesis of each of these book books hold up in retrospect?

B (35:00)

Well, I found Pfaff's book compelling on the kind of the facts about, you know, did was mass incarceration caused by the drug wars or caused by private prisons? He can kind of go through the data and say, you know, that that's not what's really moving the needle here. So on, on those points, I, you know, I, I, I guess I, I, I side with John Pfaff. I think overall I, you know, it was striking to me as I was thinking about writing my own book and sort of thinking about like, is there a place for it in the literature we have? There are lots of books out there about crime and criminal justice policy. It was really interesting that there are, you know, aside from Bapt's book, they're really. All the books are written by lawyers who do bring this very different lens to the, to the table and who are thinking about you. Maybe some sociologists, maybe some others who are really just trying to say, like, here's the problem, here's, you know, here's. We're just going to lay out for you all the bad things that are happening in the world. And that certainly has a place and I think was probably a really important part of why we started having more serious conversations about how to reform the criminal justice system. But it can't be the end of the conversation, right? First of all, I think it's just really depressing and just makes everyone feel like, well, we just have to burn it all down. Like there's no, like, where do we even go from here? This is sort of such a mess. And that's just not how I feel day to day. Doing the research that I do and working in the spaces that I do, There is so much we could be doing that has a really big impact on behavior and could give us less crime and a smaller system, all at less cost and really change people's lives for the better. And I remember seeing the new Jim crow, the forward 20 year anniversary forward or whatever anniversary it was. This was sort of a, an FAQ about what questions she gets often and her responses to them. And one was, well, so what should we do about these problems? And her response in this forward was, I don't know, that's not my job. And I was just like, what? Like this is like. And part of me was just surprised that someone who's been like, you know, clearly a leading thinker on this issue for decades at this point, doesn't have ideas that she's willing to put in front of her audience. Like that just feels like a missed opportunity. And so I was really excited to write a book that's just full of solutions, you know, And I just, I think. And the other pieces, a lot of the books in this space are really focused on because they, they give you this burn it all down mentality. It's, it's like the only solution possible that could possibly make a dent in this, in this, the scale of this problem are massive structure reforms. We really just have to, you know, tear everything down and rebuild it from scratch. And anything short of that is a waste of time and energy. And again, I, from the research, I know that's not true. I know there are a lot of things we can do that are feasible. They might be, you know, seem like very small interventions, but they'll add up and they'll have really big impacts. And also if we're just waiting for a big structure reform, I mean, I don't know about you, but I'm not sure how exactly we're going to get there. That seems to me to be more of a call for inaction than a call for action. And so I'm just a very solutions oriented person. And it's just like I think we should just let's go out there and try to make the world better and not let the perfect be the enemy of the good.

A (38:55)

Yeah, that gives me a lot of thoughts. Okay. One is that, and I think Steven Pinker made this observation when he was defending his book Enlightenment now about how much progress there's been that intellectuals, intellectuals have an incentive to towards pessimism because pessimism easily translates into charisma. I don't know why that is. I think it's just kind of a bug of human psychology that when someone is telling us the end of the world is coming, we experience that as profound. But if someone is telling us, well, actually the end of the world isn't necessarily coming, there's 50 micro solutions where we can make incremental progress and not only prevent the end of the world, but we can actually create a better world. That's like super boring or we think they're trying to sell us something. So, you know, the, the new Jim Crows of the world, the Michelle Alexander's of the world get very big and are seen as very profound without offering any solutions. Because the real solution is sort of to accept that the apocalypse is coming and to repent. But that kind of leaves you. That doesn't really leave you anywhere. Right. Like the day after the apocalypse was supposed to happen, you're still alive and the problems are still there. And so like you, I'm attracted to people that tell me that there are micro solutions, we can make incremental progress. I do notice also that national politics tends towards these massive narratives of oh, burn the system down or the system's perfect. But one of the great things about America is that at the state level, state governments are pretty wide open to reform. If you're able to get the right lobbyists, that's kind of how the system works or even just convince the right people. Republican and Democrat state lawmakers are super open to tweaking with the system and allowing you to study it and those innovations get to spread and they're shorn of the kind of hyper partisan doom and gloom narratives that we see on the national level. So maybe you can tell us a little bit about like what is the most exciting state based results that you found in your research on reducing crime?

B (41:21)

Yeah, well, I'll just start by kind of echoing what you're saying about like everything that's happening at the state level is much more exciting and fills me with optimism in a way that watching cable news about the federal system does not. Right. And so there's just so much good stuff happening at the state level and so many people really trying to solve real problems. They have to balance their budget. Right. Like they have to make these trade offs day to day. And so they're looking for solutions to these tough problems. Let's see. I mean one, one example is, is thinking about how we are deciding when to convict someone of a crime. So one, one. So in the book I kind of stepped through like how to intervene at each stage of the criminal justice system. In this chapter I talk about just, you know, if someone's, if someone's first offense, right. Their first arrest and arraignment for a misdemeanor or their first arraignment for a felony offense. Non violent. Right. Let's talk about non violent crimes. What should we do that with them? Right? Should we be throwing the book at them to make sure that they never come back or should we drop that, that case and essentially give them a warning and see if they'll course correct on their own? And this has been a big debate across the country, especially related to progressive prosecutors who've really been pushing policies like the latter. And this is a place where you can kind of see it going either way. You could argue it either way. And so you really need data to tell you what, what will happen in the real world. And it turns out that, you know, we have some nice natural experiments that essentially randomize people to either have their case dropped or getting a deferred adjudication in the felony case so that they complete a probationary period, then their charges dropped. And very, you know, identical. People who just get lucky and they get the second chance or they don't. And the people who have their, who avoid this first criminal record are dramatically less likely to come back than the people who got unlucky and have this conviction on their record. And again, it's a huge impact. It's like 40, 50%. And it's something where I now know there are states that Are thinking about expanding these deferred adjudication or deferred prosecution policies for, for felony offenses just to help them triage. Like who needs a criminal record, who needs this punishment and who, you know, can. Because this was their, this was their rock bottom moment. They, they were arrested for a crime, they, they face potential consequences. You know, that was enough for them to course correct on their own. They're thinking about expanding their, their deferred prosecution programs. And so these, these types of, I call it in the book, sort of airing toward leniency for first time offenders. That policies that can help push decision makers in that direction and dole out a first time conviction more sparingly seems to have a really big benefit.

A (44:37)

Interesting. So that suggests there's, there's a type of person that will commit a crime once and only once if they are sort of allowed to get a mulligan. They'll be so shocked by their rock bottom moment that they'll never even touch that lifestyle again. And there's the rest of the category of first time offenders that are just like sort of into the behavior and need to be deterred in other ways.

B (45:05)

Yeah, yeah. Obviously this isn't going to work for everyone. And so then when people come back we'll see like, okay, we need to do more for this person. But it's enough of a drop. Like basically the way we've been. If our default is just to convict and throw the book at everybody, we are, it's just an own goal in a way. Right. We're actually increasing recidivism rather than letting them reduce it on their own with less intervention. I do want to deflag. Like this doesn't mean there are no consequences for the crime. Right. I think a lot of people hear this and they're saying like, oh, you think we should decriminalize trespassing or decriminalized shoplifting. And that's definitely not what this means. People have been arrested, they've been probably booked in jail, they had to take time off work. They're worried about what's going to happen in their case. If this is your first offense, like that is all, like that will would be a rock bottom moment for a lot of people. Right. And it's kind of a wake up call that you've really got to change what you're doing here. And so I think just, I think the lesson here is just to learn from that, that that is, that's enough of a punishment and we don't need to also put a criminal record on someone's record, which it turns out, which, you know, we're talking about ban the box. Clean slate policies are the same way where we're trying, we're trying to all these different ways to undo the effect of a criminal record down the road, and none of those policies work. And so it's just really hard to undo the effect of a criminal record. And so if we can just be more sparing in, when we put a crime on someone's record to begin with, it turns out that gives them the room they need to keep their job, keep their housing, and course correct on their own and avoids pulling them into the system. So it is sort of interesting. I think this is an example of where the increasing accountability, increasing the likelihood that people get caught might be a place where people on the right get it a little bit more correct. This is a place where people on the left have gotten it a little more correct that the system was pulling people into the criminal system more than it was deterring them. And these are places where just data is really helpful for helping us figure out which ideas are right and which need a little bit more iteration.

A (47:24)

How big of an issue is recruiting talented police officers? Because the seems to me logical implication of increasing the likelihood of getting caught and that being more important than long prison sentences means we should have more police officers. Now, I don't know if more police officers straightforwardly always translates into a higher clearance rate for crimes, but it seems to me there's probably some relationship. And I know recruitment has been an issue since 2020. You had a kind of mass retirement of police officers in the wake of the Defund the police movement and the Black Lives Matter protests. And now I'm, you know, we're seeing. And we don't have to get sidetracked on this issue, but from my perspective, the incompetence of ICE officers in Minnesota and how they're handling situations much worse than the Minnesota Police Department would handle analogous situations just reinforces the importance of just how crucial it is to have cops that are actually good at their jobs. And so how big an issue is recruiting talented cops in deterring crime?

B (48:42)

Yeah, really important. And to one of your earlier comments, I mean, there is now a lot of research showing that hiring more police officers just on average, without thinking about like, some are going to be better than others, just on average hiring more police officers and increasing police presence in communities reduces crime, especially violent crime. And I think that's mostly coming through this deterrent effect of increasing the probability of getting caught. Right. So it might not translate into actual higher clearance rates because the crime never happens. So this is one of those challenges. We could have a whole other wonky discussion about clearance rates and how to, what it's measuring and all of that. But, but you're not going to steal a car right in front of a cop, right? Like you're very likely to get caught and if you do that. And so just, I think that's, that's the intuition there is that you're increasing the probability that you get caught. And so police reduce crime. Lots and lots of evidence. That's true now. And also, as you, as you said, we are now in a situation where the vast majority of police departments are understaffed. The way they try to handle that is like poaching cops from each other. Right. Which is not, not, you know, scalable solution. That's not going to help anyone kind of long term. And so we need to think about how to help recruit and retain more and different people. Like just by definition, if we want more than we have right now, we're going to have to widen our net and think about, and try to bring in people who wouldn't have thought about being cops before. And I do think this is the type of, it's the type of profession that a lot of people think they, they have a full understanding of what the job is day to day based on maybe what they watch on tv. And the reality is probably a bit different. And you know, there's a lot of just, you know, de escalation and just trying to, you know, help people settle arguments. And, and it's just there's a lot to the job that is different from just, you know, pointing a gun at somebody and making arrests. And so I think finding ways to communicate that. But it's hard. I mean, I think this is something my team is working a lot on. I mean, we do see more and better policing as being a really important component of public safety. And this is a place that police departments across the country are eager for help to recruit and retain officers. And I'm not sure that there's a magic bullet here in terms of what, you know, being able to have some, some intervention that's going to dramatically increase the number of officers available. And so where attention is shifted is trying to help make the best use of the resources we have. So if we have a certain number of officers, how can we use technology to help them become more efficient? You know, AI is, is affecting policing like every other industry. It's helping, you know, take better notes, generate reports. So that you don't have to spend all your time, you know, writing them up after your shift. And then we've got different, you know, high tech tools that help solve cases faster. So everything to be able to help, everything's, you know, potentially useful. Be able to help officers do more with the hours and the day that they have and with the personnel that they have.

A (52:00)

I remember seeing a paper once, I think Alex Tabarrock had reposted it, which claimed to show that, I think it was maybe it was convicted criminals or people awaiting trial or something. It was some natural experiment where people who had committed a crime, some of them also were about to become fathers. They had impregnated someone, essentially, and the fathers who had impregnated someone were vastly less likely to, to, to reoffend or something like that. Do you recall that paper? Is it a. What it was defining and is it a solid one?

B (52:39)

I don't remember the specific paper, but there are, there's a bunch of work, especially from sociology looking at these turning points. So you're much less likely to commit crime after you have a child or after you get married. So those are some, those are pretty consistent findings. And I think our data's getting better, so that's nice. You can kind of like pin it down even more. And you see these big drops which,

A (53:01)

you know, are those like, are those true natural experiments?

B (53:06)

They're. Well, so this is that maybe, maybe is going to get to sort of my, my hesitation to put too much weight on them is, is. It's not, it's not clear what we do with the information. Right. So I think it's, it's. I believe the papers that like after someone, like, I think it's probably. Well, it's. Yeah, let me back up a little bit. I think, I believe, I believe those studies that show that, you know, after you, just after you, you have a child, you're much less likely to offend. I don't think there's something, you know, biasing that data in some way. I think the problem is like, well, so how do we, if what we want to do is figure out an intervention we can use to get people to offend less. Telling everybody to have more children could work. I mean, maybe that's one hypothesis that comes out of this. But it's not random here who's having children, right? I mean, these people are presumably on a different path in some way already. They're in a relationship or something. And especially if the, the paper is showing like, if you get married, you're less likely to commit crimes. Like, well, that's a choice. Right. And so there can be a bunch of stuff going on in your life that is leading you to be making different decisions about risky behaviors related to crime as well. You're not going out and getting drunk with your friends every night when those are the sorts of things that, you know, that's the behavior that leads to bar fights and graffiti and all kinds of things that you get arrested for. And yeah, I just, I think I, I just, I, I struggle with, with what to do with, with research findings like this because it's like, it's interesting intellectually. It's just like, huh, that's interesting that people have these turning points in their lives. But if what we want is a policy intervention, we can use to try to like speed up that process or the maturation or the growing up that people do and you know, try to just spark a turning point in someone's life. These papers are not telling us that. Right. And that's what we really want to know from a policy perspective.

A (55:15)

So what about age? I think, correct me if I'm wrong, one of the solid findings in this research is that as people age, in particular men, maybe women too, but as men age, they are less likely to commit violent crime for probably hormonal and physiological, biological reasons. Yet if, you know, if I, if a 25 year old guy commits an assault, presumably everything else held equal, he's treated the same by the criminal justice system as a 45 year old guy that committed an assault, even though their predicted horizons for future crime may not be the same. What do we know about the relationship between age and crime and are there policy implications? Could there be for that?

B (56:06)

Yeah. Yeah, so you're right. So there's this, this general relationship referred to as the age crime curve. So crime, your, your propensity to commit crime increases through your like mid teens to late teens to early 20s and then starts falling. And after age 25 or so, which coincides roughly to when you know your, your brain is fully developed, you become less impulsive and less reckless and crime falls as well. So, so in general, crime is a young man's game and, and most people are going to grow up and stop doing dumb stuff that gets them arrested. I think. And, but I think the, the policy implications are even, it's, the policy situation is even worse than you're describing. It's not just that 25 and 45 year olds are treated the same. Typically youth is, is held as a mitigating factor in sentencing I do think that most, most judges know this. Right. They know that young people are, Are, Are likely to grow up, and so they're not held as culpable for their actions because it's just you're not thinking, you're not thinking straight when you're young and your brain isn't fully formed. And so people. There is a, so there is a mitigating effect. Youth can have a mitigating effect in sentencing. And I think that's right from a, a perspective of say, like, retribution or mercy or like what people deserve, which is not a matter of what data say, but those younger people are the ones who are at the highest risk of continuing to offend for at least in the short term. Right. And so if you have someone who is. If we all we want. If all we cared about was reducing crime, then locking up people when they're younger would be smarter than locking people up when they're older. And that is counter to the way we usually think about callability for offending. So it's just sort of this interesting tension in what we are using prison and punishment for. And, and then it coincides with. This interacts with this age crime curve in a way that, yeah, leads to complicated policy implications.

A (58:29)

Okay. I also recall one study I read years ago which said that just placing floodlights in a public housing installation at night reduced crime by like 10% or something like that. Is that a solid finding?

B (58:46)

Yeah, pretty solid. I mean, it was a rel. So that was a randomized trial in New York City, and it was a small, small study. So that's the only caveat. But otherwise it was a really cool randomized trial. It's consistent with other evidence that, that just increasing the level of ambient light. So if you look at like before and after daylight saving time, when you're shifting daylight from the morning to the evening, that reduces crime right during that hour. Um, and so we know that again, along with. Along the lines of just increasing the probability that people get caught, that's what deters and prevents crime. If it's easier to see, if it's easier to see either a potential threat or to be able to identify the perpetrator, or if witnesses are able to see what's happening to you, that is all going to have a big deterrent effect on crime. And so at this point, I think the evidence is pretty solid that if we improved street lighting, that would be really beneficial. And I gather there are people working on helping government officials do that in practice. It seems like a really easy thing to do. But a lot of people who are in leadership positions in cities, for instance, don't necessarily know, you know, where, where's the overlap between high crime areas and where the street lighting is really bad. And so if you can just help them with the data and overlying, overlaying those maps, they can quickly get out there and kind of add some more street lights. So there's some people who are trying to just help on a more on the technical side, I guess, of just how do you implement this in practice? But, yeah, definitely a good thing to do.

A (60:22)

Okay, final question. I forget which year it was of the first Trump administration that he passed the First Step act, which was a surprising and bipartisan effort to make sentences more lenient in the federal system. And I think Jared Kushner and Van Jones had something to do with that. How does that reform look to you a few years out? And is there anything in the federal system that you, that you would look for new legislation to solve in the next few years?

B (61:04)

Yeah, well, so the, the piece of the First Step act that I've been most excited about, I know a lot of my team has been most excited about, is, is helping. It's more on the, the back end of figuring out, like, how long do people actually need to be in prison and when could they be released on electronic monitoring to essentially house arrest. And there is a lot of evidence that using electronic monitoring as an alternative to incarceration, either for, for part of someone's sentence or even as an alternative to the entire sentence, that, that is a really smart approach. It's cheaper than prison and it reduces recidivism because you avoid some of the, you know, prison as a school of crime effects. It's just school, you know, prison kind of had about have a bad effect on people. And so it, it produces less crime at less money for less money. And so everybody likes that. And so that was a piece of the First Step act was being able to release people for earlier to home confinement using electronic monitoring. And my understanding is the current Bureau of Prisons, under the current leadership, is trying to fully fund and implement that. It has not been fully implemented yet and is something that we're really excited that the current administration is focusing on. So, yeah, I mean, I think, you know, as we were talking about earlier, so much of the action in this space at the state level, and there is a lot of movement there, especially related to thinking about how to increase the probability of getting caught as a way to bring the two sides together. I think, you know, the, the left has traditionally by kind of the fringe left talks about abolishing prisons and police. And you've got the fringe right that is focused on just throw people in prison and throw away the key. And I think, and that's. Those are really the only perspectives you hear on cable news. And the middle 90% of the country is like, those are our options. Those are terrible. And the reality is there's so much more we can do. And so increasing the probability that you get caught as a way to deter crime is one of the bipartisan ideas out there that actually can bring both sides together. Like everyone can get on board with that, right? Like if someone has committed murder, we should arrest them. Like no one, no one is opposed to that brings justice to victims. And also it will have a deterrent effect if we get increased clearance rates. So we're seeing a lot of action like that on the state level. At the federal level, things tend to be more signaling. I don't want to say just symbolic, but the federal system is just so much smaller because so much criminal justice policy happens at the state level. Like if you're, if you're arrested for homicide, that is a state level crime, that's not a federal offense. And so we're interested in things like FBI guidance on the use of DNA and having the Bureau of Prisons implement some of these policies that can set a really good example for states and be piloting things that states might want to pick up on afterwards. But there is definitely still work to do to fully implement the first stuff act.

A (64:11)

Awesome. Okay. Jennifer Doliak, thank you so much for your time and your book, which I highly recommend is called the Science of Second Chances, A Revolution in Criminal Justice. Thanks so much, Coleman.

B (64:24)

Thanks. Thanks for having me. I really enjoyed this.