Jennifer Doleac studies the economics of crime and discrimination. In July 2023, Jenn will join Arnold Ventures as the Executive Vice President of Criminal Justice.  She also hosts her own podcast:   Probable Causation, a podcast about law, economics, and crime. Her twitter is here.

We chat about trends and causes of crime. How guns, drugs and policing interact with crime trends.

…there was this huge increase in violent crime in particular in the late early eighties, early nineties. And suddenly violent crime started falling dramatically in the mid-1990s. We still aren't entirely sure why that is the case, this big mystery in the economics of crime world. But we do know that basically crime has been falling since then until very recently. So during the pandemic and since the pandemic, we've seen this big uptick in homicide and shootings, at least in the US. Again, we're not entirely sure why that change. It's kind of like trying to describe what's going on in the stock market. There are lots of sort of little blips and everything, and you can have big picture understanding of the economy and what drives growth, but not be able to predict fluctuations in the stock market. So it's similar with crime rates

But overall, we're still in a place where homicide rates and violent crime rates are much lower than they were in the early to mid-nineties. So overall things have gotten much safer, especially in our big cities; we're much safer. But of course, as you said, there's a lot of variation place to place; particular neighborhoods, particular communities, they're the brunt of a lot of violent crime that is still going on. So it's a major public safety or major public problem and concern for policymakers in particular places and that has become more of a focus in recent years as homicides and shootings have gone up, which of course we're not used to after this big decline for decades

What we know of policies that work on reducing crime, and how challenging the recent uptick in crime statistics is to ideas on reforming criminal justice.

We discuss alternatives to jail, and what type of interventions can work on crime, such as sentencing for misdemeanors, and access to healthcare. 

Jenn explains why the “broken window” theory of crime has not really held up. The mixed studies on body cameras and how deterrents (like DNA databases and CCTV) seem to work.

Jenn discusses her work suggesting some policies have had unintended consequences related to “ban the box” (where employers are not allowed to know of former convictions on initial job application), and related to her paper on the Moral Hazard of Lifesaving Innovations: Naloxone Access and Opioid Abuse (which has proved controversial in some quarters).

We play overrated/underrated on: Texas, diversity and universal basic income.

We end on Jenn’s current projects and life advice. Transcript below and video above or on YouTube. Listen below (or wherever you listen to pods).

PODCAST INFO

• Apple Podcasts: https://apple.co/3gJTSuo

• Spotify: https://sptfy.com/benyeoh

• Anchor: https://anchor.fm/benjamin-yeoh

Transcript: Jennifer Doleac with Ben Yeoh (only lightly edited)

Ben Hey everyone. I'm super excited to be speaking to Jennifer Doleac. Jen is about to join Arnold Ventures as head of their work on criminal justice, and I think she does amazing work bringing the economist mindset to challenges related to criminal justice, poverty, and discrimination. She hosts her own podcast, At Probable Causation. Jen, welcome.

Jennifer Hello. Thanks for having me.

Ben (00:27) 

So let's start with crime. When I've looked at the crime data, it occurred to me it's not exactly clear what the trends have been in all different types of crime, or why. My personal reading is it does look like long-term trends over the last 40, 50 years for violent crime seems to be down, and for overall crime as well looks broadly down both in the UK and the US and other rich nations. But there's quite a lot of flux and quite a lot of regional or city differences. And then in more recent years, the trend in crime in anything in the US looks like it might have ticked up. It looks like it might have ticked up in a couple of other places as well, although, again, unflux and unstable in some other nations. So my first question then is what do you think the trends in crime are? And maybe you can cut it in ways which you think are helpful as in violent crime, non-violent crime, crime in cities and cutting it away from the long term and the short term.

Jennifer (01:32) 

Yeah, I mean, I think those long term and short term pictures you just painted are really the main stories when we think about where we are in terms of public safety. So there was this huge increase in violent crime in particular in the late early eighties, early nineties. And suddenly violent crime started falling dramatically in the mid-1990s. We still aren't entirely sure why that is the case, this big mystery in the economics of crime world. But we do know that basically crime has been falling since then until very recently. So during the pandemic and since the pandemic, we've seen this big uptick in homicide and shootings, at least in the US. Again, we're not entirely sure why that change. It's kind of like trying to describe what's going on in the stock market. There are lots of sort of little blips and everything, and you can have big picture understanding of the economy and what drives growth, but not be able to predict fluctuations in the stock market. So it's similar with crime rates.

But overall, we're still in a place where homicide rates and violent crime rates are much lower than they were in the early to mid-nineties. So overall things have gotten much safer, especially in our big cities; we're much safer. But of course, as you said, there's a lot of variation place to place; particular neighborhoods, particular communities, they're the brunt of a lot of violent crime that is still going on. So it's a major public safety or major public problem and concern for policymakers in particular places and that has become more of a focus in recent years as homicides and shootings have gone up, which of course we're not used to after this big decline for decades.

Ben (03:26)

So when I was reading the literature, I was surprised by that observation and the comment that you make is that there isn't that much agreement as to the why on these trends, and the why for the down and the why for the up which is very perplexing, which I guess is one of the reasons why maybe the person in the street has kind of misunderstood what the trends were and certainly misunderstood as to what they were. I guess my thought is what are your best theories of why it was down and maybe why it has come up? What are your top two or three best kind of causal explanations for it?

Jennifer (04:05)

Yeah. So there have been a lot of studies trying to nail down what happened, especially with that nineties decline. All kinds of theories have been thrown around. There was a lot that was changing about criminal justice policy during that period in response to the rising crime rates. We increased incarceration, we increased policing. We know independently that both of those things reduce crime. And so it seems reasonable that we put a lot more police on the street, we lock up people for a long time, maybe that helps to reduce crime. Again, in general that is true, but that doesn't seem to explain what happened in the nineties and why we saw this big decline in crime in the nineties. Other theories out there are removing lead from gasoline and cars. So if you remove lead exposure about 20 years earlier, then those kids who would've been exposed to high lead levels grow up and are less violent, less likely to get themselves into trouble. So maybe we see crime decline for that reason.

Others have suggested that it's a similar story, but abortion legalization. So really like explanations all over the map. My hunch about the nineties is that it's honestly just a mix of a whole bunch of different stuff. I think the lead hypothesis is somewhat compelling, although we, again, whether it could explain the whole decline feels like a bit of a stretch to me. We also were benefiting during that period from a big investment in anti-poverty programs in the sixties, seventies, eighties, and having those kids grow up and have better opportunities surely reduced criminal involvement. So my best explanation is very unsatisfying, I think it was a lot of things in the nineties. For the more recent uptick, again, as you said, we don't really understand what's going on here. I really thought and I think said on some podcasts early on in the pandemic that my guess was it was a lot of everybody is inside, you don't have as many eyes on the street. So there's this Jane Jacobs urban researcher story that basically if you just have more people around who can be eyes on the street, then everything is safer. You've got potential witnesses and so on.

And so suddenly during the pandemic we're not all out and about anymore. So taking the eyes off the street means that people who want to go and cause trouble can without any consequences. So I sort of was hoping, at least, that was a temporary change and you saw an increase in homicides and shootings, but then that would all decline when we started moving around again. It didn't. And so that suggests we're at a new equilibrium almost, so things just shifted. And in addition to a lot of people buying more guns at the beginning of the pandemic, seems like there are just more access to guns. Maybe we got to a point where everyone just has more guns and also there are sort of cycles of retribution if it's gang violence and so on. So we seem to be in a new equilibrium in a lot of cities where gun violence is just higher than it was pre pandemic. So we're going to need to actually do something about it which is harder.

I think all of that said, I mean, what makes me optimistic as a researcher is I think we're much better at figuring out what to do about these problems than we are at explaining why we are in this situation to begin with. So we don't necessarily need to understand why crime is higher now in order to figure out what works to reduce crime. So that is helpful. We don't need to fully under-- Of course, understanding root causes can help us come up with ideas about what to do, but it's not necessary.

Ben (07:50)

Yeah, that's a really good point. So empirically we can do things that you know. So the pollution thing, improving neighborhoods, we might talk about some of this access to healthcare, some of the jobs, cognitive behavior therapy; a lot of things that have been shown to work. I guess speaking from here in the UK and Europe, one thing Europeans can't get their heads around is the whole gun issue. I guess as you've raised it, it seems to European people that guns must be a really big answer to it. I guess this is a little bit contested in the US. From your reading of it, what is the intersection of the whole gun issue and is it a case that because this is all a minority or quite a large minority who own a load of guns and there seems to have been a tipping point in certain communities or regions or zones, like you say, with intersection or with poverty or the environment and communities and it spills over. There also seems to be some intersection with mental health because you've seen mental resilience issues also spark up. So you put some of these small factors together and like you say, you're getting one and one and one at one equals like this really big thing. I'll be interested in your views from that and explaining it to us here in Europe and the UK where we just don't quite understand the issue around guns.

Jennifer (09:16) 

Yeah, I think a lot of people in the US don't really understand either why we're in this situation. A lot of people believe very strongly in Second Amendment rights to be able to own guns. So that makes meaningful gun control just very difficult politically to implement. And so I often think of conversations about gun laws as being a little bit of a distraction if we're focused on like what policies can we implement to actually change things because it's just going to set off; it's just going to cause lots of fights. But all that said, I think one challenge here and the reason it's not sort of an immediate one-to-one, like someone buys a gun and we see crime go up. I think that there are lots of law abiding people who buy guns. One big challenge is that a lot of those guns get stolen. So I've actually been having a lot of conversations with people recently including police officers who say like they know-- I live in Texas-- You see a pickup truck in a parking lot, they could guarantee you there's a rifle in the back of that pickup truck. Thieves know that too. And so you wind up seeing a lot of guns stolen out of cars. A lot of people just store their guns in cars and they're not locked up. So it's interesting, I've just been having a lot of conversations about like, "Maybe we could get policies passed about having to lock up your guns when they're in the car."

So there are some places like that where you could imagine some sort of interventions that just try to keep the guns in the hands of the law abiding citizens rather than the people who are stealing them. But otherwise, it's a much bigger conversation. Surely if we implemented really meaningful gun control, I think the best evidence shows that that would dramatically reduce violent crime. There is uncertainty there, of course. These studies aren't perfect. We don't randomly pass these laws, they're really contentious. We can have a much longer conversation with the literature there. But I think in my mind the big question is whether we're actually going to do it, and it seems very unlikely to me, so I don't spend a whole lot of time talking about it.

Ben (11:33)

So the economic case is clear, but the political world is not. In some ways that's a little bit like carbon tax, although carbon tax is probably more contested. Most economists, I think some ridiculous number like 80 or 90% of economists think carbon tax is the answer. “We do that, we price the externality, what's the problem people?” And then all of the political economy people are going, "Well, in a democracy it doesn't quite work like this." So that's another example, but it's quite interesting on that. We mentioned guns. I guess the other two big ones on there are kind of drugs and poverty with some of these intersections. I was really interested in the drugs one. Mostly I was reading the stats on alcohol in particular which seems to be true across-- Well, I only looked at European nations and the US. But something like alcohol was involved in between 30 to 50% of violent crimes, 30 to 50% of murders across all of these nations, whether you're Scandinavia, continental Europe, UK or the US, which was really amazing; domestic violence. So all of these violent ones. Obviously that's kind of legal drug, but then the second order on drugs and things. So it seems to me that people think that the link of alcohol is pretty causal. Would you agree, and I guess are there some policies that we can do or think around that, or is that again tricky because of people on the other side law abiding and the rights to alcohol and things like that?

Jennifer (13:05)

Yeah, the evidence on that is very strong. Alcohol is really bad. Occasionally, some reporter will write up a story about, like, "Imagine there was a drug that we knew increased violent crime by X percent and had all these terrible outcomes, and we heavily regulate or make illegal all kinds of substances that because we were worried about the criminal effect. And then they reveal at the end what we're talking about alcohol, which is obviously quite legal and easily available.” So there are policies out there, a lot of which are contributing or allowing these studies to figure out what's the effect of alcohol on crime. Taxing alcohol; so increasing alcohol taxes tends to reduce violent crime.

Something I've been really interested in and most of my own work is on how to help people reintegrate into society after a conviction or time in prison. So I spend a lot of time thinking about the kind of reducing recidivism piece. There are a bunch of studies of this program called 24/7 Sobriety, which was started in South Dakota and is now expanding to some other states where they frame it as basically like if you are engaged in crime that is alcohol involved, you essentially lose your license to drink. And so as part of this program, you have to basically come into the local jail and breathe into a breathalyzer at random several times a week, or actually wear one of those bracelets that measures your blood alcohol content. The idea is if you are caught drinking, then you are immediately put in jail for a night or two.

So it's basically a hundred percent certainty of getting caught if you drink, but a very light sentence or a very light consequence. And basically they dramatically reduce the amount of drinking that people are doing. Also, most of these policies are targeted at people who have DUIs; driving under the influence offenses. But they also see reductions in domestic violence and they actually see reductions in mortality down the road. So it just has big benefits. I think as you said, if we were thinking we could just ban drinking in general in order to reduce crime, I think that's a political non-starter. Lots of people see lots of benefits to drinking and enjoy it. But I think it does raise questions of if we trust people for the most part to do this responsibly, perhaps we could trust people to do other things responsibly and then punish the violations of that rather than banning the substance outright. But finding ways to discourage it would certainly help public safety benefits.

Ben (16:07)

Sure. That makes a lot of sense. But it's kind of interesting because-- I guess it's slightly mixed, but I tend to read that a lot of people think the war on drugs though hasn't been particularly helpful or at least have had pros and cons. So strangely, the alcohol thing we kind of agree on that. But then on these other ones, the side effects of the war driving crime into the black markets and things, with an economist mindset, again, not sure it's politically palatable. But that seems to have gone the other way. Is that also kind of your reading? It's a little bit more contended I guess.

Jennifer (16:43) 

With the war on drugs?

Ben (16:44)

Yeah, with the war on drugs.

Jennifer (16:45)

I mean, obviously in the US, we're in the middle of a big experiment about legalizing marijuana, the easiest of the drugs to legalize. In general, yes, my read is that the war on drugs has not been particularly successful, has created black markets, has increased crime in other ways. All of that said, there are a lot of drugs that do have really negative externalities in econ terms. If using substances only affected you, then perhaps we'd all be okay with everybody just making their own decisions if we assume you have full information and everything else. But knowing that a lot of drug use does have the possibility of really negatively harming people around you-- as we just discussed alcohol does-- then there's an incentive for the government to regulate it. And so the question just what that regulation looks like is complete criminalization and throwing the book at you and putting you in prison for a long time. Is that the best way to handle it? Probably not. So I think in general we're in the middle of finding a middle ground and finding ways to back off this thinking that we can just convict and incarcerate our way out of people using drugs. But we haven't found the best solution yet.

Ben (18:07)

Sure. So there's going to be a lot of good, natural design experiments coming up now with all of this.

Jennifer (18:12)

Let's hope so.

Ben (18:15)

Which is a good segue into I think one of the papers I read on your work, or it might have been your work sort of looking at this, which was looking at misdemeanors and particularly sort of small offenses and how harshly you come under judgment and things like that. The work kind of basically suggested-- I'm going to say it in real lay terms-- But if you were kind of given strong warnings for small offenses and you weren't put into the system, the chances are it was better for you in terms of then not getting into system or re-offending and that type of thing. So that's one possible policy that could work. And actually there seemed to be some natural experiments would support that would also help sort of the fact that prisons are really expensive. So that type of thing. Is that a true reading and how certain do you still feel of that idea on that work now?

Jennifer (19:13)

Yeah, I think there's growing evidence that we've been, especially in the US again, very concerned for a long time about mass incarceration and the effect that really long prison sentences have on people and then communities. And this idea that locking people up then makes it really difficult for them to get a job later and everything else. There's more and more evidence that really seems to show it's not the incarceration that seems to be doing most of the damage, it's the initial conviction. So if you put anything on someone's criminal record; that initial misdemeanor, even an arrest honestly can often show up. But let's just imagine here for the sake of argument, it's a misdemeanor conviction, a felony conviction. A lot of people who are convicted of these crimes never go to prison but still have a very difficult time then finding a job, keeping a job if they had one, finding housing. These records are permanent. It's public record that you have this kind of record probably for good reason. We don't want the government to just convict and incarcerate whoever without telling us.

But this all means that giving someone a criminal record is really costly over the long term. So I think the evidence in general, my read of it is that we've gone too far basically in how easily and quickly we give out criminal records and we should be airing toward leniency, especially for first time offenders. So that research you mention with Amanda Agan and Anna Harvey when we were working with Suffolk County in Massachusetts where Boston's located. We basically had a really nice natural experiment where non-violent misdemeanor cases; stuff like disorderly conduct and minor drug possession and shoplifting are essentially randomly assigned to different district attorneys who decide in this first court hearing whether to just drop the case, it's not worth the government's time, or to move it forward and let a colleague take it and pursue it.

We found that if you get lucky and get one of these lenient prosecutors that drops the case, you're 50% less likely to show up back in court again with a new charge. So it actually dramatically reduces recidivism. There's other work showing something similar with felony defendants. If you kind of put them on probation and then wipe the charges off if they successfully complete probation, similarly, about 50% are less likely to come back. So the way I often describe it to people is like, especially these first time offenders when they come into court, they're really at a fork in the road and you can either give them a conviction and pull them into the system, or you can just send them on their way and hope that this initial contact was enough of a wakeup call and they'll kind of figure it out themselves. And it turns out for a lot of people, they would. That is enough of a wakeup call and they would course correct without the conviction and punishment that the court system would implement.

And so knowing that, that just all strikes me as meaning that we are currently over convicting and over prosecuting, and we should just pull back a little bit and let people course correct on their own. Doesn't mean no punishment, doesn't mean like decriminalize these offenses or don't arrest them, but we don't need to be convicting people and throwing the book at people for everything.

Ben (22:34) 

First time offenses, particularly more minor offenses, although even felonies. And it does seem to be true outside of the US as well. What I like about what I was reading is, like you say, it's replicated across a different couple of areas. The numbers are also quite big. It's not one of those where you've got just a five or 10% could be kind of within forecast era. So essentially replicated. And also the causal model just really rings true. Interviews I've read, there are some people-- Yes, there will obviously be some bad apples or not. But some people will say, "Yeah, this was a wakeup call, I changed my life. This was not the person I wanted to be." So it makes a lot of sense. My view generally is that I find policymakers tend to be open to economists. There might be political constraints which mean it's not practical. Do you feel this is something that policymakers have potentially been able to pick up and maybe the politics of wanting to be seen to be hard on crime means that it's not possible? There has been a little bit of take up. I mean, where do you think we are with this policy?

Jennifer (23:43)

So one of my favorite things about working on criminal justice policy broadly is that there's just really broad bipartisan consensus that our current criminal justice system is not working very well. Incarceration rates are too high. It is helpful here that incarceration is so expensive. So even if it's just like small government conservatives are worried about spending lots of money on a field program, incarceration is a great example. So that means that there are a huge number of people of all political stripes that are interested in what the evidence says about how to fix things. And on most of these topics, there isn't a left and right policy yet. That's probably because most of this is state local and the federal government doesn't get involved as much. So it becomes less fodder for cable news, which is great.

I think all of that said, one challenge in the current conversation about whether prosecutors should be airing turn leniency is in the US, we've been having this conversation about progressive prosecutors. So they've been the ones really pushing policies like this. A lot of them have been elected around the same time that the pandemic happened and you see this increase in homicides. So it has become a very hot political issue whether these progressive prosecutor policies are what are causing the rise in homicide. My read of the evidence is that that is absolutely not happening. But there are variations in policies across places. Maybe there's some places where contributed, who knows? But it is at least an easy story to tell.

So I think that is sort of a one short term blip in this conversation about, "Could we go easier on defendants, especially first time defendants?" But I'm quite confident that over the longer term it'll get a lot of traction. There are a lot of policy makers in the space that honestly just are trying to figure out how to make things work well. Their courts are overwhelmed and crime rates are always a concern. So if they have a policy that is this effective-- And again, we're talking here about just like airing more turn leniency. It's not like everybody gets one crime free or something. It's not a totally guarantee of leniency.

Ben (26:07)

And it's cheap.

Jennifer (26:11)

Exactly. It's cheaper than the status quo and you get less crime. It feels like this should be an easy sell. So everyone I talk to gets it. It seems intuitive. And I suspect these sorts of changes are happening. They're just a little bit-- They're not advertising them and it's not the sort of thing you want to get into a fight on cable news about. But I think a lot of offices are paying attention to the evidence and using it accordingly.

Ben (26:46) 

Sure. That makes sense. That also brings to mind the question that Tyler Cowen asked, and I think you kind of part answered, which was with this recent uptick in crimes. “Does this mean the end for criminal justice reform?” And I think your answer was, "Well, it's a blip, but hopefully over the longer term actually because we don't quite understand why there's this uptake, but we do know things which would make it go back down. Regardless, we should still be pressing ahead." But yeah, that's his question. Does this mean a criminal justice reformance is harder and has it even kind of stopped making any headway?

Jennifer (27:24) 

I know that a lot of people, especially early in the pandemic when we saw homicide first start to rise, a lot of people in policy circles were really freaking out about this; that this was going to be the end of the criminal justice reform movement. The idea being that a lot of people are very open to the idea of reform when they feel safe, but as soon as they feel like their personal safety is threatened or the safety of their kids or their friends is threatened, then forget it, lock everybody up. So I think this just requires us to be more deliberate in our... It's going to be a harder conversation and already it is a somewhat harder conversation. Something that I have noticed is that often you see almost like the right left divide almost becomes like the left, only wants to focus on the fact that the criminal justice system can be very unfair and inefficient and racially biased. But they don't want to acknowledge that there's a real public safety problem.

And then the right is only focused on rising homicide rates and doesn't want to acknowledge the inefficiencies and inequities in the system. And obviously both of these things are problems. We can fix the system and make it more efficient and more fair and lower crime and make people safer. And it seems like we have to do both of those things. But it has been interesting, I think, the political dialogue around whether we acknowledge that crime is increasing has seemed a bit contentious. I think part of it is because there are some folks that worry that if we acknowledge that homicide is increasing and that there are cities that are less safe now than they were four or five years ago, then somehow people won't be supportive of reform anymore. And I, in my personal view is you have to acknowledge the facts on the ground in order to have a productive conversation.

Ben (29:29) 

Yeah. I call it the kind of walk and chew gum problem. You definitely can do both. Maybe not that exactly, maybe it's a little bit harder than walking and chewing gum. Actually, we have the same arguments here in the UK on our health system where it's obvious if you don't really tackle it from a left right perspective, that it actually needs more money and more investment and it needs to be more efficient. Those two things are actually both true. It's really inefficient, and I guess generally the right leaning would say, "Oh, but we've got to make it more efficient. We are wasting money," and all of that, which is true. But it also needs capital investment and needs investment in human capital of this. So it needs more money as well. You can walk into government. You need to do both if you're going to do that. And actually probably most things are like that. So maybe thinking about it...

Jennifer (30:21):

One other quick thought on that. One thing that I do hope that we've learned from this experience in the nineties where we saw a big uptick in violent crime then, and we responded by putting everyone in prison for a really long time and just basically-- I mean, mass incarceration was our response. We've learned since that that's not a cost effective policy. So jail and prison can keep actively dangerous people off the streets for a certain period of time. That is genuinely helpful. But it has zero or close to zero deterrent effect on long-term behavior. People do not respond to the knowledge of a long sentence. They respond to the probability of getting caught.

This is all the same. We've learned a lot since the nineties about what works and what doesn't. And so my hope is genuinely that this time around if-- I mean, already it seems like crime rates are leveling off. So hopefully we don't get into this kind of situation. But if crime were to continue to rise and it continues to be a major public policy issue, my genuine hope is that we will respond smarter this time and actually use everything we learned over the past several decades to do this better and achieve more public safety in a smarter, more cost effective, more fair way.

Ben (31:48):

I'm hopeful. I'm always a kind of cautiously optimistic person though on a lot of these things.

Jennifer (31:54):

Same.

Ben (31:54):

Maybe thinking about some of these other win-wins or the intersectionality. I mentioned healthcare because I think it was in a couple of your review articles that access to healthcare helps reduce crime. And actually, I do a lot of work in health and this is one of the things I think people really miss, is this intersection. There's a lot of intersection with social care and all of these other things. But I'm saying, "Well, yes, access to healthcare will help life expectancy if you want your health KPIs." But you get a lot of these other side benefits. And the one particularly in poor areas, and particularly now with the mental health one on crime seems to be absolutely massive. And particularly if you then try and put it in dollar terms, you're getting huge returns to the system, which don't go into the standard health economic ones because they're not interdisciplinary enough, which is one of my real bug bears around how specialized most economists have gone.

I said, "Well, you make this health economy argument, but you've missed two or three of the really big wins, which is if your crime goes down and then you try and put a value on that, you've just blown out 10 times.” So I was wondering how strong do you think that access to healthcare element still is. Maybe talk about that. And the three other ones if you wanted to comment on them as well, which I thought were really interesting, were improving neighborhoods because people like nice streets, it tends to be a win for the community, seems to also reduce crime; a little bit more niche, but these ones on summer jobs. And then also for certain people are kind of essentially cognitive behavior therapy. But therapy for people who could spiral downwards. Again, relatively cheap interventions seems to be win-wins. Be interested in how you've reflect on that sort of literature and that intersectionality of these-- I guess these are kind of non-police other win-win ways of reducing crime, which kind of seem to work. And I kind of feel that there would be, or there is support for. Am I right here?

Jennifer (33:50):

Yeah. I spend a lot of my time talking about interventions like this when I talk to policymakers. Basically when I get invited to talk now, it's all about how to reduce violent crime. I'll start with the criminal justice interventions because we do know some stuff works; like putting more police on the streets works, putting cameras everywhere works, stuff like that. But if for whatever reason you want to know about what else you could do, there's a lot of other stuff you could do. So healthcare is, I think the one I've become most bullish on. There are a lot of really nice studies now showing that increasing access to healthcare through programs like Medicaid which is available to lower income residents in the US. So there've been recent expansions that have especially expanded Medicaid to low income childless men which is exactly the group that you might worry would be engaged in crime. And those all show big reductions in crime rates.

The exact channel is still a little up in the air. It could be an increase in mental health care, it could be an increase in substance use treatment. Of course, those things can heavily interact and intersect. It could be just the reduction in financial stress that might lead to more drinking or something like that. So people are still working out the exact mechanisms. But there's this one amazing paper by Alicia Yakima where she has data from South Carolina on who's on Medicaid as kids, and then sees basically if you have Medicaid as a kid, it's just much easier to be on Medicaid as a child. Then once you're an adult-- They kick off most people at age 19.

So she can see what happens to everyone at age 19. She just sees like at age 19 for young men that are kicked off, suddenly just a huge increase in incarceration rates; almost immediate, you can see it in the graph. They're all just locked up. And that is driven entirely or almost entirely by young men who were having medication for mental health issues covered by Medicaid. And so basically they lose that medication and almost immediately are locked up in our criminal justice system. And it's just like, "God, what a waste of money, of time, of..." This is such a burden on these kids. It's just such a cheap intervention when you think about it that way; healthcare. So there's more and more evidence coming out like that.

I talk to some people sometimes and they're like, "Yeah, I mean, anyone who looks at this system and has any interaction with this population knows immediately healthcare matters.” I always like to say, "But now we have the research so it's useful. Not everything that seems obvious to us when we look at it firsthand actually is borne out by the data; this one is.” So there's more and more really strong causal evidence that increasing access to healthcare reduces crime and reduces recidivism. I kind of put cognitive behavioral therapy in a similar bucket, although there it's obviously CBT can be useful for people that don't have what we traditionally diagnose as a mental illness. But the story around CBT and the kinds of programs that have been tested are more around like changing the scripts that we all have in our heads about how social interactions will go. And that is perhaps more beneficial to people who live in neighborhoods where you have to interact with some people in some way and then sort of change your interactions to interact with other people.

So for those of us that live in safe, affluent neighborhoods, you sort of defer to authority or if anybody gives you a hard time; if someone mugging you on the street, you hand over your wallet and know that there'll be a cop around the corner. If a cop pushes back on you and tells you to do what you're told, you do it. But kids that are growing up in higher crime neighborhoods might learn over time that they need to stand up for themselves and push back if they're threatened in the neighborhood because that's the only way they survive. And then of course, if they do the same thing with a cop, there could be terrible consequences. So when you have to do that sort of script shifting across different places that is just much more cognitively taxing. So these CBT programs basically push everybody to just slow down and think about what is the story you have in your mind about how this interaction is going to go and just make a more deliberate choice about how you're going to respond, and don't respond with your immediate impulse.

I hope I didn't butcher that program too much to anyone who's listening who studies these programs intensely, but I think that's the general idea. And they've been shown to be really effective when they're introduced with kids in schools, high risk kids in schools, and even tested in juvenile detention centers. You see big reductions in recidivism going forward. I often joke with folks that if we just gave everyone a therapist-- I mean, truly everyone a therapist, I think we'd all be better off and we'd see a lot of social returns. But especially if you give people that are at high risk for mental illness and for emotional trauma that comes from violence and living in rough neighborhoods and everything else, if we gave all of them a therapist, I think we'd see big reductions in crime and other sorts of social ills. So it seems like a really good investment.

Ben (39:17):

That's one of my left field ideas for AI therapy. I think there's some evidence that AI chatbots are okay. They're not as good as an in-person therapist, but they're better than nothing; at least for overall therapy. So obviously not proved here. So if they become very cheap and everyone has them on their phone and there's a little bit of a push, then maybe they can give us the positive stories that we need. The last one on that was the improving neighborhoods. Is that strong on built in environment. Are you positive on that one as well?

Jennifer (39:47):

Yeah. I would say in general there's some suggestive evidence on this. I think, again, the channel here is a little unclear. I saw something recently saying that maybe part of it could be that police spend more time in nicer neighborhoods there would be like green vacant lots or whatever. So it's actually the police presence that's doing this. So I think we still need to figure this out. But there are some neat studies. There's one where they looked at neighborhoods where there was some sort of moth or something that was eating all the leaves on the trees and killing trees of certain types, but not other types. So it was like this random shock to where the trees were. They saw what happened to violent crime in the places where these moths susceptible trees to the ones that that were okay. And they saw crime go up in the places where the trees all died.

So it's sort of like these neat where it's like, "Oh, so it could be the tree." Even there, we also have evidence that increasing temperatures increases crime or increasing pollution or reducing air quality seems to increase crime. So it could be trees also help with air quality. They help cool everything down, there's shade. So is it really the greenery or is it the shade or is it the better air quality or is it the cops who like to hang out where there are nice trees? Who knows? But there is a good amount of evidence on this. And there are a whole bunch of criminologists in particular that are really interested in just the connection between our built environment and how we behave in that environment. So there are other psychological mechanisms that are potentially at play and that we just behave better in certain types of places than others.

Ben (41:32):

Yeah. There definitely does seem to be a link with a built environment. But as you say, it's complex and actually may not even hold from one location to another location, even with the same physical makeup. The one I see here in London, although I think they might have run out money because I don't see it as much so often, but we have fly tip zones. Do you call them the same thing in the states where people dump their rubbish and they're not meant to dump them there? You occasionally get these hotspot places where people just dump all of their old fridges or toys and stuff. They are often ugly corners of the street anyway. Local councils did efforts like they cleared them up and then they dumped them again. But they cleared them up and then they planted plants and made it look really nice. Then actually people decided not to fly tip there anymore because it's kind of like, "Oh, we wouldn't want to spoil it." I think maybe it just moved it to other places so I'm not sure the point there. But at least those were no longer fly tip.

Jennifer (42:38):

But it improved that neighborhood.

Ben (42:39):

And maybe I think the idea at least in that then if it's then you have to go so far enough that you can't fly tip on your street, then you do what you're meant to do, which is just call up the local council and then they come and take it away.

Jennifer (42:52):

It increases the cost of the alternative.

Ben (42:55):

Yeah. That's exactly right.

Jennifer (42:58):

I mean, this is in some ways a bit of a theory behind broken windows which was really hot in the nineties, which is this idea that if we really track down on low level offenses and quality of life offenses, like littering or dumping your trash on a street corner or breaking windows which is where the name came from, then everybody notices that this neighborhood is now being cared for and people are paying attention to it. And so then they will stop behaving in criminal ways. So this again, was hot in the nineties. A lot of people thought that implementation of these broken window strategies and policing was a reason that crime declined so much in the nineties.

We now know it doesn't seem to have contributed, doesn't seem to explain why crimes fell so much in places like New York where it was especially important. But this basic idea that if you take care of a place and lots of people seem to be paying attention and there's like a neighborhood watch or something like that, then it does seem intuitive that people are going to be less likely to do things that would disrupt that place. So might generally be deterred from committing crime in that place if they feel like people are paying attention to it. The question is how to actually implement that? It doesn't mean we need to arrest everybody and throw them in prison for 10 years. It can mean lower level consequences or it could mean plants and flowers can deter people from dumping their trash there. There might be other ways. But I think the basic point is that sometimes it's low level investments can have big payoffs.

Ben (44:50):

Yeah. And I think, like you said, it's probably an intersection of a few things. So community policing or being aware in your community and your neighborhood being good. More aware, police more likely to be walking on the beat and all of those other types of things. I was interested in your views on what alternatives to jail else we could do. So like you said, there's kind of just letting people off for minor misdemeanors, but there seemed to be some evidence that some of these other alternatives to jail also work in a kind of win-win because jail is really expensive.

Jennifer (45:25):

Yeah. So I've become really interested in electronic monitoring as an alternative to especially short jail and prison sentences. So in the US we often use electronic monitoring as a supplement to community supervision. So if we're going to let you out pretrial or we're going to put you on probation, then maybe we'll add electronic monitoring so we can keep track of where you are and we'll get notified if you leave your house and you're not supposed to. Increasing supervision doesn't seem as helpful. But as an alternative to locking you up, we now have a lot of evidence from other countries that do this routinely that it's really beneficial and seems to reduce recidivism quite dramatically. Some of that seems to come from reducing the destruction that jail and prison can have. You'll see you're able to continue working, continue spending time with your family, maybe going to school. It also could reduce the negative effects you might get from spending time in prison with other people that are criminally active. So those sort of negative peer effects can be reduced if we just put you on electronic monitoring and send you home. 

There's one recent paper that just came out from someone who's on the economics job market this year; Roman Rivera. And he's finding similar effects in the US, that using electronic monitoring as a substitute for incarceration seems beneficial. So that's the first evidence we have in the US which has been a major hold up with policymakers because everyone's like, "Well, we're not Australia, we're not the UK, we're not France. So it's totally different here." It turns out it's not totally different; it works the same way here. So that's really promising and it feels to me for various reasons we might not want to just let everyone that we're currently sending to prison out. We might not even want to let everyone who's on the margin of going to prison out. But it could at least make it more politically palatable to reduce our reliance on incarceration and has the big benefit of actually dramatically reducing recidivism. And like we were talking about earlier, it's also cheaper than sending someone to prison. So there are a variety of reasons that this could shift us in a better direction and is something that I'm paying a lot of attention to this research space.

Ben (47:47):

Yeah. So that's all of your work on sort of technology and justice and crime. Those arguments from policymakers actually really irritate me sometimes because sometimes it's clear. So if you've got differences in countries like America with guns and Europe, okay, there might be differences. But when the causal model you have is the same kind of human and the same kind of thing, it's kind of really silly to do that. It's like you are putting it back. There's a reason that there's these similarities. And if you ask the experts, they will generally agree when they think there's read through or when they think there's not read through. And they're also generally right as well, that's the other thing. If you do the research and say, 'You are to guess in advance, do you think this will replicate or not replicate a couple of these things?" Actually, economists generally get it right, which is also kind of unsurprising there.

Maybe we should touch a little bit on also what doesn't work. I recall you said earlier cameras can work, but I think you were referring to kind of CCTV cameras and cameras on the street which we have a lot of in London. Actually, we might have reached over saturation because once you have one on every block-- maybe three on every block is not going to do you any further, but one on a block versus not. Whereas I think your work suggests or we've looked at that body cameras, long story short, kind of probably don't work and are also quite expensive. Another thing which maybe doesn't work or is a little bit more mixed-- and this is kind of on the flip side, is when you're applying for jobs and you're kind of now when you're resume blind. So you don't have to say things about...

Jennifer (49:23):

Ban the box policies.

Ben (49:24):

Yeah. Ban the box. I forgot the buzzy buzzword on that which is kind of interesting because you would've thought that, "Oh, maybe not." But the second order effects which might be the same with the body cams, which I thought was quite interesting. And maybe in the third one which also seems to be a little bit more contested was on alternative drug use for naloxone and methadone and things like that; whether there's any displacement on that. So I'd be interested in your views on some of the things which actually may be second order when we've looked at them in more controlled cases, may not seem to have worked how we thought or either contested or neutral or you kind of think actually they may not be worth the money. Those were the three I picked up on in in your work.

Jennifer (50:13):

Yeah. So I'm talking a lot here about what works, what have we learned. The drum I usually am beating with policymakers is these are really hard problems to solve, most things we try will not work. We should aim to fail fast; not to fail because the reality is most things we try are not going to have the effects that we are hoping for. And often they're going to backfire just because these are such complex situations and human behavior is really hard to predict a lot of the time. So these are great examples, those sorts of situations. So cameras in general like CCTV cameras seem to be really beneficial because they're increasing the probability that you get caught. And again, as I mentioned earlier, we have a lot of evidence now that increasing the probability you get caught has a much bigger deterrent effect than increasing the punishment.

So they have a privacy cost and so that needs to be our public conversation; like different types of surveillance technologies are going to have bigger benefits than others and also bigger costs than others. So weighing those things becomes the conversation we should have. But we should know that in general they're going to be more effective than putting people in prison for 20 years. The body-worn camera literature is actually really interesting because there were all of these really nice randomized controlled trials, which is really unusual in the space. We had all these RCTs across different countries randomly give body-worn cameras to some police officers and not others. The working hypothesis there was that if the officer is wearing the camera and knows their behavior is observed, they're going to be less likely to behave badly which could include knowingly escalating a situation to an arrest or use of force when it didn't need it.

So to the extent that police are behaving badly in a conscious way rather than just because they're afraid for their safety, for instance, then body-worn cameras could reduce that bad behavior. But across all these RCTs, basically we found very little evidence of any sort of benefit or any behavior change at all. So that was really discouraging. All these places kept their body-worn cameras presumably because they weren't actually counting on that behavior change as the main goal of the policy. They also just wanted transparency and they wanted to be able to hold people accountable when something bad happened, even if it was somewhat rare. So that was sort of the conversation for a long time.

But we've since had a couple studies come out that have made the point to the extent that these body-worn cameras-- that if some officers are wearing them and others aren't, it could still have a broader impact on the community in a way that within community randomized control trial is not going to pick up. So part of this is just spillovers across officers or across ships depending on how these things are randomized. The studies already always talked about this, but it was hard to tell exactly how big the spillovers would be. Anyway, so there have been a couple of other studies that have just compared communities that adopted body-worn cameras with those that didn't and compared trends over time. And those seem to find at least suggestive evidence of big reductions in police use of force in the places that adopted body-worn cameras, which suggests that these community-wide effects could be happening. This is still a very active research area, active conversation, but it feels like maybe there's something there that we didn't see with the RCTs, which is just fascinating from a research perspective because we usually love RCTs. It's like the best gold standard, at least.

Ben (53:49):

No design is perfect.

Jennifer (53:50):

No design is perfect. You learn different things from different designs.

Ben (53:53):

Yeah. And they can’t capture everything, right?

Jennifer (53:55):

Exactly. Yeah.

Ben (53:57):

Interesting. So the jury is still out?

Jennifer (53:59):

The jury is suddenly still out. It's back out. The jury was in for a while and now it's back out because of this new evidence which is just-- science man, it's great. On Ban the Box, yes, this is an area I've studied directly. So basically the idea behind Ban the Box is that we are worried that a lot of people with criminal records cannot get their foot in the door to get a job interview or to be considered for a job because employers see-- Like, they'll have a box on an application saying, "Have you ever been convicted of a crime?" If you check yes on that box, anecdotally, we know that employers will often just throw those applications out. We also just have a bunch of really nice experimental research where people send fictitious resumes from different people with the box checked or not. We know employers discriminate against people with criminal records. So there's good evidence on that.

The hope was that we could stop that discrimination by just removing that box. So banning the box. But this of course prompts economists like me to say, "Well, then what will they do next without the box? Why is it that they were discriminating in the first place?" And if they're worried about, for whatever reason, an employer doesn't want to hire someone with a criminal record, removing that information doesn't remove that worry for whatever reason. They still don't want to hire someone with a criminal record; you haven't changed that. Now, you've just left them in a position where they have to guess who has a criminal record. And the United States, there are huge racial disparities and who has a criminal record. So it would be statistically accurate for them to assume that young black men are more likely to have a recent conviction they might be worried about than young white men, for instance.

So this kind of policy could increase racial disparities in employment rather than reduce them. And indeed, we had previous studies showing that when employers first got access to criminal records, like in the nineties when the internet became more widely available, we actually saw reductions in racial disparities; reduced racial disparities in employment because it seems employers were racially discriminating before because they were worried about hiring people with criminal records and not because of race itself. So adding the box on actually was a policy limine at the time to racial discrimination, even if that wasn't exactly the design. But it turned out that was a good intervention.

So now we removed the box and we actually see big increases-- this is what I find in my study-- big increases in racial disparities in employment for young black men. A bunch of other studies have shown these policies don't even help people with criminal records get jobs. So we see that young black men who don't have a record, their employment is falling because they can't signal they don't have a record upfront. But the people with a record aren't even getting the second chance that the policy had hoped it would get, because eventually the employer can do a background check and reject their application then. So this policy has just been a complete disaster and it's still very popular. I think there are a lot of advocates who view it as just like a step in a broader movement which can be-- I am sympathetic to that concern and the broader desire to want to increase second chances for people with criminal records and want to improve reintegration-- something I study a lot myself. But this particular policy is not working and actually seems to be making things worse. So, yes, there are many other things we could be doing instead and so I have that conversation a lot with folks. But yeah, great example of how just well-intended policies do not always have good effects. So we need to be making sure we're meeting the goals that we set out to meet.

And then on the drug front, I have a study on Naloxone and opioids which is easily the most controversial paper written which is sort of amazing. I write a lot on race and crime. It turns out this paper on, I mean, essentially noting that depending on whether you want to use the words moral hazard or risk compensation. If you make something safer to do, people are going to do more of it. So basically Naloxone is this amazing drug that basically reverses overdoses of opioids. So it saves your life if you're in the midst of an opioid overdose. And there are these policies that just dramatically broadened access to this. So just sort of you could easily get it from pharmacies, anywhere without a prescription. A lot of people thought like, "Great, this is just an obvious clear wind. You're just going to save a lot of lives. This could dramatically change the opioid epidemic." An economist might like me, looks at it and says, "Well, probably we'll have a lot of benefits, but could also increase risky opioid use. If we're going to make something safer, people are going to do more of it. So how much of that benefit does it undo?" 

And so I have this paper with Anita Mukherjee where it seems to undo a lot of the benefit. We see no net decline in deaths due to opioid overdoses even though people go to the ER more with overdoses. So it seems like there's some which together suggest there's more actual risky use. In general, our takeaway is even if we obviously want to save people's lives in the moment, but we have to recognize that this policy does not seem to be a slam dunk on its own, and we have to be investing in other things like treatment or other solutions. So, yeah, that paper has generated a lot of anger from the public health community. It occasionally comes up for one reason or another, and the harm reduction folks become irate and send me lots of angry emails. But look, I am also interested in saving lives. And I think the question here is just what the best way is to do it.

Ben (01:00:01):

Yeah. The answer is it's complicated. I spent a few minutes on the internet. I was surprised at the strength of anger. I mean, they were really rude. It's kind of really interesting because-- I don't know why I wouldn't expect it in academia. I kind of expend it to my day job is all about investments and markets and I kind of expect us to be rude with one another. We're trying to make money and obviously we think some person is wrong because otherwise we wouldn't do that. But I kind of thought with people in good faith trying to do something complicated to get to the right answer. It's complicated, humans are humans. So there was that. But I guess that the strength of pushback probably continues to surprise you today then.

Jennifer (01:00:50):

Totally. It's just amazing. Similarly, I'm an economist and academic economists are not known for being warm and fuzzy and nice. You're all used to heavy pushback, very sharp criticism in academic seminars, but all in sort of an intellectual-- It's all part of an intellectual exchange where we're trying to make the ideas better. We're pushing each other on the ideas and the data. And if you have a disagreement, you go to the data with it. It's like, "Well, let's see. Let's test these different hypotheses." The pushback that I have gotten from the public health community, Lisa's corner of it, over the years seems much more in line with the idea that there is a consensus on what the "right answer" is on a policy topic like this. And we found the "wrong answer." And so that goes against what they've been advocating for in policy circles. So they need to come after the paper and sort of tear us down as opposed to talking about like, "What could be wrong with their research methods or whatever." Actually talking about how to make the paper better.

So I've had a lot of very frustrating conversations with people since about the sort of detrimental effect that that kind of attitude within a scientific community has on the research that's done. It's not surprising to me that you haven't seen other papers reach similar conclusions from within that community because people get the very clear message that they will be ostracized and torn apart publicly if they even suggest if they ever find a result like this. I've got a pretty thick skin and I'm willing to work on controversial topics, but it definitely deterred me from ever writing a paper on health myself again. And I'm sure it deters others from working in this area which I think is what a lot of these folks want. I know a lot of wonderful people who work in public health and are public health researchers, and they're doing excellent work. But it does seem to be a corner of that research community where there are more advocates than scholars. And I think that is the segment of the community that comes after me on the internet every year or so. It's just sort of amazing. I don't take it personally anymore. It's just clear that this is just something else that's going on, but it's fascinating to watch and I'm curious to see how it evolves over time.

Ben (01:03:15):

Sure. Yeah, I saw a glance of it and obviously don't know the ins and outs. But it strikes me that this is a symptom of wider challenge, which I see across many sectors. I think we see it within economics. I guess also with a kind of more recent me too phenomena or the fact that I have a lot of women friends and also younger female friends who find economics are kind of unfriendly. They also find this about law, investment banking, quite a lot of other sectors. So I mean, we can brush quite a lot of that. But also amongst the social sciences-- And I had this conversation with economists here in the UK, with Diane Coyle, who makes similar things. It kind of shapes also the research agenda, also how the diversity of thought, and that there seemed to be so many reports in the last couple of years. Again, this is just skimming the internet, occasionally seeing things on social media or whatever was sort of ‘me too’ and these allegations, which seem to sit quite awkwardly. I guess from your viewpoint being a woman in economics, but actually soon to leave academia, what your viewpoint is. Obviously, everyone just got their own personal story. This is not necessarily N equals 10,000, but N equals one. But I seem to get a lot of N equals one of women who haven't necessarily been super happy with the state of where economics is in terms of diversity of people or thought and things like that. I'd be interested if you had any thoughts on that.

Jennifer (01:04:51):

Yeah. So this is a topic I have thought a lot about over my career. In general, I love being an economist. I love the toolkit and the insights that economics brings to important social problems like criminal justice policy and reform. Being in academic and working at a university as an economist can be different. So then we're talking about working within certain industry. So first, I mean, I think there has been a lot of talk for a long time about just the way that women and various racial groups are just tremendously underrepresented in economics. I think in general we hover like full professors-- It's around like 20% of full professors are women and somewhere around a third of all assistant professors are women. Maybe half is too lofty a goal, but it'd be nicer to get closer to half.

It seems to be worse than in other STEM field. So a lot of times people are like, "Oh, well, it's just such a math-based profession." Of course for me, if you are women, that doesn't seem to be the holdup. It's something different about economics that does seem to make it worse. There've been a bunch of really nice empirical studies recently in recent years showing that women do seem to face biases in publication in terms of being burned with more service within the academy and all this other kind of stuff that reduces productivity or parent productivity, even if our actual intellectual contributions are the same. And like you, I've heard a lot of N equals one stories from lots of friends at this point who just have faced lots of what feels like one-off challenges and biases and lack of recognition of their contributions and their professional value in the workplace. They're just really frustrated. I think at this point I am solidly mid-career, I've been a professor for 11 years.

Most women I know who are sort of in my general cohort are really struggling and really starting to look for other jobs. But they're trying to figure out what those alternatives are. And I think a lot of us, if you spend all this time in the academy, you work so hard for tenure, it feels crazy to give tenure up. I personally got to a point where both and just experiences in the universities, but also in this broader economy too conversation that I was heavily part of in this past fall. It just kind of got to a point where it seemed clear to me that while tenure guarantees you a job, it doesn't guarantee you a good job. It just made me more open to the possibility of giving tenure up and considering something else. Then once I did that, all of a sudden, there are really amazing opportunities that are available. So yeah, I'm moving to Arnold Ventures. I'll be running their criminal justice program. So excited, it's such a good fit for me. I'm going to have an opportunity to really influence research and policy in this area in ways that I've already been trying to do, but it will just be so much easier to do from this other platform. So very excited. It's definitely a move for the best. Now, I find myself, when I talk to women who are like, "I don't know, this is just such a terrible job. I don't know..." I've stopped just commiserating and now I full on encourage them to consider other options. So I think they are out there.

There are all kinds of people that are really working hard within economics to make it better. In general, I still strongly encourage women to go and major in economics, even get a PhD in economics if they love research. I think it's an amazing toolkit and an amazing source of expertise and it allows you to have an influence and interesting insights on wherever you want to apply it. But I'm also very glad that there are now lots of different ways that you can apply that expertise. You don't have to become a professor. You can work in tech, you can work in industry. Everybody loves economists at this point and sees their value. So, yeah, that's the not so short version of my current thinking on economics and the profession.

Ben (01:09:44):

Sounds very clear. I'm going to ask about Arnold Ventures at the end. I guess I had one question then on this. What would you propose to the economics profession? Are there any policies or ideas or culture change? I will link it back to policing, because why not? What I've seen here-- this is more of a London metropolitan police issue which I've seen. But it does strike me that particularly for London which is quite a multicultural city, having a police force which reflects your city really seems to help. So more women and more minorities in the police. And again, this for me is anecdotal. But I've seen some little bit in literature and kind of because it causely makes sense that if you're part of the black community and you see black police officers, you kind of feel like you're going to be listening to people in your community and you'll be listened to.

Jennifer (01:10:35):

They'll take you seriously.

Ben (01:10:35):

Take you seriously either if you have a complaint or if you feel like you're being picked on. "Oh, it's not being picked on. Hey bro, this is what's happening." You'll feel you are listened to. And again, all sorts of good reasons for having women. So I'm actually quite interestingly a fan of kind of-- I don't know whether you'd necessarily get there by quotas, but diversity In police force, diversity to a lot of organizations seem to have these system-wide benefits, which obviously I could see happening in economics. I wonder whether you think that's true of policing and then what would you do for economics?

Jennifer (01:11:10):

Yes, there's definitely evidence that that's the case in policing; that having more women officers increases reporting of gender-based violence, reduces homicides due to domestic violence, for instance. Increasing the hiring of black police officers seems to reduce victimization of black people in the community. So basically the story you laid out is supported by the data, at least studies we have so far. And I think this is a place where seeing if this continues to hold up in current day-- a lot of this was based on litigation in the seventies and eighties. So we're a different baseline, but it’s definitely something I think we have good reason to believe would be a really good fix for policing or at least contribute moving in the right direction.

I think the same holds in economics. I mean, we know that there are important role model effects. It's hard to be what you can't see; that whole thing. So having more senior women encourages more women to come into the profession. There also is evidence that people appreciate having mentors who are like them in some way. And you could imagine women being able to be better mentors to women because they just understand their experience. So there have been interesting papers suggesting that one way affirmative action can be beneficial in the long run is that it brings in at least a first cohort of senior people from this minority group, and then that has long-term benefits. Even if the affirmative action stops, they're able to mentor people further down. So then it can kind of continue the flow of people from that minority in. So you could imagine there being big benefits there.

I mean, economists in general hate quotas, but there have been all these interesting studies showing that quotas, at least in the political space requiring a certain number of women to be elected from certain places. They all seem to have huge benefits and it seems to be mostly crowding out more of mediocre male candidates. It changes the way that residents think about women in leadership. Right now, they are actually seeing a woman leader and so they're less likely to think women can't lead or they're maybe more able to be better judges of women candidates because they now have a better-- they've seen some examples and are able to… Maybe they didn't know how to tell who would be a good woman leader from a bad woman leader before, but now they can; something like that.

So you can imagine similar stories in economics. As someone who has organized seminar series in various places that I've worked, I have definitely in the past thought about I want to bring in more junior candidates from underrepresented groups because I want my senior white male colleagues to think about these people as economists; these are examples of what economists look like. That has been a very active type of intervention I've made in the past. I think all of this stuff matters. And then as you said before, I mean, there's a reason. This isn't all just to make us feel good. I honestly think the science will be better if we have the best and brightest people at the table; and you can't do that if a huge segment of the population either feels excluded or is actively excluded and discriminated against.

But then also it changes the research questions that we're asking, if we don't have people with all kinds of life experiences in the room. We will come up with better policy solutions, we'll come up with better hypotheses if we have a variety of perspectives there. So this is good for science ultimately; to have a wider variety of people there and to break down the different silos that are separating us currently. So a lot of this is motivated just by like, "I'm just tired of injustice. It's just really frustrating to see my friends not valued by their colleagues.” But part of it is, honestly, I think this is just really important for science. And really, I think, economics is often offering some of the best policy solutions in really important spaces and we have seats at the table a lot more than people from other disciplines. So if we're falling down on the job because of stuff like this, feels to me like a relatively easy fix.

Ben (01:15:40):

Yeah. Better for the science, better for humanity. I wasn't aware of that political leadership work you mentioned there, but makes a lot of sense to me. Great. So I thought I'll finish with a little overrated, underrated, and then future questions. So if you're up for it, you can just go overrated, underrated, or you can pass or neutral. I'll start with Texas.

Jennifer (01:16:07):

Totally underrated.

Ben (01:16:09):

Of course. It's got to be.

Jennifer (01:16:10):

I love Texas so much. I grew up in...

Ben (01:16:12):

You're Texas born?

Jennifer (01:16:12):

I grew up in Massachusetts. I grew up in Massachusetts and have slowly been moving farther south. I like warm weather unfriendly people, and Texas is great on both counts. So I love it here.

Ben (01:16:24):

What's most misunderstood about Texas?

Jennifer (01:16:29):

I think on the coasts of the US it's very-- it's like people really enjoy making fun of Texas as being this crazy outlier where everyone is just politically super hard right and it's just not true. It's like any state. There are rural areas and urban areas. Also, it's a huge state. There are like five major cities in this state. So it's just a really diverse place; I think that's the biggest thing.

Ben (01:17:06):

Yeah, it's like Nashville and Tennessee also. People think of it as one thing, but it's more than that. So overrated or underrated, universal basic income.

Jennifer (01:17:21):

I think probably overrated. Feels like it's getting a lot of hype. It's not a policy that I am super excited about as like a panacea.

Ben (01:17:37):

Yeah, that seems fair to me. Overrated or underrated, DNA databases.

Jennifer (01:17:47):

Underrated. I have research showing that they have huge deterrent effect. As I was saying before, there are going to be some surveillance technologies where there are really big privacy costs and others where there's less. This one I think people think because it's DNA, it's really invasive, but it's not what you're imagining. They're not analyzing your whole genome. It's really just like a higher tech fingerprint. So low privacy costs.

Ben (01:18:13):

I think they're overblown. In fact, so on my health work, I think this overall privacy concern is holding back all of our healthcare data analytics and we don't realize we give so much more information to Facebook and Google and Apple anyway. We've ticked the bot and given it to them. Actually, because of the rules and regulations we've held onto healthcare, but actually the healthcare would really help us out. And the data to Google only allows them to sell you more widgets and shoes. So we've got it backwards. It is more targeted shoes, so it's the shoes that you want. So there is maybe a little bit of consumer welfare, but it's not the same in health. We've definitely got that the wrong way around. But anyway. And it's through the same channel like the CCTV is because your probability of getting caught, it puts you off

Jennifer (01:19:05):

Exactly. Yep.

Ben (01:19:06):

Overrated or underrated, daylight savings time.

Jennifer (01:19:12):

Written in all my papers here. I think underrated. So from a crime perspective, it turns out shifting daylight to the evening, which is what daylight saving time does, reduces crime because people are more likely to get caught when the sun is out.

Ben (01:19:27):

Yeah. So I only learned that this week. So a lot of people don't seem to like daylight savings time, but I hadn't heard this angle. So actually when you put into the crime effects, it actually seemed to be more robust, a slightly bigger number than I thought.

Jennifer (01:19:44):

Yeah. People often think of daylight saving time as the switching, but it's actually like daylight saving time is the summer when sunlight is in the evening and then standard time is when the sunlight is in the morning. We could eliminate the switching and just keep daylight saving time year round. And so people occasionally propose this in Congress. So you could imagine like three different fights. You want to have, “Do you want the switching?” And all the costs that people worry about seem to come from the switching of the time change. So you don't get as much sleep and there are more heart attacks and car accidents and all this stuff because people are thrown off. But we could do away with the switching and just keep daylight saving time in terms of when we want our daylight.

Ben (01:20:27):

So you get one step up. And that's because more crime happens at night, is that right? Or in darkness.

Jennifer (01:20:33):

More crime happens at night, right.

Ben (01:20:33):

It does happen at night. Is it actually easier to do crime at night or is it just perceived to be from the stories we tell ourselves so that's why we do it.

Jennifer (01:20:45):

Yeah. I mean, I usually quip that criminals aren't morning people. They're not getting up to rob you at 6:00 AM or 7:00 AM on your commute. They're only going to rob you at 5:00 PM on your commute. There might actually be more people out and about at 5:00 PM too and you might be more likely to walk home than walk to work. But for whatever reason, there's a lot more crime in the evening. And so shifting the daylight then does actually reduce crime overall, it doesn't just get shifted to the morning.

Ben (01:21:16):

Sure. And the last one on this, the links between I guess climate and crime, although I guess this might be more specifically heat.

Jennifer (01:21:29):

So, yes, we have all of this evidence that higher temperatures increase crime. I have work with Jonathan Colmer showing that that also intersects with gun laws, going back to the first part of our conversation. And gun laws are very effective at mitigating this crime intersection. But the all matters because climate change is happening and our temperatures are rising so we should expect violent crime to rise too, and all of these anti-violence programs are only going to be more important as temperatures go up more.

Ben (01:22:10):

That makes a lot of sense. Great. So last couple of questions. How about your future work and ventures? So what's on your agenda for your work with Arnold Ventures and what's all of the other things you're getting up to?

Jennifer (01:22:25):

That's the main one right now. So yes, I'll be the executive vice president for criminal justice at Arnold Ventures starting in July, which means I'm overseeing their research and advocacy funding and strategies. In general, what is really attractive to me about Arnold and what has drawn me to them and why I think I'm a good match, is they are very focused on rigorous research informing policy. And not only trying to support and generate more of that really high quality causal research, but also getting it into the hands of policymakers and making sure that it's having an impact. So yeah, I'll be doing a lot of that. I will keep the podcast going. I've realized in the close conversations that a lot of the stuff I have been doing as a professor has mostly been structures to help me keep better tabs on what is going on in the field, which of course is now going to be only more important in this new job. Just knowing who's working on what and what new evidence is out there. So I have this online seminar series; plan to keep that going; a conference every year, keep that going, and the podcast. So that's the big thing.

I'm moving to Houston, I'll be spending a lot of time in DC and New York. A lot of our team is there. Basically, I'm really looking forward to learning from the incredible people that are on the Arnold VJ team. It's just a tremendous group of people who have been working in the criminal justice policy space for a really long time and I'm going to learn a lot; both about the research they've been supporting, but also just these policy areas which hasn't been directly part of my job to be actively engaged in the policy conversation. And so that'll be much more proactively part of what I'm doing. Other than that, I'm working on a book called, The Science of Second Chances. So that will keep going. It's due to go to my editor in December, so I'm hoping to get that in on time That book is going to be on basically what we know from the research on the best ways to intervene at each stage of the criminal justice process to give people a real second chance and break the incarceration cycle. So I'm keeping that going while I am starting to think about this transition out of academia and into the real world.

Ben (01:24:57):

That all sounds great. Got a bit of a great advice. So the last question would be, any life advice for people?

Jennifer (01:25:21):

I think the main thing that I try to remind myself and hopefully will be relevant to other people too, is I think that we are all probably too cautious in terms of taking leaps, making changes, leaving employers, leaving industries, leaving relationships. I think there are a lot of things that we could imagine either like sunk costs; there's a sunk cost fallacy here, which economists love to think about. Like, “I've already invested so much in this thing, so I don't want to leave." But I think also it's just a fear of the unknown. And so we imagine that the alternative is going to be the worst possibility. We know what we have right now, but in practice I actually think the alternative tends to be much better.

So I think a lot of us are too risk averse in terms of being willing to make changes. And most of the time when I talk to friends after they've hemmed and hawed for ages about whether to make some leap, they're so much happier once they do. This is somewhat related. There was a study a couple years ago by Steve Levitt and probably some co-authors looking at-- they randomized people into just like, if they had some major life decision to make, they flipped a coin and then did whatever... They were sufficiently torn that they did whatever the coin told them to do. And the people where the coin randomly told them to make the change and take the leap were actually much happier later. So that confirmed my general hypothesis that I think a lot of us are too risk averse in making major life changes. So when in doubt, I try to take the leap.

Ben (01:27:12):

Take more risk. I had someone who's a master improv actor and there's a theme in improv, which is always say yes. So that's one of the things. Well, on that note, Jennifer Doleac, thank you very much.

Jennifer (01:27:30):

Thank you. Thanks for having me. This was really fun.