The nationwide jump in shootings and homicides early in the pandemic and the rise in other crimes that followed in some places have made crime a hot topic again in the United States. It has been a prominent one for academic research for a while, with economists in particular flocking to the field as a testing ground for research strategies that aim to sift causes from data. To get a sense of how recent findings fit with the national discussion on crime, I talked to Jennifer Doleac, an economist at Texas A&M University who not only studies crime but hosts a podcast on new research, Probable Causation, and has organised the Criminal Justice Expert Panel, which sums up expert opinion on crime questions. This summer, Doleac, who has also written a few columns for Bloomberg Opinion, will become executive vice-president of criminal justice at Arnold Ventures, a leading funder of crime research. Following is a much abridged transcript of our conversation and a list of research papers referred to in it.

⁎⁎⁎

Justin Fox: My overarching question is, what just happened over the past three years with crime in the US, and what can high-quality research tell us about why?

Jennifer Doleac: I’m going to have lots of really unsatisfying answers for you.

JF: That’s the idea.

JD: OK, good. We saw violent crime go way up in the late Eighties and early Nineties and then suddenly start declining. We still don’t really understand why that happened. It’s similar now. We don’t understand why crime has been falling all this time, and we don’t understand why it suddenly was not falling any more, and homicides and shootings started to go up. Those unknowns freak people out and lead everyone to start looking for answers and scapegoats. That’s why you wind up with all of these news stories about progressive prosecutors. That seems like the main scapegoat.

JF: You did a paper about this.

JD: We used data from Suffolk County, Massachusetts, where Boston is located. People who are arrested and charged with low-level, nonviolent misdemeanours have to go to an arraignment hearing. The prosecutor who’s standing at the table on the day that they have their hearing is lenient or harsh or somewhere in the middle. That person decides whether to pursue the charges and press the case forward or just drop it right there.

If they get lucky and get a lenient prosecutor who drops the charges, they go on their way. If they get unlucky, the case continues. The question becomes, what does this do to this person, to their future trajectory? It turns out if they get lucky and their case is dropped, they’re about 50 per cent less likely to come back with a new charge in the coming years. Recidivism seems to fall, which is very much in line with the progressive prosecutor’s view that throwing the book at people for these low-level offences can have what we call criminogenic effect — it can create more crime rather than improving public safety.

We also looked at a reform that happened in Suffolk County when [former District Attorney] Rachel Rollins took office and implemented a presumption of non-prosecution for these types of offences. We see a reduction in recidivism, no change in crime rates. If anything, crime seems to fall a little bit.

I definitely changed my priors. I did this paper with Amanda Agan and Anna Harvey, and was the most sceptical of the three of us about the extent to which this criminogenic effect would cancel out the deterrence effect. I have changed my mind about it. First-time offenders are sort of at a fork in the road. We can either hope it’s enough of a wake-up call that they’ve been arrested and had to come into court, and they’ll change course on their own, or we can pull them into the system. I’ve become a big proponent of erring towards leniency in those sorts of situations.

There has been other work to suggest similar things with nonviolent felony defendants. There’s a whole bunch of work on pre-trial detention and the fact that locking people up pre-trial has a really detrimental, causal effect on their future trajectories. They’re more likely to plead guilty in that initial case but also more likely to reoffend in the future.

So there’s a lot of evidence that the policies that progressive prosecutors seem to be implementing would likely not be harming public safety and certainly not be increasing homicide rates. Amanda, Anna and I have this other paper where we look at when progressive prosecutors were elected across different cities, what happened to crime rates in those cities. It’s noisy, but there is no increase in homicide, for sure.

A lot of these prosecutors were elected during Covid or around that time. There are also a bunch of cities that didn’t elect progressive prosecutors during Covid. That is what allows us to look and see if it’s different — and it’s not different across these different places. But part of what is spurring this conversation is that, “This prosecutor just took office and three months later homicide started going up. It must be because of him.” It’s not. It’s just a coincidence. Homicides started going up everywhere during Covid, so it’s something else, but we don’t really have a good answer for what that something else is, which is what allows the story to continue.

JF: To me a more promising explanation is what has been called the “Ferguson effect”. At least in some places, there was pullback in policing either in reaction to protests or just in reaction to Covid.

JD: There is a researcher at the Public Policy Institute of California, Deepak Premkumar, who has a nice paper showing some de-policing effect in the wake of these high-profile police killings that seems to be increasing homicide a bit. There is a question about how much of the overall magnitude that we’re seeing in homicide is due to this versus other stuff. My hunch is this is not going to explain all of it, but I agree there’s certainly more evidence in favour of that theory than the progressive prosecutor theory.

Part of what’s going on — and now this is me speculating — is that there were some events that triggered a change in behaviour and then triggered a whole bunch of other changes in whether people got guns or there was gang violence and retaliation, and led to sort of a new equilibrium. A big part of it was just that there weren’t people out and about and we didn’t have eyes on the streets, we didn’t have potential witnesses. People were free to go out and wreak havoc, and also there were all these gun purchases.

JF: There was a huge increase in gun sales ...

JD: Because everyone’s like, “The world is ending.” I’ve had a lot of conversations with police recently about how if you want to steal a gun, the best possible place to look is in the back of a pick-up truck that’s sitting in a parking lot. It’s just really easy to steal guns, and in a situation where there was a huge increase in gun purchases, you wind up with a lot more guns out there to be stolen. Even if you assume that everyone who purchased a gun is a law-abiding citizen who is not going to go out and shoot someone, the guns get stolen.

There is a really cool paper by Bill Evans and some colleagues looking at the effects of the crack epidemic and how that pushed us to a new equilibrium where everyone was incentivised to buy more guns and carry a gun more often. They show that among young Black men it has persisted even as violent crime rates have fallen for everybody else. That is what has gotten me thinking about this idea of maybe we’re just in a new equilibrium now, especially on the gun side.

The main thing I try to point out to policymakers is we don’t have to fully understand why we are here to come up with ideas of what to do about it. We can have ideas about what to do about violent crime that do not require us solving this problem that we might never solve.

JF: What are some top candidates?

JD: Putting more police on the streets reduces homicide, reduces violent crime. There is plenty of research on that. There are also plenty of discussions now about the potential social costs of over-policing, so it’s reasonable to have conversations about whether that is the route you want to go. Also, it is really hard to recruit police right now.

We know that increasing the probability of getting caught for crimes has a big deterrent effect in a way that potentially locking people up for 20 years on the back end does not. No one is looking that far ahead. Putting cameras everywhere, adding more people to DNA databases will increase the probability that you get caught if you offend. We have lots of good evidence that would deter crime.

On the not-law-enforcement side, summer jobs programmes seem to put children on to a different path and change everything from reducing violent-crime arrests to actually reducing mortality from gun violence. Even stuff like day-to-day changes in air pollution, which way the wind is blowing from the highway. If it’s blowing this way, violent crime goes up over here. If it blows that way, the violent crime goes up over there. Given that, putting better air filters in classrooms, in offices, in public housing, wherever we could think to put them, would probably have a beneficial effect.

There is a bunch of research on things like greening vacant lots. If you’ve got this dilapidated house, you come in and plant some gardens and trees and stuff. That seems to be really useful. Certainly that is a cheap, easy, uncontroversial intervention.

Leniency towards first-time offenders in the long run is probably a good investment. Another thing is increasing access to mental healthcare. There is this amazing paper using data from South Carolina showing that when we kick children off Medicaid at age 19, when it becomes much harder to stay on Medicaid, you just see all the children get kicked off and then in the other graph you see everyone immediately locked up. It is these children who were using Medicaid to get mental health treatment, they’re the ones that are now at very high risk of being locked up.

JF: What is it that makes evidence compelling? What is good research?

JD: Honestly, a control group. You need to have some sort of control group that plausibly tells you what would have happened without this intervention. The ideal would be you randomise people to either get a summer job or not, and then you just watch what happens for the next couple of years. In some cases, like in the summer job literature, they actually did run those randomised trials. But in a lot of cases it’'s impossible to randomise that way, so you have to find natural experiments, which is the fun exercise that especially economists get really obsessed about.

JF: In that Suffolk County misdemeanour paper, how do you know that the prosecutors who let people off didn’t just think those people seemed less dangerous?

JD: We had information on who the prosecutor was, which was key. We used their decisions in all of their other cases as a measure of how lenient or harsh they were. Cases were randomly assigned across prosecutors, so we know there’s no difference in the composition of the cases. It is sort of a randomised experiment.

JF: Are there any topics where this kind of research has flipped the consensus?

JF: There’s definitely stuff that we know doesn’t work. I have worked on “ban the box” policies, which are not really a violence-reduction policy, but there was a lot of hope that it would help people with criminal records get their foot in the door with an employer if we tell employers they can’t ask on a job application if you have a criminal record — we “ban the box” that people used to have to check saying they have a criminal record.

Economists like me look at that and say, “We haven't changed any of the incentives facing the employer if they don’t want to hire someone with a criminal record.” Now they are going to guess and, based on the racial disparities in the US in who has a criminal record, they might guess that a Black applicant is more likely to have a record than a White applicant is. We can look at the data and find natural experiments to test what happens when “ban the box” policies go into effect. We find that employment for young Black men falls after “ban the box” goes into effect. Young Black men who don’t have a criminal record can no longer signal that upfront, and they can’t get a job any more. But it does not seem to help people with records, which is the icing on this terrible cake.

But then there are also policies where, like, who would have thought that summer jobs would be as effective as they seem to be? We now have all these amazing studies showing dramatic effects on violent-crime arrests for a really long time. Having this kind of research allows us to find needles in the haystack of programmes. There are lots of well-meaning programmes out there. Most of them are not doing very much. This one really does something. So now we can use that as evidence to go to funders and cities to say, “Invest your money here, this is actually worth it.”

JF: You were talking before about how when you go into these experiments, you have priors, but sometimes it comes out different. Clearly there are groups in a lot of these fields that when you do that, it makes their heads explode. I’m thinking of the Naloxone experiment. Describe that.

JD: Naloxone is a drug that will save your life if you’re in the midst of an opioid overdose. It basically stops the overdose, and that is amazing. This is a major medical innovation. So there have been a bunch of policies passed to make it easier to access naloxone. It used to be you needed a prescription from your doctor. We’re in the middle of an opioid epidemic; for obvious reasons people wanted to loosen those restrictions so you could have it in your medicine cabinet in case someone in your family happens to overdose.

As the cynical economist, you are always looking for unintended consequences. On the one hand you’re saving someone’s life. On the other, everybody now knows the risk of death when you use opioids has fallen if you’ve made naloxone more widely available. If you’re reducing the expected cost or the risk associated with using opioids, then we would expect people to do more of it. The question Anita Mukherjee and I were going into this study with was: how much of the benefit does that cancel out?

We look at the passage of these laws across different places as a natural experiment, comparing trends over time in different places. We find that on net, people go to the ER more. Which could be good — they’re seeking help. Crime seems to go up, which is not so good. There’s no net effect on mortality. That suggests to us that there is more risky use that is cancelling out the benefits. Our takeaway is this might be great for saving lives in the moment, but it is clearly not a solution to the opioid epidemic, so we need to make sure we’re investing in treatment, investing in other things.

There is a certain corner of the public-health and harm-reduction world that gets very angry when you suggest that their preferred policies are not working as well as they say they are. Those folks come after me on the internet about once a year, and I have stopped taking it personally because clearly this is just what they do.

JF: Do you think if there is more and more research, it will eventually change people’s minds?

JD: That’s the hope. I have a lot of faith that, in the long run, the literature will converge on the right answer. Science will win out. We will figure this out. One challenge is that it’s clear that within this corner of the public-health community, part of the goal is to stop researchers from publishing studies that have this kind of finding. Then it becomes not surprising that you don’t see studies coming out of public health that show that harm-reduction efforts don’t always work. Not everybody wants to be beaten up on the internet once a year. There is a very clear incentive to toe the party line.

That is worth talking about because, again, I’m confident in the long run we’ll get the right answer, but it would be nice if we got to the right answer sooner, especially when we’re in the middle of an opioid epidemic. Wasting our time on interventions that are not working well is really costly. These are life-and-death policy areas, and I believe the fastest way we will get to solutions that will have meaningful effects on people’s lives is to be testing and iterating. Don’t assume that the thing we try is going to work; assume it won’t work. Our goal should be to figure that out as fast as possible so we can move on to the next thing. This is my soapbox.

⁎⁎⁎

Round-up of recent research

Prosecution

Michael Mueller-Smith and Kevin T. Schnepel, Diversion in the Criminal Justice System (The Review of Economic Studies 2021). Ungated version.

Amanda Agan, Jennifer L. Doleac and Anna Harvey, Prosecutorial Reforms and Local Crime Rates (working paper, 2022).

Amanda Agan, Jennifer L. Doleac and Anna Harvey, Misdemeanour Prosecution (Quarterly Journal of Economics 2023). Ungated version.

⁎⁎⁎

Pre-trial detention

Arpit Gupta, Christopher Hansman and Ethan Frenchman, The Heavy Costs of High Bail: Evidence from Judge Randomisation (The Journal of Legal Studies, 2016). Ungated version.

Emily Leslie and Nolan G. Pope, The Unintended Impact of pre-trial Detention on Case Outcomes: Evidence from New York City Arraignments (The Journal of Law and Economics 2017). Ungated version.

Paul Heaton, Sandra Mayson and Megan Stevenson, The Downstream Consequences of Misdemeanour Pre-trial Detention (Stanford Law Review 2017).

Megan T. Stevenson, Distortion of Justice: How the Inability to Pay Bail Affects Case Outcomes (The Journal of Law, Economics and Organisation 2018).

Will Dobbie, Jacob Goldin, Crystal S. Yang, The Effects of Pre-trial Detention on Conviction, Future Crime and Employment: Evidence from Randomly Assigned Judges (American Economic Review 2018).

Alex Albright, No Money Bail, No Problems? Evidence from an Automatic Release Program (working paper, 2021).

Stephen Koppel et al, Examining the causal effect of pre-trial detention on case outcomes: a judge fixed effect instrumental variable approach (Journal of Experiment Criminology 2022).

René Ropac and Michael Rempel, Does New York’s Bail Reform Law Impact Recidivism? A Quasi-Experimental Test in New York City (working paper, 2023).

⁎⁎⁎

Ferguson effect

David C. Pyrooz et al, Was there a Ferguson Effect on crime rates in large US cities? (Journal of Criminal Justice 2016). Ungated version.

John A. Shjarback et al, De-policing and crime in the wake of Ferguson: Racialised changes in the quantity and quality of policing among Missouri police departments (Journal of Criminal Justice 2017). Ungated version.

Tanaya Devi and Roland G. Fryer Jr, Policing the Police: The Impact of “Pattern-or-Practice” Investigations on Crime (working paper, 2020).

Cheng Cheng and Wei Long, The effect of highly publicised police killings on policing: Evidence from large US cities (Journal of Public Economics 2022).

Deepak Premkumar, Public Scrutiny, Police Behaviour and Crime Consequences: Evidence from High-Profile Police Killings (working paper, 2022).

⁎⁎⁎

Enduring effects of the crack epidemic

William Evans, Craig Garthwaite and Timothy Moore, Guns and Violence: The Enduring Impact of Crack Cocaine Markets on Young Black Males (Journal of Public Economics 2022). Ungated version.

⁎⁎⁎

Police presence

John M. MacDonald, Jonathan Klick, Ben Grunwald, The Effect of Private Police on Crime: Evidence from a Geographic Regression Discontinuity Design (Journal of the Royal Statistical Society Series A: Statistics in Society 2016).

Aaron Chalfin and Justin McCrary, Are US Cities Underpoliced? Theory and Evidence (The Review of Economics and Statistics 2018). Ungated version.

Steven Mello, More COPS, less crime (Journal of Public Economics 2019). Ungated version.

Emily K. Weisburst, Safety in Police Numbers: Evidence of Police Effectiveness from Federal COPS Grant Applications (American Law and Economics Review 2019). Ungated version.

Sarit Weisburd, Police Presence, Rapid Response Rates and Crime Prevention (The Review of Economics and Statistics 2021). Ungated version.

Aaron Chalfin et al, Police Force Size and Civilian Race (American Economic Review: Insights 2022). Ungated version.

⁎⁎⁎

Surveillance cameras

Mikael Priks, The Effects of Surveillance Cameras on Crime: Evidence from the Stockholm Subway (The Economic Journal 2015). Ungated version.

Eric L. Piza et al, CCTV Surveillance for Crime Prevention: A 40-Year Systematic Review with Meta-Analysis (Criminology & Public Policy 2019). Ungated version.

⁎⁎⁎

DNA databases

Jennifer L. Doleac The Effects of DNA Databases on Crime (American Economic Journal: Applied Economics 2017). Ungated version.

Anne Sofie Tegner Anker, Jennifer L. Doleac and Rasmus Landerso The Effects of DNA Databases on the Deterrence and Detection of Offenders (American Economic Journal: Applied Economics 2021).

⁎⁎⁎

Summer jobs

Alexander Gelber, Adam Isen, and Judd B. Kessler The Effects of Youth Employment: Evidence from New York City Lotteries (The Quarterly Journal of Economics 2015). Ungated version.

Erin Jacobs Valentine et al, An Introduction to the World of Work: A Study of the Implementation and Impacts of New York City’s Summer Youth Employment Programme (report prepared for US Department of Labour by MDRC, 2017).

Jonathan M.V. Davis and Sara B. Heller Rethinking the Benefits of Youth Employment Programmes: The Heterogeneous Effects of Summer Jobs (The Review of Economics and Statistics 2020). Ungated version.

⁎⁎⁎

Air pollution

Malvina Bondy, Sefi Roth and Lutz Sager Crime Is in the Air: The Contemporaneous Relationship Between Air Pollution and Crime (Journal of the Association of Environmental and Resource Economists 2020). Ungated version.

⁎⁎⁎

Greening

Charles C. Branas et al, Urban Blight Remediation as a Cost-Beneficial Solution to Firearm Violence (American Journal of Public Health 2016).

Charles C. Branas et al, Citywide cluster randomised trial to restore blighted vacant land and its effects on violence, crime and fear (Proceedings of the National Academy of Sciences, 2018).

Michelle C. Kondo et al, Neighbourhood Interventions to Reduce Violence (Annual Review of Public Health 2018).

⁎⁎⁎

Mental healthcare

Elisa Jácome, Mental Health and Criminal Involvement: Evidence from Losing Medicaid Eligibility (working paper, 2022).

⁎⁎⁎

Ban the box

Jennifer L. Doleac and Benjamin Hansen, The Unintended Consequences of “Ban the Box”: Statistical Discrimination and Employment Outcomes When Criminal Histories Are Hidden (Journal of Labour Economics 2020).

⁎⁎⁎

Naloxone access

Jennifer L. Doleac and Anita Mukherjee, The Effects of Naloxone Access Laws on Opioid Abuse, Mortality and Crime (The Journal of Law and Economics 2022). Ungated version.

• Justin Fox is a Bloomberg Opinion columnist covering business. A former editorial director of Harvard Business Review, he has written for Time, Fortune and American Banker. He is author of The Myth of the Rational Market