Do the Police Have a Management Problem?

This is a scene we recorded recently at a professional training session in Chicago:

Luann PANNELL: We got to train the way we expect people to perform. And sometimes when they don’t perform in those use-of-force situations, we kind of have to take a step back and go, “Well, why would they, based on the PowerPoint that they just got?”

This training was being held at the University of Chicago.

Sandy Jo MACARTHUR: And if you start doing it in your district, you’re going to make a difference, and you’re going to see either uses of force go down or the amount of force used.

This is a brand-new program called the Policing Leadership Academy. Now, you may ask yourself: police training at the University of Chicago?! This is a school known for its intense undergrad core curriculum; for its economics department and business school; for its medical and law schools. Why would they be teaching police officers? Here’s why:

Jens LUDWIG: Spending time looking at policing and realizing, these aren’t high-performing organizations of the sort that you would expect. 

That is Jens Ludwig. He’s an economist at UChicago; he teaches in the public-policy school, and he also runs a research center called the Crime Lab.

LUDWIG: It’s not hard to go sit in a courtroom, visit a prison or a probation office to understand how they work. But policing, really — real policing is the most opaque part of this. And so we went out and spent a lot of time just watching what police do, sitting in police cars and police stations. One of the biggest surprises to me is that, you know, I think in most cities in the United States, they just haven’t made the shift to high-performance, professionalized organizations yet. 

DUBNER: And what do police departments that you work with say when you say, “Hey, on average, police departments are not very high-functioning”?

LUDWIG: Well, in my experience, having spent a lot of time around Chicago cops, nobody complains more bitterly about the terrible functioning of the Chicago Police Department than Chicago cops.

Just to be clear, Ludwig is not anti-police.

LUDWIG: They have a very difficult job in a very complicated set of social environments.

It’s how police departments are run where he has a problem:

LUDWIG: If you look inside a police department, you take a person who’s a very good patrol officer with a college degree and no management experience, and based on being a really good chaser-downer of criminals, you say, “Congratulations, now you’re running a police district in Chicago that has maybe 500 police officers and 200,000 people living in the area. I’m sure you’ll figure it out. Good luck.”  

What Ludwig is describing here is a phenomenon known as the Peter Principle: someone is good at doing a particular job, so they get promoted into a different job, typically a management position — which they may not be so good at. We once made an episode about the Peter Principle; it was called “Why Are There So Many Bad Bosses?” This happens in software firms, it happens in hospitals and universities — and Ludwig says it certainly happens in police departments.

LUDWIG: Nobody would look at that and say, that sounds like a perfect recipe for success. And then you ask, why doesn’t the public demand better? The public doesn’t know.

So today on Freakonomics Radio: let’s let the public know! And we’ll hear about the new training curriculum that’s meant to help:

MACARTHUR: The curriculum has to be relevant to the job, it has to be realistic, and it has to be rigorous. 

We’ll hear how police managers respond to the curriculum:

Michael WOLLEY: Within the first two days, I was in awe.

And we’ll hear the origin story of this unusual research project:

LUDWIG: The project didn’t really start as a research project. It started as a, “Holy s***, the City of Chicago is on fire, someone’s got to do something” project.

*      *      *

Stephanie Drescher is a police captain in Madison, Wisconsin.

Stephanie DRESCHER: I love my job. I love my job. I can’t imagine doing anything else.

Although Drescher did start out doing something else.

DRESCHER: I didn’t go to school to be a police officer. I have a degree in history and religious studies. Edgewood College, in Madison. I did insurance for many years, and part of my job was to do risk management for the city of Madison through worker’s compensation.

One day, her job brought her to the police academy in Madison.

DRESCHER: I was just there to fill out the forms for everyone. And then Sergeant Koval — became literally, later, the chief of police — said, “How old are you? Where’d you go to college? Have you ever thought about being a police officer?” And I said, “Twenty-five, Edgewood, and absolutely not.” But I frankly wasn’t really thrilled pushing papers around every day doing insurance. So, I said, “Let’s give it a whirl.”

Soon after, Drescher was back at the police academy.

DRESCHER: Six months of Madison training specific, but also state qualifications — so, defense-and-arrest tactics, professional communication, firearms. And then you go into field training for about three to four months. I put in for the job in November of 2008. I started the academy in May of 2009, and I finally went solo in February of 2010. So, it’s quite the process.

Drescher spent her first six years as a patrol officer.

DRESCHER: I got promoted to sergeant in 2016. I was promoted then to lieutenant in 2019, and I was promoted to captain in 2021.

You might think that each of these new jobs would come with another extensive period of training, like when Drescher first became a police officer. But that’s not the case.

DRESCHER: You get promoted to sergeant, you get three weeks. And then they tell you, go do 40 hours of subject-matter expert training, but figure it out on your own time. You get promoted to lieutenant, and then they’re like, “You’re going to go to the North District. Have fun. Figure it out.”

One thing that’s important to say is that the way things are done in Madison, Wisconsin, may differ from how things are done in Chicago, or New York, or L.A. — and especially in the smaller police departments around the country. Law enforcement in the U.S. is famously decentralized: there are 18,000 federal, state, county, and local law-enforcement agencies, each with its own way of operating. There is some coordination, and standardization — but not nearly as much as you might think. So when it comes to how police departments are managed — well, that’s where Jens Ludwig and the University of Chicago Crime Lab thought they could help. This goes back a few years.

LUDWIG: My first year at the University of Chicago was 2007. That fall, there’s a doctoral student in chemistry from Senegal. The University of Chicago’s chemistry department is one of the best in the world. You can imagine this kid’s family, when they get the admissions letter a few years before in Senegal, this is like the Wonka golden ticket. We’ve won the life lottery to get a Ph.D. from one of the best chemistry departments in the world. Kid defends his doctoral dissertation in the fall of 2007. Two weeks later, at 1:30 in the morning, he’s coming home from an ice-cream social, of all things, is at the door of his apartment, and someone tries to rob him for his backpack, something goes wrong, he gets shot in the chest point-blank and dies. So the University of Chicago likes to go around telling people we won more Nobel Prizes than any other university in the world, which is great. But here you could see our research hasn’t had impact literally in our own backyard.

Ludwig is right that Chicago has more Nobels than anyone in economics. But overall they’re ranked fifth for Nobel Prizes.

LUDWIG: So it led to a lot of soul-searching here about how we can be more useful to the city of Chicago and cities more generally. And so, we looked around trying to understand why we have social problems like gun violence. And there’s politics and there’s lack of funding. But we were of the view that another really important root cause of social problems is lack of understanding about how to solve them. And you can see the government has its hands full just running operations day-to-day. They don’t have the bandwidth nor the — it’s not their thing to do R&D on top of that. And so we thought maybe we could make a difference by being the R&D partner to the government.

DUBNER: When you say that there is a “lack of understanding” — in other words, there are things that can be determined and acted upon that can change the outcomes in the realm of crime, in this case — persuade me that that’s actually true and not just the sort of wishful and/or high-minded thinking of academics.  

LUDWIG: Yeah, well, tell me if this is a responsive response or off-base. The government spends an insane amount of money on all sorts of different programs to try and prevent crime from happening and prevent people exiting prison from coming back. And the large majority of those very well-intentioned programs don’t work. A small share of them do work. And yet the government just keeps funding this as if we didn’t know anything. And for a lot of the things that they spend their money on, we just have no idea. I was at a conference recently, and someone from M.I.T. claimed that 99 percent of what the government spends money on has just never been evaluated with data at all.

DUBNER: If indeed it’s — let’s say, even above 50 percent, why do you think that would be so common? It’d be idiotic for a family to spend a lot of time and money and other resources on something and then not think about how to determine whether it’s successful or not. So why do you think that’s so much more common in government, if indeed that’s the case? 

LUDWIG: Let me offer you a hypothesis, which is that in our daily lives, we get a lot of feedback about what works and what doesn’t work. So, if my wife says, you know, “How do I look in these pants?” I very early in our marriage learned what the right answer is to that.

DUBNER: Which is? 

LUDWIG: Which is, “Great!” And similarly in the private sector, if McDonald’s comes out with a new, like, McPizza, and it’s horrible, they very quickly can tell that it’s horrible because they just don’t sell anything. But the problems that the government is trying to solve are incredibly complicated, and have a long list of determinants. If you live in Chicago and you were to ask yourself, how good is the Chicago public school system? That’s an incredibly difficult question to answer. Is the police department doing a good job? It’s really, really hard to know. 

But trying to figure out things that are “really, really hard to know” — well, that’s what academic researchers do, or at least try to do. So Ludwig set out to understand what police departments do well, and what they don’t. This would yield a research paper called “Policing and Management,” which Ludwig co-authored with Max Kapustin and Terrence Neumann. In the paper, they note that other economists have firmly established that a given manager can make a big difference in how a typical company is run. Here’s Kapustin:

Max KAPUSTIN: Having read those papers and thought about the issue, it does seem like there are a set of management practices that seem to be, you know — I think you can call them best practices, right? About how you set targets and how you use data to plan your operations and how you promote people who are talented and how you hold people accountable when they don’t perform. I mean, it’s really not rocket-science kind of stuff. And at that point, you’re like, “Well, okay, if this works for private firms, why wouldn’t it potentially work for public-sector organizations, including police departments?”

So Kapustin, Ludwig, and Neumann analyzed several years’ worth of data from the 50 biggest police forces in the U.S. They zeroed in on every time there was a turnover in management. And then they looked to see what happened to the crime rate and to police use of force. What did the data say?

LUDWIG: Managers really matter a lot in policing. Even holding the police resources and strategy constant, just getting a better versus worse leader in place can generate huge swings — huge improvements in public safety and huge reductions in police use of force. 

Okay, so that’s what their research shows. But for the sake of the larger story we’re telling today, we need to point out that this research project didn’t really start out as a research project.

LUDWIG: It started as a, “Holy s***, the City of Chicago is on fire, someone’s got to do something” project, right? So the murder rate jumped by something like 60 percent from 2015 to 2016. Everybody’s looking around thinking, “Oh my God, what do we do?” 

That spike in murders wasn’t the only problem the Crime Lab was thinking about. There was also a growing lack of confidence in the Chicago Police Department, especially after the police murder of a 17-year-old named Laquan McDonald. That killing led the federal government to launch an investigation of the Chicago Police Department.

LUDWIG: Then the Obama administration’s Civil Rights Division of their Department of Justice connects us to this guy, Sean Malinowski. And he says, “I’ve got an idea.”

Sean MALINOWSKI: Yeah. My name is Sean Malinowski, and I’m a retired chief of detectives from the Los Angeles Police Department.

Malinowski was still with the L.A.P.D. when Ludwig reached out. But things were so bad in Chicago — the U.S. Department of Justice had said so — that there was a clear path for an L.A. cop to come in to Chicago to consult.

MALINOWSKI: They had a dramatic increase of about 300 homicides in 2016. So they wanted a team to go in and take a look at why that happened and what might be able to be done to turn that back. We just started talking to folks about where were there holes in the system. And what we found was there was a lot of technology that had been purchased and deployed in Chicago, but not a lot of training on that technology, and not a lot of direction to officers in terms of what they needed to be doing in the field.

This led to the creation of what Malinowski’s team called Strategic Decision Support Centers, or S.D.S.C.s.

MALINOWSKI: That’s kind of a mouthful, I realize, Strategic Decision Support Center — it’s not even a fun acronym that someone came up with. But it makes sense if you think about what they’re trying to do in those centers. So they’re decentralized. A lot of police departments have now a real-time crime center, and that’s centralized. And Chicago had that. But what we found is that information was not getting back out — about crime trends, about what was happening in the neighborhoods — was not making its way back down to the officers on the street, nor were they getting information from officers who were out there every day. So we essentially went in, built a room, we put some technology into the room so that you could view some of the C.C.T.V. cameras and other technology that was out there. You know, they already had gunfire-detection systems, they already had license-plate recognition. They had systems that could show you where your police cars were as calls for service were coming out. The problem is there was nothing really being done with it unless you had a very ambitious commander who, with everything else that they have going on was going to take the time out of their day to focus on what was happening. So we just put it all in one place, and we teamed up civilians with police officers. And what’s interesting is, Chicago Police Department in 2016 had no civilian crime analysts. I mean, for comparison, we had at least 100 on L.A.P.D. for a smaller police department: we had 10,000 officers and 100 analysts. They had 14,000 officers and no analysts. And that’s when the University of Chicago Crime Lab jumped in and really when everyone else was kind of wringing their hands about what can be done about this, the Crime Lab jumped in with both feet and said, “We’re going to commit to this thing.”  

The Crime Lab sent data analysts to the S.D.S.C.s.

MALINOWSKI: They hired crime analysts extraordinaire. They were really data scientists. They were people that the department really wouldn’t have been able to attract, I don’t think. 

Their job was to synthesize all the incoming data for local commanders.

MALINOWSKI: And we just presented the information to the officers who could make an impact on the street. 

One of the first S.D.S.C.s. was built in Englewood, a neighborhood on the South Side of Chicago with a lot of gun crime.

MALINOWSKI: When I got out there to Englewood, I looked around roll call, and I just saw how defeated the officers felt. They felt like they weren’t making an impact. And, you know, these are the officers that are going, every night, responding to multiple shootings. They’re dealing with people who’ve been shot, they’re getting blood on their shoes. I mean, it’s a horrific job sometimes. We set up a system where every day, the commander was to get briefed on what happened yesterday, and what resources they had to deploy today, and how we were going to solve some of these problems.

LUDWIG: The first month after Sean pushed the button, the newspapers were starting to report — what in the world is going on in Englewood? 

MALINOWSKI: We had great success in Englewood during the first month. It was a short month, February, but we went without any homicides and very few shootings during that month. So, when you can prevent shootings and homicides from happening so that the people in Englewood, both the officers and the people living there, had a reprieve of a month without a homicide, everyone notices that. 

If this approach to policing sounds familiar to you — well, it is. The New York City Police Department has for a few decades now used a system called CompStat, which gathers and distributes crime data in very much the same way that Sean Malinowski just described. CompStat has been considered a huge success in New York, and it’s been exported to police departments around the world. Chicago has tried it — several times — but for a variety of Chicago reasons, it hadn’t taken root. But this new S.D.S.C. setup seemed to be working. Here’s Jens Ludwig again, from the Crime Lab:

LUDWIG: One of the commanders, he got quoted in an interview with the newspaper where he said, before these management changes that we implemented in his district, he said, “My guys would basically just normally drive around randomly.” Sean goes in and gets them to do something that looks a lot more like standard operating practice in a high-performing organization — use data to figure out what you’re going to do, do it, use data to figure out if it worked, and adjust, and so on. So I think the real kind of “a-ha” here is not — if I just described what they were doing after the intervention, you would say, “Where’s the intervention?”

DUBNER: It also sounds kind of like common sense.

LUDWIG: It does! All of this only makes sense once you realize what the status quo was beforehand. 

The Crime Lab helped Sean Malinowski’s team open Strategic Decision Support Centers in other high-crime neighborhoods in Chicago. The results were similar to Englewood: a significant decrease in gun violence. But: it didn’t last. A year later, gun crime in those places was back where it was before Malinowski and the Crime Lab showed up. Why? Here’s Malinowski again:

MALINOWSKI: If you think about the concept of the Strategic Decision Support Center, it’s to provide information to that decision-maker — so, who’s managing that district at the local level. And in a police department the size of Chicago, there’s all different levels of maturity and professionalism and training in those spots. Sometimes you get someone who was a lieutenant yesterday running a squad, and then the next day they have the whole district they’re in charge of, and with very little training on how to navigate that. And so sometimes they fall back on what they know. So if they had a certain background, then the district starts going in that direction. You know, if they have a narcotics background, they start looking at that, because it’s what they’re comfortable with.

In other words: yes, data can be a powerful tool. But there needs to be someone capable of managing the data and managing the police officers. And that’s how the University of Chicago’s Crime Lab came up with its next big idea: what if they could create a sort of M.B.A. program for police commanders? And not just police commanders from Chicago, but from all over the country. Coming up: we go inside the first class of the Policing Leadership Academy.

*      *      *

The first cohort of the Policing Leadership Academy, or P.L.A., began meeting in May, on the campus of the University of Chicago. The program’s Curriculum Design Director is Sandy Jo MacArthur:

Sandy Jo MACARTHUR: So the 30,000-foot view is the University of Chicago, in particular Jens Ludwig, had this concept, as he had seen in smaller, bite-sized pieces in Chicago, that if we actually train commanders to use good leadership techniques, if they’re data-driven, if they have the skill sets, that we can reduce crime and violence in the districts, reduce gun violence in particular, and increase trust in the community. And that is the basic concept of the Policing Leadership Academy.

They recruited one manager each from two dozen police departments across the U.S., as well as Manchester, England.

MACARTHUR: We have Los Angeles; New York; Cleveland; Miami; Miami-Dade; Chicago; Indianapolis; Arlington, Texas; Dallas; Waco, Texas; Aurora, Colorado. 

Also, Boston, Washington, D.C., New Orleans, Detroit, Phoenix — and Madison, Wisconsin, as we heard earlier, from Captain Stephanie Drescher.

DRESCHER: You get promoted to lieutenant and then they’re like, “You’re going to go to the North District. Have fun. Figure it out.”  So, I was selected to come here for a traditional formal leadership training, which is not something that we get frequently in-house.  

Madison, it turned out, was not an outlier in the lack of training for police leaders. MacArthur again:

MACARTHUR: We actually took a poll and only three, I believe, of the 24 departments had some sort of training before they took the seat.  

The P.L.A. assembled in Chicago for one week a month, over a period of six months. The curriculum was a mix of classroom sessions, seminars, field trips — and a lot of just hanging out.

MACARTHUR: You know, having a drink and breaking bread. The bonding that happened immediately that first week really surprised me. And it really allowed us to go much, much deeper into the week on trauma, for example. Because we had strategic thinking. Everybody likes that. Risk management — no one likes it, but everybody knows we need it. But we had a whole week that was on violence-reduction and the role of creating healthy and resilient communities. And we were able to look at their own sense of resiliency. How do you keep yourself resilient while you’re managing and leading the community? And what a powerful week that ended up being.  

I asked MacArthur what her goals had been for this first P.L.A. class: 

MACARTHUR: I’d love for them to be data-literate, where they understand taking data and changing it from information into intelligence. I do want them to be trauma-informed. Because when you understand trauma and the impact trauma has on you personally, your men and women in your department, as well as your community — boy, you can totally depersonalize responses from the community. And then the third piece I would really like is to be culturally competent. One of the things with being culturally competent is you are naturally curious. And you don’t make an assumption about a certain community, or a person who lives on the certain side of the tracks, or whatever.

MacArthur is a law-enforcement consultant; she does most of her work for police departments and universities.

MACARTHUR: And I had spent 35 years, full-time, with the Los Angeles Police Department, retiring in 2015.

She retired as assistant chief, the highest-ranking woman in the L.A.P.D.

MACARTHUR: I do a lot of work in the world of training, use of force, officer wellness. Those were probably my three biggest loves when I was at the police department.

MacArthur joined the L.A.P.D. in 1980, and she was a beat cop in the Foothill Division, in the San Fernando Valley. That was the place where, in 1991, Rodney King was beaten by L.A.P.D. officers, including a sergeant named Stacey Koon. The beating was caught on video, and broadcast around the world.

MACARTHUR: So literally 20 days after the King incident, I was working in South Los Angeles, and I was sent back to Foothill Division. I had been there as an officer, and I was sent back there as a sergeant. So I was given the unit that the sergeant who had previously been in the unit — Stacey Koon, and he and his team, a lot of them were stood down because of what happened that night, that was caught on video. I was given that unit. So, of course, they thought I was an Internal Affairs plant. I even asked the captain at the time, “What is your goal for me coming in here? I’m really happy in South Los Angeles. Why are you doing this to me?” So internally, there was a lot of angst. But externally, every call I went on, the community was outraged. We would be handling a robbery-in-progress call where the owner of the store who called us was glad we were there, but the people surrounding us were yelling and spitting. And my officers — it felt very personal, right? I was very, very involved in rolling on as many calls as possible, just knowing that there was this clash constantly and that it could be like putting a match on gas. But optics-wise, I had to listen to the community first, or they immediately would think I was siding with the officers.

So I would say in roll calls, “Hey, guys, this is the way I’m going to handle these calls. If you need me to come to the scene, send me a little information in the computer about what the call is. Know I will come and talk to you. But I needed to go to the community first, because they feel like they had no power, right? And just by the sheer nature of our uniform, we look like the power guys, you know? So I started to do that, and it was amazing because the officers are then able to depersonalize it, sort of get their amygdalas hijacked, all that stuff, right? And so they’re in fight-or-flight, just like the community is in fight-or-flight. And just by giving a little bit of space, and giving the party who feels the most harmed their opportunity to first tell their story, that set a different tone right out of the chute. So I use that example when I teach supervisors, how important optics are. And then I would go in not with the tone that the officers might have had where they’re giving commands, “Stay back,” you know, “You’re crossing a police line.” I went in with a much more, I would say, conversation like we’re having today. You know, going over, they’re screaming and yelling at me and say, “Hey, you know, come on over here. Tell me what your concern is.” 

DUBNER: Now I’m thinking about your role in the Policing Leadership Academy and thinking, “Well, you’re training these supervisors.” And I’m wondering about your supervisor back then in that moment in Foothill, what do you wish had happened or what do you think would have been a positive thing to happen for your commander, at that moment? To speak not only to you but the rest of your colleagues, to say, “Okay, here’s something we might try.” In other words, it sounds like you figured something out on your own, but I’m curious how much benefit there might have been if it were more systematic.

MACARTHUR: I agree with you 100 percent. Because if people were able to depersonalize these conflicts — so, I know the community is not going to be able to depersonalize easily. But if we as an organization, as a leadership team, could have sent that message, that we support officers doing the right thing, but we are the ones who have to try to depersonalize these conflicts. It’s about the badge, it’s about the uniform, it’s about the history of L.A. I used to say this all the time in training and roll calls: “For the sins of the fathers, the sons must pay.” It’s history. It’s policing history. And you weren’t even born. I wasn’t even born. And we’re still paying for things because nobody got in there, rolled their sleeves up and said, let’s start to work through this. So, you can go all the way back to how officers in the South were used to manage slaves. We had much more corruption within police departments in the ‘20 and the ‘30s. Some agencies dealt with it harder than others. For example, L.A.P.D., you get Chief Parker coming in, and he’s like, “I’m going to manage this just like the military, and we’re going to root out corruption.” So he comes in with a very militaristic tone — which, in some places it was super-effective, and in other places it sort of backfires. And then you get into the ‘60s, and the turmoil across this country in the ‘60s politically, and police were called in to manage that. You can see across this country how police were utilized oftentimes for things they shouldn’t have been, but then it became their job. I always say the peacekeepers, I wish that’s what we became. But we became the enforcer. 

This new Policing Leadership Academy is trying to update that enforcer tradition. There are sessions called “The Psychology of Conflict” and “Organizational Morale.” We asked some of the participants in the first cohort what they’d been learning, or found particularly useful. Captain Drescher from Madison mentioned a tactic called Show Me Clear. It’s about teaching her officers to make a mental reset after every call they respond to.

DRESCHER: It’s something that we train, but it’s something that wasn’t institutionalized. So, it’s just take a breath, reset, frame yourself up to go to the next call.

We also spoke with Michael Wolley. He’s the Deputy Chief of Operations with the Indianapolis Metropolitan Police Department. He came away with some procedural ideas, thanks to a classmate from Washington, D.C.

Michael WOLLEY: We were having violence in our bar districts. We had a couple shootings, there’s homicides, stuff like that. What D.C. brought to the table that we didn’t necessarily think of, or at least I had not thought of, was using other city resources to close down streets to make sure that there are safe routes for pedestrians, adding additional lights, having probation come out. So, just bringing in other elements or government entities to kind of help us tackle this problem.

And we spoke with Kenneth Corey. He spent more than 30 years with the New York Police Department; and he’s one of the people leading the P.L.A. program.

Kenneth COREY: We did a whole piece on social-network analysis. So when we talk about violence, and we talk about the small numbers of people who drive violence and are always kind of encircled by violence, knowing who they are is very helpful when violence starts to flare. So, if we know that this particular individual was the victim of a shooting, who’s likely to carry out the retaliation for that shooting on their behalf, right? Because that’s where I need to focus my disruptive efforts — prevention, outreach, community violence interrupters. But I also know where I need to push uniformed police officers to serve as the deterrent until the detectives can complete their investigation.

Corey also preached the need for police leaders to treat their officers well.

COREY: Just letting your people know that you care about them, right? You know, learning people’s first names, learning something about them, like inquiring about their fam — making them feel like a valued member of the team. And the officers are as happy as they’ll ever be. I mean, cops complain, that’s the nature of the beast. Doesn’t matter how good they have it, they’re going to complain about it, and we like to say that the only thing the cops hate more than change is the way that things are.

Each of the participants also had to complete what Sandy Jo MacArthur called a capstone project — basically, a mini-thesis.

MACARTHUR: Yes. So the capstone project — our commitment to the chiefs, and the superintendents, and commissioners was: if you allow your person to come, we will ask them to work on a violence-reduction problem that they are dealing with. One took on mental health, which is huge, right? So she took on bringing all the stakeholders in the city together, and looking to see how can we do a co-response model that they didn’t really have. 

DUBNER: When you say a co-responder, you mean a mental-health professional, is that right?

MACARTHUR: A co-responder is when you have one sworn officer who’s dressed in jeans, very casually, responding to a call with a mental-health professional.  

DUBNER: Now, what you’re describing there is already underway in some places, as I understand it, right? Or at least being piloted in some places?

MACARTHUR: It is underway in a lot of places and it’s been very effective, but it hasn’t been really the big focus down in Florida. So she’s looking at all that. I’m excited to hear more about hers. 

The project MacArthur’s talking about is the work of Sindyanna Paul-Noel.

Sindyanna PAUL-NOEL: You think of Indiana Jones and you just say Sindyanna. Really easy.

She’s a lieutenant with the City of Miami Police Department.

PAUL-NOEL: Yes, my capstone project is the co-responder model, where we have licensed clinicians along with our crisis-intervention officers reaching out to our high utilizers of police services when they have mental crisis. So, we have a list of individuals who continuously use 911 and end up getting committed, which is a 72-hour hold for mental assessment. So, trying to reach out to those individuals before they need to call 911 for their mental crisis. We did have one of the service providers drop out. So, then we had to find another service provider, which was very tedious. Licensed clinicians are very hard to come by, who would want to go out in the field. A lot of them have the opportunity and the luxury to work from home. Telehealth is a major challenge for us that we go against. But I was able to reach out to one of the service providers in our area, so they’re on board.

I asked Sandy Jo MacArthur to give an overall assessment of the first cohort of the Policing Leadership Academy.

MACARTHUR: Overall, I’m like — I don’t want to say shocked, but I am so happy with how well it went. My expectations were that we would have a group of 24 or 25 commanders who, if we were lucky, we’d maybe have two-thirds of them that thought that was valuable. I’m feeling — and this comes from both in the classroom, but afterwards, you know, having a drink and breaking bread — I don’t think there’s one person that hasn’t found this incredibly valuable. And I think that’s partly because we have created this incredible network of people from all over the country. 

So that’s good to hear. But of course the important question is: will it work? Can a leadership program like this, over time, help modernize American police departments?

*      *      *

So Jens Ludwig, an economist at the University of Chicago, has accidentally become a police reformer. Economists do like to stick their noses in areas that don’t seem to have much to do with economics. So I can imagine that some listeners — especially if you are critical of modern policing in general — that you might be skeptical. On the other hand: as an economist, Ludwig may have found an angle that others wouldn’t have.

LUDWIG: You know, I’m living in the city that has the highest unfunded pension debt of any city in the country. And any solution to the public-safety problem in Chicago that requires the infusion of huge sums of money already faces a huge problem. You go to the mayor and say, “I’ve got a great idea, I only need $2 billion a year to implement it,” you’re going to get crickets as a response.

DUBNER: And your idea doesn’t require so much money because why?

LUDWIG: Because training compared to salary is cheap. And because we’re training leaders, not frontline personnel, there’s a huge amount of leverage from that training. And so those two things together make this an insanely low-cost bet on making things better. Two things that the national conversation would make you think are intrinsically in tension — you can have the police be more fair and also have more public safety at the same time, right? They’re not intrinsically in tension. And so then we realized this is an amazing opportunity to have impact at scale, partly for the reason that you don’t need to train tons of people.

DUBNER: If I were to just ask you about the role of the police in crime reduction, how do you think about that?

LUDWIG: How do I think about the role of police in crime —?

DUBNER: Yeah. In other words, is crime reduction primarily delegated to the police in most cities?

LUDWIG: Oh, I see. Here’s what I would say is, common sense and the data together suggest that there really are very big, powerful, important sort of macro root causes to the crime and gun-violence problem. And the other thing that we can see is, those problems have been incredibly hard to change in the United States. And even if we could organize ourselves as a society and get much more unified than we have been to make a big push and solve those problems, that is not going to be a short-term project. So then you start looking around and saying, while we’re trying to solve the root causes of crime and gun violence, what do we do? And I think policing is one of the most important levers that we have in our toolkit that lets us start making progress quickly. Anything that quote “merely addresses crime and gun violence” often gets kind of crapped upon by people who focus on root causes.

DUBNER: Because they say you’re treating the symptoms —

LUDWIG: Treating the symptom and not the disease. Exactly and I think that’s misguided because what it misses is what a headwind gun violence is in trying to address the root causes.

DUBNER: You’re also thinking about this like an economist who’s looking at costs and benefits. And the fact is that gun violence has massive costs across society.

LUDWIG: And it’s not just costs of the sort that you would think of, like medical costs and lost wages and whatever. Our friend Steve Levitt did a paper many years ago that showed that every serious crime that happens in a city reduces the city’s net population by one person. Every murder that happens in the city reduces the city’s population by 70 people. Which shows you exactly how petrifying gun violence is — most of those murders are committed with guns. Now imagine that you’re trying to do economic development in Englewood, one of the high-violence neighborhoods on the South Side of Chicago. Imagine trying to convince a business to move in to Englewood when, you know, I’ve talked to cops who work in Englewood who are afraid to drive in to start their shift at the police station. If you can get the gun-violence problem under control, that’s a huge tailwind for trying to do economic development, to build the tax base, to get people jobs, to make the schools better, and so on.

DUBNER: In other words, treating the symptom can be really useful.

LUDWIG: Treating the symptom can be enormously useful in treating the underlying cause. The reason that we started the Crime Lab in the first place is you’re looking around the country and you’re seeing these enormously difficult, challenging social problems that are really ruining thousands or millions of people’s lives every year. Gun violence, unfair policing, you know, the list goes on and on. And so then the question is: how do you make progress on that? So when we started off, we were doing a lot of work on the social-policy side, like social programs with kids, and the scale problem there — how do you have a national impact — gets really — you know, there are 50 million school-age kids in the United States. If you want to move the needle at the national level, it’s an incredibly challenging problem. Now, the flip side of that is, take gun violence for starters. The gun- violence problem in America is very disproportionately concentrated in, say, 200 cities. And in every one of these cities, there’s, call it 10 to 20 to 30 key leaders in every police department, the superintendent and then a bunch of middle managers underneath them. And what we have found through some of our research is, who is running the police department matters enormously in what the police department does. And so I think the big idea behind this initiative is: if the gun-violence problem is concentrated in 200 cities and call it 20 people per city are responsible for making the decisions that are incredibly important in affecting the gun-violence problem, that’s, like, 4,000 people to train. Could we and partners train them? And there the answer is, that’s not a crazy aspiration. That’s sort of the scale play here, is that by training the leaders of the departments, you’re getting a huge amount of leverage on the problem.

DUBNER: How were the participating departments in this first cohort chosen? 

LUDWIG: One of the things that we really did not want to do was fool ourselves into thinking that we were doing social good if we weren’t. So what we wanted to do was set this up in a way that we could really measure if it’s having the impacts that we want. Like in K-12, if there’s a good school, there’s excess demand for that, way more kids apply than you have slots, and so what public school systems do is they use a lottery for admissions. We’re doing the same thing here. We’ve got more people who want to come. So we’re randomizing admissions. And so departments had to be willing to share data with us in order to be in.

DUBNER: And then how do you know if you are succeeding, what kind of data are you getting from the police department end, and how long will you be gathering that data?

LUDWIG: We’ll know if it’s a success if we see outcomes like gun violence go down in these districts. And also outcomes like police use of force or arrests that have low public-safety value go down in these districts as well. One of the measurement challenges for the research team is how to then also figure out what’s happening to sentiment by front-line officers and the community as well. But that’s going to be harder to figure out.

DUBNER: So historically, a lot of police data is self-reported data, essentially. How trustworthy do you find police data to be generally, along all the dimensions that you just mentioned?

LUDWIG: Yeah, I think that there’s a lot of variability within police data in trustworthiness. Things like shootings, I think the consensus is that shootings are reasonably accurately measured — 

DUBNER: In part because that’s the type of event that’s hard to hide, correct?

LUDWIG: Exactly. 

DUBNER: But in terms of the way a police officer might interact with someone from the public, how do you feel about that?

LUDWIG: Yeah, that’s really sort of the frontier of the science here. I think the big data revolution is going to help on the measurement. There have been clever people who have capitalized on the shift to body-worn cameras by police departments, to use things like artificial intelligence tools to measure the nature of police-citizen interactions out in the field, for instance. And so I think everything’s on the table in terms of trying to figure out how to measure that.

Jens Ludwig and his Crime Lab will be gathering data from the police departments that were represented in the first cohort of the Policing Leadership Academy. And we’ll report those results to you when we know them. They’re convening their second class in January. I would love to know what you think of this project, and what you think of this episode. Our email is radio@freakonomics.com.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Ryan Kelley. We had recording help in Chicago from Shane McKeon. Our staff also includes Alina Kulman, Eleanor Osborne, Elsa Hernandez, Gabriel Roth, Greg Rippin, Jasmin Klinger, Jeremy Johnston, Julie Kanfer, Lyric Bowditch, Morgan Levey, Neal Carruth, Rebecca Lee Douglas, Sarah Lilley, and Zack Lapinski. Our theme song is “Mr. Fortune,” by the Hitchhikers; all the other music was composed by Luis Guerra.