How Criminal Justice Policy Gets Made In The White House With Rachel Harmon

What happens when the people shaping national crime policy don’t actually have the data they need?

In this episode, I sit down with Rachel Harmon, law professor at the University of Virginia who previously served as a senior policy adviser for criminal justice for the White House Domestic Policy Council. 

Rachel provides a rare inside look at how crime policy really gets made, breaking down  what it’s like working inside the White House -- where decisions move fast, data moves slow, and the pressure to respond to public fear doesn’t wait for evidence to catch up. She explains why even basic questions, like how big a problem carjacking actually is, can be nearly impossible to answer in real time, and how that gap shapes policy decisions.

The conversation dives into the messy reality behind “data-driven policy,” including:

• Why crime data often arrives too late to guide decisions
• How political pressure competes with long-term strategy
• What it takes to actually implement policy after it’s announced
• And why the U.S. still lacks basic data on policing, charges, and outcomes

Tune in for a fun, informative conversation on the messy business of making Federal criminal justice policy. 

Rachel Harmon is the Harrison Robertson Professor of Law and directs the Center for Criminal Justice at the University of Virginia School of Law. She is one of the nation’s leading scholars on policing and the law.

Transcript

SPEAKER_00 0:00

I'm Jeff Asher, and this is the Jeffalytics Podcast. Crime policy doesn't always follow the same timeline as the data. The problems are happening in real time, but how can institutions respond when the data that help explain those problems often shows up months or even years later, if at all? My guest today is Rachel Harmon, a law professor at the University of Virginia and a former federal prosecutor who spent years studying policing, accountability, and the role of law in shaping how agencies operate. She also served as a senior policy advisor for the Biden White House, working directly on those issues at the national level. In this episode, we talk about what that experience actually looks like. We get into how policy gets made by the White House when information is incomplete, what it takes to move priorities across federal agencies, and why implementation is often the hardest part of policy making. Let's get started. My guest today is Rachel Harmon at the University of Virginia now. Rachel, how are you?

SPEAKER_01 0:55

Very good. Thanks for having me.

SPEAKER_00 0:57

Good. Thanks for coming along. So, first question for all guests: what is your background? What brings you here today?

SPEAKER_01 1:05

Well, you can tell me that better than I can in terms of what brings me here today, but I can tell you that I am a law professor at the University of Virginia Law School, where I also direct the Crim Center for Criminal Justice. And I have worked on and off on crime and criminal justice policy for a long time. I started my career. I went to Yale Law School. I clerked for a federal court of appeals judge and for Justice Breyer at the Supreme Court, and then became a federal prosecutor in the Department of Justice. I spent most of my time there in the civil rights division, prosecuting civil rights crimes, which included crimes by public officials and especially law enforcement. And then I decided that I wanted to understand more how law could be used to prevent problems in law enforcement rather than dealing with them afterwards. And I became a professor. And since then I've studied policing, its effectiveness and accountability and the way it responds to communities and how law influences the police. But I think you invited me here today because I got to know you when I spent a year helping out at the White House as a senior policy advisor on criminal justice. And there I worked on crime policy and criminal justice policy.

SPEAKER_00 2:25

Not the only reason. I certainly am very interested in delving into sort of the answer of how can the law help police do their job better and more effectively and more constitutionally. But I do want to talk about the White House because that's how we got acquainted. And I I really just I want to have a thousand questions about that. Like, how does one get picked to do that sort of role within the federal, the federal government?

SPEAKER_01 2:52

Well, you know, that that's a little bit puzzling even to me. I think I'm uh an expert on policing and the law. That was obviously a big concern when the Biden administration came in in 2020. And people in the White House had reached out to me early on, and as they were formulating the executive order on policing and criminal justice policy. And that conversation continued. Some of the uh policy advisors there used me as a resource. And then when someone was leaving, they asked me to step in. And initially I couldn't do it for personal reasons. Then they came back around to me another time. And by then I could. And so my youngest had gone off to college, and I thought, okay, you can't cart from the sidelines forever at some point to say, ask you to participate in the policy that you've been talking about for years, then you've got to go do it. So I had just written an article on the federal influence and local law enforcement. It felt like it couldn't say no. And so I went and uh served as uh on the domestic policy council.

SPEAKER_00 4:04

So, first overarching big question is is just like, what is it like working at the White House?

SPEAKER_01 4:10

It's actually an incredible institution in many ways. It's very, very different from my experience working at the Department of Justice or and other even policy work. I before I went to law school, I worked for the Manhattanboro President in New York. I had done policy work there. But it it's just such a different institution. First of all, the place gets built every four years from the ground up. And so it's it it's a um uh um really challenging thing to do. It has a vast uh responsibility and and and this, you know, if you think about the portfolio of the White House, it's just completely vast. And yet the personnel, you know, there's really just uh not that many people working there. Uh if it's a thousand, I'd be surprised. I don't know exactly the number. And so it it has this incredible responsibility and intensity. And the, you know, when I first got there, I didn't even know what my job was, and nobody had time to slow down and tell me. I mean, it's just the things are, you know, somebody compared it when I for when I first got there, it's like you're it you're it's a crew race, and all eight people in the boat are rowing as hard as they can, and someone jumps out of the boat and you jump in and you just have to start rowing. And so there is just it the the speed at which the White House works and the intensity of the responsibility, it's and and the things that could happen on any given day. So you're pursuing both long-term policies and then short-term responsiveness is really quite incredible. And especially from my stayed office at the University of Virginia where things moved pretty slowly. I mean, of course, I'd been a prosecutor for years and I had done trials. I I know what intensity feels like, but this really felt like a, you know, the the emails would start to pick up at 6 a.m. and they wouldn't finish up until 11 or 11:30. And you couldn't be, I mean, I I must I I did a lot of cycling before I got to the White House, White House. In fact, before I went there, I just finished a 700-mile ride from Portland, Oregon to Missoula, Montana. And then I got to the White House and didn't get on my bike again for a year because I couldn't be away from my laptop for an hour. I mean, and and to be at a cell service was just impossible.

SPEAKER_00 6:36

You have to learn to ride your bike with your laptop sort of on the front bar balanced, I think is the is the ethics.

SPEAKER_01 6:41

I watch movies, but I'm not that good at the at the at the responding to emails on my bike.

SPEAKER_00 6:46

All right, fair enough. How how do you compare having worked at the White House to having clerked with the Supreme Court? Is there any comparison? How do you how do you conceptualize those two experiences?

SPEAKER_01 6:58

Yeah, so that they are extremely different. Um the White House is responsive in a daily way, in an hourly, minutely way. I mean, it's just so in the world. And the work of the Supreme Court is much more insulated from both public view, but also it it's slower and more thoughtful, and the justices choose their cases, and then it takes months to brief them, and then they get argued, and then it takes months to write and draft the opinions. And so it's just sort of a time frame on the two is very, very different. I also think that it's not to say there are, I mean, there are all sorts of things to say about politics and the Supreme Court, and I am hardly an expert on them, but it is not fundamentally a political institution in the same way that the White House is. And you feel that every day. So power within the building, you know, in the White House, there the president and vice president are elected, and the president, everything really funnels down from that. Everyone else in the White House has a job because the person above them has a job, and the person above them to the president. And so there's just this looking up sense of, you know, responding to the leadership that it is so it's both power within the building, I guess is another way to say it, and the responsiveness to the world, that it just feels very different. You know, that each Supreme Court justice is um they interact to be sure, but they're each an island. They each have a constitutional responsibility that's separate. They each have their own staff, in addition to the collective library and other staff. And they each make decisions, um, certainly in interaction, but they're kind of you know more independent than anyone in the White House is you all working towards one die.

SPEAKER_00 9:07

Getting a little more granular, what what specifically is the domestic policy council? Because until we started interacting, uh you just kind of assume there's this monolith and there's a person at the top and they have underlings arranged in some sort of random pattern and they do the thing. But there's a there's a hierarchy in organization to it, which makes sense. So I'm curious, like how how would you summarize the domestic policy council and more specifically, what is the role that that you played and the policy council plays in sort of setting criminal justice policy at a national level?

SPEAKER_01 9:38

So the domestic policy council exists to help the president form and implement his domestic policy across a whole range of issues. And so that means it's a very uh diverse internally. You know, I shared an office with somebody who was working on disability policy, and I worked every day with people who were doing the national anti-Semitism policy. And so there were people doing all sorts of things that had very little to do with the stuff that I do, which is crime and criminal justice. So the policy council as an office is diverse. It is led by the domestic policy advisor, and the domestic policy advisor is one of several senior advisors to the president who help formulate and implement policy in different areas. And each of the councils is organized differently. So the domestic policy council doesn't look like the National Security Council, which is or it has obviously a different portfolio, but it's also organized differently. We were helped the domestic policy advisor help the president uh form and implement policy. And that meant both, given President Biden's priorities, that meant both crime policy and it also meant criminal justice policy. So he, his executive order, for example, on effective and accountable policing and criminal justice was intended to pursue policing and criminal justice policy that both worked to make communities safer and also that was responsive to the law, to uh community concerns about the ways that policing and criminal justice is carried out.

SPEAKER_00 11:29

And are there there's subject matters? I don't know the degree to which you could talk about it, but the the things that you worked on. Are there subject matters that that you helped push along the line, specifically things that your academic research helped to sort of inform the policy going forward? Is there anything that you worked on that you're particularly proud of that like you were, this is a thing that I did that suddenly was a a thing nationally?

SPEAKER_01 11:52

Well, so I I will give all credit to the president for any policies that we helped implement. But what I would say is that I was glad to be there in the moment I was there because I think the the events around 2020 and the death of George Floyd and the interest in reform and accountability and policing that grew really from the ground up and that involved police chiefs all over the country trying to be responsive to communities and build trust. And though this was not, you know, this wasn't didn't start in 2020. You could think of the events of 2015, around Ferguson and and and beyond that drew attention. But President Biden's EO, I think it is probably the most important effort at national uh uh law enforcement reform ever in this country. And I came with a lot of expertise in policing and the law. So that was a really exciting thing for me to be a part of implementing. You know, from the outside, I had sort of contributed a little bit. And then from the inside, I got to work on trying to make that happen. And, you know, one of the things that's not at all obvious from the outside is how hard it is to implement policies so the president can announce whatever he wants, but that doesn't make it so. It's not self-executing. So it means working intensely with agencies like the Department of Homeland Security, like the Department of Justice, to get the actual things done that are in an executive order. And some of them are really, really challenging. So one of the things I worked on a lot was try helping get the national law enforcement accountability database, which we called NLEED, National Law Enforcement Accountability Database, off the ground. And that database is intended, was intended. It was the executive order had told the agent to the Department of Justice to create it and federal agencies to contribute to it and try to promote it beyond the federal government to state and local agencies. And it was intended to collect information that is useful to law enforcement agencies in hiring new personnel. So things about previous terminations or decertifications of law enforcement officers by state agencies or civil suit judgments against officers that might be relevant to hiring decisions. And it turns out to be far more complicated than anyone imagined to create such a database and have federal agencies contribute to it in a way that law that agencies could then use, or to create a database for local state and state actors, which did not get off the ground by the time the the end of the administration came. But it was that kind of thing where I had a lot of experience on law enforcement accountability issues that I felt like I could contribute something substantive. And on the crime side, the administration came in at a time when crime had just experienced the you know murder rates in 2020. I learned everything I ever knew about crime data from you. And murder in 2020 was at its highest, you know, went up its highest one-year increase ever. And that's how the administration started. And we had to respond to that. And so, even as we're trying to make policing more accountable and it effectiveness and support for crime prevention and control in communities was absolutely essential. And so that I had done a couple of years ago, participated in the National Academies of Sciences work on proactive policing. And so I felt like I was kind of up to speed on the research on some of the strategies that are effective in crime, control in terms of law enforcement activities. But of course, there was also attention to lots of non-law enforcement activities. And this was the administration invested more, really the first focused investments in community violence intervention ever, and other ways, you know, mental health services, other ways to try to prevent and address crime beyond law enforcement. But of course, you know, I came in focused on crime and law enforcement, mostly through a law enforcement lot.

SPEAKER_00 16:08

And as you're making these sort of or trying to make these policies, what are the biggest hurdles that you face, especially like coming in, murder's risen the fastest ever? It sort of as as you it wasn't really until towards the latter half of your time in office that it was really clear that it was falling and falling fast. How do you respond to these, these big national issues and sort of what are the biggest challenges to doing so effectively from the federal government?

SPEAKER_01 16:36

Yeah, so it's funny, I felt very much like as an academic and as someone whose career did not depend on anything that happened in the White House or on anyone liking me or liking my opinions. And you know, if they didn't want to hear from me, they could just say I could just go home. I there was no sense that I had to be there, I had to contribute. And I feel like that gave me a little bit of privilege and and perspective as an almost outsider, now insider at the White House. And what I found was that it it can be very hard for the White House to sort of quickly respond to current events. So you're you know the the speed at which things go and the vast number of things that the White House has to deal with means that you don't always have lots of time to stop and think, okay, what are going to be the most effective long-term strategies to address this problem? And you don't have a long clock, right? You're you're on a four-year clock, which means whatever you're gonna get done, if you're you're either you're gonna have to make the case for it at the very least, um in a midterm election or in another presidential election. And so there's this very, there's a little bit of a conflict between the long term and the short term. And then there's the collecting enough information and having enough time and space to make good policy and then sell that policy both internally and externally. So, in terms of formulating policy, and and we could talk about crime data and the role crime data plays in that, but formulating policy is not so easy. And then once you have the policy in place, the president decides what he wants to do, he issues an executive order. You have to get agencies on board and and not just agencies, but outside actors who can contribute to the policy agenda and to implementing policy. And that turned out to be extremely challenging. And it's not because people are obstructionist or because, but because they have jobs too, and they have incentives and they have, you know, the Justice Department has to do a lot beyond implement crime policy of the White House as it's responsible for law enforcement in a way that at least should be and was in the Biden administration, independent of the White House's political interests. It it it you know, it operates a vast bureaucracy. Um and you have to get you have to align the gears to make a priority in the agencies the things that are priority in the White House. And I I I found, I mean, I have utter respect for the actors in the Department of Justice. Um, but that sometimes was challenging. I mean, for the White House and the Department of Justice to get aligned. And in under the Biden administration, the Justice Department was treated with a respect and a distance and to in order to preserve its independence. And that was very important on the crime enforcement side, but that distance, which was created to insulate that decisions and assure the American public that they were neutral and fair, could sometimes bleed over to things like the policy agenda on funding priorities, which isn't, doesn't require the same level of independence or insulation. So I don't know, maybe that's not a real really great answer, but I would say there are political obstacles, there are climbing obstacles, there are agency obstacles, there are personnel obstacles, there were lots of obstacles.

SPEAKER_00 20:18

And since you brought this up, what role does data and and specifically, you know, the I mean the whole podcast is about this lack of perception of what's actually happening with this issue? How can data sort of overcome that in the policy-making process? And and in your experience, was it mostly a crime is out of control, we need to do something about it, or was it at a hey, things are, you know, here's where things are sort of leveled off and now it's starting to fall. These are the policies that might be working. How did data play a role in any of these decision making?

SPEAKER_01 20:51

So, for one thing, the the the you know, one of the biggest challenges for the White House is if you think about the White House as a short-term enterprise that has to be responsive on a daily basis to the American concerns. Crime data operates in a very different time frame. And the so your crime data aside, and that's pretty new, that's that is post to the administration or at the very end of the administration. But the, you know, one of the things we struggled with is the American public's concerned about carjacking. It seems like there's a rash of carjacking in DC, and carjacking numbers look like they're going up. But we don't have national prime carjacking data. And even if we did have really good national carjacking data, we wouldn't have it for 18 months. And that's not that's very, very challenging if you're trying what you're trying to do is say, how big a problem is this? And what are the and what kinds of strategies might we use to intervene to stop it? And yet you can't ignore carjackings because you don't have the data. Um, it I think of carjacking. Checking is actually in what one of the crimes that most affects Americans' perceptions of safety. It's a pretty rare crime. It doesn't happen in vast numbers, but but it's also one of the most violent random crimes, right? It mostly doesn't happen between people who know each other. It's very unlike assaults and gun violence. It and it's sudden, it's scary, it can be violent, it often involves a gun. And it so people feel it shows them that things are out of control in a way that few other crimes do. And yet we have very little good data about it. And that so responding to that concern, you know, the Washington Post reports about a rise in carjacking, and everyone in the White House reads that report from my boss, the domestic policy advisor, the president, everybody. Um, and so it's impossible not to respond to those concerns. And yet it it we're often hamstrung in our ability to do so um by the absence of data.

SPEAKER_00 23:08

It's really funny you mentioned specifically carjackings, because one of the projects I'm working on now is with NIBER's data, we can they the carjacking is defined as a robbery where a car was stolen.

SPEAKER_01 23:21

Yes.

SPEAKER_00 23:21

Yeah. And we can actually pull that data. And because the FBI is running these things monthly, they're publishing new data. We actually suddenly are starting to have decent carjacking data. And so what I'm doing now is I'm sort of building an AI that's scraping all of the data, getting the master files from the FBI. And if you want to know how many carjackings were there in Los Angeles, it'll pull that data from the FBI and have it through, you know, Los Angeles through January 2026 right now, which is the most recent like sort of updated data. I guess how, because this all of these structures are improving so fast, how do we get people, and I don't know if you have any insights from your work, like how do how do we get people to understand that like the infrastructure, our awareness, our ability to understand crime trends is changing. This is this is much less of a like, how do you do it from the White House? But like, do you have advice from your experience of someone sitting in my seat that's like, we can actually answer a lot of these questions fast, but nobody knows about it?

SPEAKER_01 24:21

It's a very long-winded question for No, I so I think the work you've done here is critical. As you know, I'm a big fan of your work. I I follow your Substack. I I I used you as a resource when I was out the White House. I needed to understand data. I I I think you're doing critical work. And I think one of the things you're doing, which I hate to do, but you do very, very well, is translate that into things like op-eds and responding to media calls and the Substack and to try to help people see both what the data says, but also where the data comes from and how to understand what it means, the difference between, you know, the uniform crime reports and NIBERS and the victim survey and the like. I agree with you. One of the things that is amazing is, you know, I was in the White House, I don't know, what was it two years ago? It wasn't that long ago. And what we know about crime is very different than it was then. So in the two years since, we have much more current data than was available to the White House at the beginning or middle of the Biden administration. And so what do we do with that, or how do we convey that we have that? I think one is in making you know resources available to political actors at all levels. And that is partly a publicity issue and partly it's a it's an experience. Um I think politicians learn in part from politicians and from law enforcement and making, you know, they'll use the data that they that builds trust and that is reliable and that's consistent over time and that is produced on a regular calendar and that gets press attention. And then and then they start to say, oh, I get it. I can rely on that data, and that helps them in their policy making. I think also, you know, we shouldn't, it's so crime data is obviously extremely critical, but crime data in some ways is our best case scenario. There's lots of other data that we don't have that would help us in making communities safer. So we have very little data about what law enforcement is doing, you know, what strategies they're using. We know a little bit about arrests from the arrest data. We don't know anything about criminal charges. There's no national data on that. And then we don't know how many people are convicted or how many charges are dropped or why they're dropped, or, you know, there's just a lot we don't know about the criminal justice system and the criminal process. Though, you know, so crime data is getting better in some of those areas. Data is no better than it was 20 years ago.

SPEAKER_00 27:09

Is there a way to improve those data sets in your experience?

SPEAKER_01 27:13

Yeah, I I mean, I I think so. Like one of the efforts that the Biden administration did was to start a use of force database and the National Law Enforcement Accountability Database. Both of those are resources that agencies in theory could use to check their own performance and also to give us some handle on some of the costs and benefits of policing and how to make policing better. That's, you know, we need national data. We don't even have national standards for data collection, for example, on the use of forests. Something that's of vital importance to understanding how police are interacting with the community and also that affect community trust in policing, which in turn affects, we think, the effectiveness of law enforcement and getting cooperation and the like. And so that seems important, but communities can't even check themselves. If you're a police chief and you say, hey, I really want to know in LA how I'm doing compared to other cities, you would need national data standards on the use of force and some national way of collecting and distributing that information, ideally. And we have that in some other areas. We have that in the National Crime Information Center or the DNA database or fingerprints, but we haven't done that as much in criminal charges. We haven't done it in sentencing, we haven't done it in um uses of force and other policing data. And as a result, I think we're we're a little bit, you know, we're operating half in the document.

SPEAKER_00 28:50

Yeah. Response times is the one that always comes to mind is we have no idea what's good and what should be standard and how fast should a police department get to an emergency. And and yet each department sort of collects it. And if it's good, they say about what it is, and if it's not, they don't.

SPEAKER_01 29:05

So yes, but you know, one of the the most important things about policing in America is how local and diverse it is. So by constitution and history and tradition, policing is mostly a local phenomenon. And local conditions, at least around response times, are really different. If you're talking about an agency that operates over a vacuum, I mean, if you if you look at agencies in Alaska, it could be hundreds of square miles. And New York City is the very, very few square miles. Obviously, response times in different places might look different. And so it's not that we're always going to say, oh, well, you have a worse response time than this agency, therefore you're doing a bad job. And the same thing's true of crime, and the same thing's true of use of force, and the same thing's true of lots of other aspects of um safety and disorder. You know, homelessness looks different in different cities. Um, car crime looks different in different cities. Well, you know, I grew up in New York. When I first went out to LA, I had no idea that cars could play such a role in American society. I mean, I who knew? It looks very, very different. So there's a lot of diversity, and that I think is challenging to policy. I think one of the hard things about the White House, this is back on the White House issue, which is that people really don't realize is of course people hold the president responsible for national policy and national and national conditions. And it's not that I think we shouldn't, but we should also recognize that in some areas, in some areas, the federal government really is in charge, and they are the primary determinants and implementers of policy, and they should be held accountable. Immigration is one of those. But crime looks very, very different because the federal government isn't the primary actor here. There are, I don't know what it is now, but it might be 17,000 law enforcement agencies in the United States, most of which are local agencies and sheriff's offices. And what the federal government does is offer them funding and collects and distributes information to them and partnerships like joint task forces and crime gun intelligence centers and technical assistance and enforcing federal law, which includes both criminal law and civil rights law, and acting as a role model, you know, by getting its own house in order. And that's very different from, you know, if you talk to a police chief and a mayor, they're in charge of the policy. Now they're constrained by state action and federal action, but but they they have a much more daily effect on the quality of lives in that community.

SPEAKER_00 31:59

So, like if you go back and had a time machine and you went back like seven to ten years, you'd see criminal justice policy and criminal justice reform as one of those things that there's maybe a path forward in a bipartisan sense where the parties can come together and actually make policy change. It feels like that has evaporated in the last three or four years. Is that something that you think is possible, or do you just do you disagree with that assessment that I've sort of assertion that I've made?

SPEAKER_01 32:28

Well, so I think bipartisan work is more challenging today in many issues, not just crime and criminal justice policy. So, you know, do I think this is so it might not be uniquely challenging, but yeah, I do think that there, you know, we had the the President Biden signed the bipartisan Safer Communities Act that expanded background checks and funded state red flag laws and expanded mental health services and community violence intervention. And you know, the it the he signed the American Rescue Plan that led to the biggest federal investment ever in public safety by states and localities. And that included hiring police officers and but also supporting community violence intervention and mental health professional responses to mental health crises and things like that. And so I think we did have a period in which there was a sense that we could work together, and that has faded some. But I don't have no hope. I I'm not confident under the current administration that bipartisan efforts on crime and criminal justice policy will thrive. But I also think that dropping crime rates gives us some runway, which is not only have crime rates, I mean crime rates dropped during the Biden administration and biggest, I don't know, or biggest and 50-year drop in 2023 and then a bigger drop in 2024, if I remember correctly. That started, you know, a trend, which is, as you point out in your work, continued in dramatic fashion. And so now we have this, we do live in a safer America. I don't know that everyone experiences that way because they still, because actually the conflict and disorder and conflicts over things like immigration policy create a sense of fear that probably pervades people's experience of crime and as well. But I do think that that crime drop, especially if it's sustained, could help us get together on criminal justice policies and crime control policies and re-entry policies that are both supported by the evidence and that are likely to actually make things better. And so I don't have any hope. In fact, I think lower crime gives us room to start identifying policies that we could implement over the long run. And they'll face less opposition when people don't feel like they have to respond to the rise in car thefts by cracking down on kids or shoplifting, you know, videos of shoplifters. We could talk about, I mean, I became obsessed at the White House with car crimes, including car thefts, carjacking, and thefts from cars, especially of guns, and then also shoplifting and organized retail crime, which was a a huge political issue at the time and now has faded in significant part because people are paying attention to other things. And of course, we have no idea whether it went up or went down then or now or anytime, uh, because we have no data on that. But we do know that it was of significant public concern and now maybe less so, which might give us a way to say, okay, so what are good policies at the state and local level? And how can we look at experiences that have worked well and share them? And, you know, so I don't know. That's a long-winded way of saying, I mean, I'm an academic, so of course I'm long-winded, but the the it's a way of saying, you know, I don't have no hope because I think people do get together on smart on crime, smart and criminal justice policies when they are not under the gun.

SPEAKER_00 36:24

I'm going to give a quick plug to the FBI again. They have a really cool data set. They did a report a couple of months ago on this on was it flash mob shopliftings. You take NIBER's data, you find shoplifting offenses where there were six or more offenders in only one business. And they're calling the, so those that's what they're calling the flash mob shopliftings. Obviously, you have the reporting issue, so we can't really tell whether it's going up or down because that shoplifting is sometimes reported and sometimes not. But it's really cool that like conceptually we can start to think about measuring those things. So again, uh you know, giving a plug again to the NIBERS discussion.

SPEAKER_01 37:00

I love it. And it shows you want to move to NIBERS, which was like a technocratic policy issue. Right. So it was it was terrible to do. It was really hard for the agencies. It's still hard. You're getting local, you think about the lift that it took to get NIBERS underway. And obviously, it was not a smooth. I mean, you could talk more about this, but it was not a smooth path. And yet think about what it has empowered us to know now. I mean, you're describing things that we could not see in the uh, you know, from crime report data that we now can see in NIBER's data. And hopefully that will get better with time as we refine that. So I that I I I definitely give credit to the FBI. I mean, I think the lift, you know, nobody was celebrating their efforts in this regard. Nobody was, you know, saying to the director, you know, what you've done a great job at, is do doing this massive lift, which is yeah, yeah. So I I think that, you know, data isn't sexy. Information collection isn't sexy, evaluation isn't sexy either.

SPEAKER_00 38:09

You know we're gonna have to agree to disagree on all of those points.

SPEAKER_01 38:12

Well, it is to me, of course. And it is to you, I know. And yeah, I think you've made it sexier, actually. I think you've done a good job at that. Uh yeah, you've definitely made date cron data hotter than it used to be.

SPEAKER_00 38:25

That is that is a that is the plug for my newsletter now. You can call it making cron data hotter since 2022.

SPEAKER_01 38:34

I I think that's absolutely right. Um, but yeah, I that's a really interesting area. I mean, the so i i if we though it's true, I I would the one thing I would say to you is that when you're thinking about the flash mob issue, and that of course it's true mo that shoplifting is an under-reported crime, and it's partly under-reported because the in incentives to report are very limited. But I bet flash mobs, which is to say shoplifting events involving six or more people, are more consistently reported than individual shoplifters. And so uh you might get slightly better data there, especially once it's publicized that there is a data set, because then there might be an incentive. And one of the things I got to do is talk with the some of the retail organizations that were concerned about retail crime and what they were doing and how trying to coordinate with local agencies and helping facilitate that. And that was just fascinating to see people responding to something that was both of concern to businesses, concern to local law enforcement, and definitely it with the rise of social media, concern about it, concern to the public.

SPEAKER_00 39:57

So have you given any thought, because I guess the craze hadn't really hit quite as much as it has now from AI when you were at the White House. Have you given any thought to how that would change your job if you were starting a new stint there, or even how is changing your job sort of as an academic working in this world?

SPEAKER_01 40:14

Yeah, it's already influencing certainly the work I see and how people are thinking about research and data and information and conveying information. I worry a lot about AI. There was certainly concern about it when I was at the White House, but it was not, you know, we weren't where we are today on AI. And I've seen a dramatic change in when I use, for example, ChatGPT. It was really bad at law six months ago. If you asked it any legal question, it could not come up with a plausible answer. And it's gotten significantly better. I run all my exams through AI to see through ChatGPT to see whether, you know, what I'm likely to get from a student who used AI in an exam just to try to figure out like what's going on in the world and also to see how good it is at doing the things that lawyers do. And it's still got a ways to go. So I'm not totally persuaded. I am, in terms of the work on crime and criminal justice, one problem is that it has always been true that social media, both you know, if you look at body camera footage from officers, or if you look at cell camera footage from the members of the public, that has been very critical in helping people understand policing in a way they didn't. So this starts with, you know, in 1992, with Rodney King, the video of Rodney King's beating. And that's really the first time that Americans had their views about law enforcement challenged by video evidence. That it took a while before cam cameras were common, but even when they were, you know, YouTube didn't yet exist yet, phones didn't exist yet, and the American public had relatively little and relatively mediated information about what law enforcement looked like. That all changed not only with the invention of the iPhone in 2006, but with the invention of YouTube, which meant that it wasn't media companies that were distributing information about what people were seeing, but it was people themselves could share it. And of course, that's gotten much, much more expansive. I am actually not on social media. I'm probably one of the last people in America. Not on social media. So I but I but of course that because I work on policing issues, I do get that uh that people share those videos with me. And the big change, this has been a radical transformation in our understanding of policing and really an opportunity actually now with national with natural language processing in AI to start turning body camera footage into data about policing. You know, we had this new opportunity, but AI actually threatens that as well because people can't trust the video they see anymore. And so, of course, it's always true that it could be edited, it could be altered, but now we're in a different world around video. And I think as the distrust for video goes up, we we it feels like we had this heyday where suddenly we could see a lot about crime, both what crime looked like and what criminal uh criminal enforcement looked like. And now we're gonna be in this world where we're we don't know again. So we we it gives us better data, but I think it also threatens what we know.

SPEAKER_00 43:53

Of course, we had Ian Adams on, who talked about AI and policing specifically a few weeks ago and talked about that case in Utah where The body worn camera picked up the Princess and the Frog movie and then the draft one software picked it up as oh, the officer transformed into a frog. And that was and they and nobody checked the report. So that's a an outlier example, but certainly a a amusing anecdote of what could actually be the worst case scenario. So so Rachel, what's what's next for you?

SPEAKER_01 44:23

Well, I'm as I have been, I'm back in my office at the University of Virginia. I'm working on papers on criminal justice and including policing. I have a paper coming out this year with a philosopher and uh legal philosopher Kim Frizan at the University of Pennsylvania about how we should think about innocent people in the criminal process and how we treat them and also whether we should consider compensating innocent people who are are stopped or arrested or charged or released because of the they're really doing a kind of public service of the same kind that jury service is, or that being a witness to a crime is in court because they did nothing wrong and yet are caught up in the criminal process. And so I have a little work in that vein going on. I have a bunch of projects on policing and police accountability. I continue to do some consulting or advice giving to nonprofits and agencies, government agencies about policing issues. And, you know, I'm trying to have fun.

SPEAKER_00 45:26

That's good. That that's the goal for everyone, right? Rachel, thank you so much for joining. Uh, this has been a wonderful conversation. You've given my newsletter a new tagline. So you know, very, very useful, very interesting. And uh, I really appreciate you coming on.

SPEAKER_01 45:40

Jeff, I really want to thank you for the work you've done in this area. It was important to me as a policymaker, and it's important to me as an academic. So thank you.

SPEAKER_00 45:49

Thank you. That's lovely to hear. Thanks for listening to the Jeffalytics Podcast. Be sure to subscribe and to learn more, head on over to ahdatalytics.com for more information and previous episodes. If you like what you heard, please leave a glowing review, which will help others to discover the show. Until next time, I'm Jeff Asher.