fbpx Social Impacts of Police Violence with Desmond Ang | Harvard Kennedy School

December 2, 2021

How do acts of police violence affect diverse communities? Listen to this Wiener Conference Call with Desmond Ang, assistant professor of public policy, to hear answers to this question and more.

Wiener Conference Calls recognize Malcolm Wiener’s role in proposing and supporting this series as well as the Wiener Center for Social Policy at Harvard Kennedy School.

View slides from this presentation.


- [Announcer] Welcome to the Wiener Conference Call Series. These one hour on the record phone calls feature leading experts from Harvard Kennedy School, who answer your questions on public policy and current events. Wiener conference calls recognized Malcolm Wiener's role in proposing and supporting this series, as well as the Wiener Center for Social Policy at Harvard Kennedy School.

- Good day, everyone. I am Mari Megias, in the office of Alumni Relations and Resource Development at Harvard Kennedy School. And I'm so pleased to welcome you to this on the record Wiener Conference call, which is kindly saved by Dr. Malcolm Wiener, who supports the school in this and many other ways. Today, we are joined by Desmond Ang, who is assistant professor of public policy at Harvard Kennedy School. Professor Ang is an applied economist with a particular interest in the intersection of race, government and media. His research examines, the social impacts of police violence, the effects of voter protections on turnout and polarization, and the role of media on racial prejudice. We are so fortunate that he's agreed to share thoughts today with the Kennedy School's alumni and friends. Professor Ang.

- Thank you so much, Mari, thank you so much to everyone for coming and taking part in this. As Mari mentioned, today I'll be talking about some of my research and specifically, I'll be focusing on my research thinking about the social impacts of police violence. So as a bit of motivation to the topic, which may not be necessarily, given the state of the world. Obviously after the police murder of George Floyd last year, there's been a lot of the discussion, public discourse, political discussions around police officer use of force, officer use of force, police violence. You know, how these events occurring, what can we do to sort of mitigate some of these events? And I think it's important to emphasize the extent to which this current moment that we're seeing the police killing of George Floyd, the police killing of many several individuals each year in this country, including a number of unarmed minority individuals is the product of countless historical and explicit policy decisions that we've made about what policing should look like in this country. So that includes decisions about where we should be policing, who we should be policing, as well as how we should be doing those things, right? And so these decisions and what we're seeing today aren't made... You know, the circumstances that we see today, aren't doing the current vacuum. They occurred as a result of how we think about policing in this country. And that in itself is gonna be informed by research and empirical research and academic research as well. And, you know, as a researcher here at the Kennedy School, as an economist here, that's sort of the explicit role of what we're trying to do, the end goal of why we do this research is to make... To help inform policy makers, to make better economic decisions, social decisions, political decisions. And so for today's talk, I wanna do two things. Hopefully, we'll have time to do this. The first is obviously to go over some of my own research more generally, what do we know about the effects of police on local communities from the literature as a whole, but I also wanna spend some time taking a step back and thinking about what we know about police, the questions that we asked about police and how that in turn is gonna inform policy decisions is very much gonna intersect with representation and diversity in academic institutions. So I think both of these things hopefully will come out from this talk. So if we look at sort of, the research landscape historically, we can see what questions have empirical researchers asked about the effects of police. And so as sort of a toy example, we can just go to Google Scholar and say, "Okay, what do we know about the facts of police?" Here, I'm just looking at articles that were published before 2005. And these are the articles that come up. These are the sort of the first five articles that come up when we think about, what does academic literature say about what police do? What impact they have on communities? And you'll see, the first article is titled, "Using Electrical Cycles in Police Hiring to Estimate the Effects of Police on Crime." This is by an economist, Steven Levitt at UChicago. And essentially this article, looks at the affects of the police on crime deterrence. The second article was actually very similar in theme, "The General Deterrent Effects of Police Patrol in Crime Hotspots." Again, thinking about the effects of police on crime. Third articles again, the same, the fourth articles looks a little bit different, asking, you know, whether the determinants of how stressed officers are, how much work they're doing their relationships. And the fifth article again, comes back to this very common theme, which is thinking about police and how they impact crime. And so obviously this is a very important question, perhaps the very first or their question about, what do police do is thinking about how they impact crime? And so you'll see... We've seen, you know, just looking here, two things emerge. The first is obviously there's this very well-defined and somewhat narrow research question that researchers have asked historically about, what do police do? And we're asking that really in the realm of thinking about how police affect crime. The other thing that emerges that is perhaps less obvious from just looking at this page is that, there's actually a very narrow author set of the types of people who are asking these questions. These articles are all written by white men, right? And so you can see that there's been... Historically, there's a very narrow set of questions that we asked about police there's a very narrow set of people asking those questions. And that's again, not to say anything about the quality of that research, that research in general is very well done. If we look through the research, we find one generally a true consensus, which is that, on average, as we increase more cops, areas that increase more cops have less crime. And so this is a causal claim that we can say with some confidence. And again, it's relatively agreed upon in the research, and that has led to... And that has had implications for policy. And so here's one example. This is article by Matthew Yglesias in "Vox," saying, "The Case for Hiring More Police Officers." This is a chronic finding idea that actually works. Again, the research shows that more cops equals less crime and that's something that's relatively popular across all racial groups. And obviously, this is a pre George Floyd article, but even in recent times we've seen, sort of this discussion of returning to this idea that, now maybe we should have more officers because that's gonna reduce crime. And again, I'm not saying anything, you know, that conclusion is relatively well founded in empirical literature that we have. And we can see this, you know, these research findings translate and reflected in actual policy. So if you look over the past 50 years in the United States, we look at a number of police officers that we employ in this country, as well as the rate, the number of police officers per capita in this country, you'll see, there's a huge increase in that over time. This is true, both in big cities and in all cities. And we've also seen a growth... Coincident with this growth in number of police officers in the types of intensity of policing that we do, there's growth in this sort of proactive, broken windows policy policing in this country. So here you can see in New York City, during the 2000s, this is large increase in stop and frisk in New York City. This explosion in a number of sort of street stops that were occurring, in which police would just stop people on the street to check if they had guns or not. Before the court mandated end of stop and frisk due racial discrimination. We can clearly see those as large increase in stop and frisk rates, large increase that's specifically targeted towards black and Hispanic communities. We can also see this research translating then into outcomes, into large growth in our incarcerated population. So here, we're just looking at incarceration rate in the United States over time, going back to 1880, you'll see over the past 50 or 60 years, there is an increase in incarceration rate of something like five times, like a 5X increase in the incarceration rate. The U.S today has more people incarcerated than any other country in the world. It also has a higher rate of people incarcerated than any other country in the world. This is not just with regards to other Western democracies. You can think of any country, the U.S is gonna beat them in terms of having people in incarcerated. And again, this is something that has led to large increases in racial disparities. So here, we're just looking at incarceration rates by race, you'll see large differences in likelihood of being incarcerated. Specifically, black individuals are just far, far more likely to be incarcerated than white or Asian individuals. Also higher rates for Latin X, native American, and Hawaiian individuals. And all of this also is to inform what we're seeing with regards to police violence. And so if we look over the data, if we look over across a of different data sets, there's not one sort of like, standard for thinking about... For measuring how much police violence are charged in this country. But if we look across all of these things, you'll see there's been a large increase over time. And the number of people each year who are killed by police, nowadays, it's something like 1,000 individuals each year are killed by police in this country. And that's generally stable in the recent past. It's also generally stable that most of these events are gonna involve racial minorities. A large share of them, something like 40% of these events involve, the police killing of somebody who didn't have a gun. And again, it's very important to emphasize the extent to which this is a explicit outgrowth of policy decisions that have been made. And that officer use of force in the case of George Floyd, and in many other cases is a product of decisions around how much destruction we afford the officers. And across the United States, less than half a percent of officer involved killings result in an officer conviction. And all of which has emphasized that use of force is very much sort of a feature, but we're seeing with George Floyd, we're seeing with other police killings of unarmed minorities, and more generally, use of force is very much a feature of how we do policing in this country. Not so much a book, okay? And again, you can imagine that it's very much an outgrowth of this research where we're saying, "If the primary goal of police is to reduce crime, well, then we should have more policing, we should have more intensive policing," et cetera. At the same time, I think... And I'm sure all of you will have different definitions on what the purpose of policing is. But I think most of us would think that it's a far broader goal than just reducing crime. So here's one definition provided by the Charles Koch Institute, saying that, "The purpose of law enforcement is to promote public safety and uphold the rule of law so that individual liberty may flourish." And so crime reduction is not so much the end, but the means to the end, to promoting safety, to ensuring community and individual wellbeing. And so in this lens, crime is, you know, just one input for public safety and community wellbeing, obviously an important one, but it's not the only one. And we can sit here and think of a number of other ones that we might imagine to be very important. Some of these things might be things like, procedural justice, your sense of justice, your sense of fairness in the criminal justice system, others are trusting government, perceptions of safety, social cohesion, economic wellbeing, psychological wellbeing, et cetera, et cetera. These are all things that we think policing might impact. And that we think police may have a role to play in sort of improving these outcomes as a way of, you know, benefiting public safety and community wellbeing. These are all different aspects of those larger concepts. And so, thinking about policing only with respect to crime is a relatively narrow frame for thinking about the impact again that police have on community is much less for determining how much policing we should have, what types of policing we should have, et cetera. And this isn't something that, you know, we just are sitting around theorizing in the wake of George Floyd. There's been a long history of public discussions about police, about use of force, about how police actions can impact a number of things that aren't just current. A number of very important margins, societal margins that aren't just current. So you can go back to the "Turner Commission Reports" in the 1960s, in which the federal government was concerned about large amounts of social unrest in this country. And so they investigated that. And one of the conclusions they came to was that, some of the social unrest was driven by a widespread belief among African-Americans in existence of police brutality and a double standard of justice protection. So again, this idea that policing might have some implications for how we view a discrimination, how we view justice, how we view the criminal justice system. More generally, we can go back even further to the "New York Times" this from 1850s. There's editorial about a police killing of a known criminal named Hollis. And in this editorial, you'll see the "New York Times" editorial board says, you know, "This guy, Hollis, he was no joke in some sense, a notorious and most pestilent ruffian," he clearly deserved to die. But then they go on to say, "Nothing can more forcibly illustrate degradation into which local criminal has fallen. It's bad enough to have a gang of lawless uncontrollable scoundrels among us, but it's much worse to have their lawlessness imitated by the man who's special duty is to enforce the law." So again, this idea that... This is a very long running idea that police might have some impact on how we trust the government. The policing and law enforcement, obviously one of the most visible forms of government and our interactions with them, what they do can inform how we actually view the government, how we view other institutions as well. In more recent periods, you can see that all the largest urban protests that have occurred in the United States over the past 75 years were all sparked by acts of police violence. So we saw this last year with George Floyd, we saw in 2014 with the Ferguson riots, I can think of the Rodney King riots in the early 90s, you know, a Watts riots in the 1960s, all of these incidences of large social unrest, driven by actions that police conducted. That being said, the social impacts of these events, thinking about how police affect these other margins of community wellbeing, just were never really rigorously examined in the academic literature until very recently. So it was only recently that research that started to really interrogate these broader questions about what do police do, what impact the police have on communities? And so, we can run the same Google Scholar search, thinking about the fact that police limiting to the studies that came out in more recent years, last year, this year. What we find is a very different set of papers. So one, this bottom paper, at "Overlooked Perspective of Police Trust in the Public," thinking about how the community trusts. Again, something that we think is very salient to help policing... The police efficacy, but then also an outcome police can impact might depend on police actions. This other paper affects the body worn cameras. It looks at the effects of body-worn cameras on police actions and civilian complaints. So how do body-worn cameras? What are the different things that policing can do, and how does it affect whether or not civilians trust police, whether that they have grievances or concerns about police? Again, very different from thinking about crime. This paper affects the police on mental health of black families. Again, something very different than thinking about police, the effects of police on crime. Again, here thinking about effects body-worn cameras on civilian complaints. And then this first other field by this guy, that sounds like a really interesting guy, "The Effects of Police Violence on Inner City Students," really thinking again about this very holistic idea of, "What do police do? What are the different margins they can impact in the community in terms of community wellbeing?" And so here, two things will emerge. Obviously one is that, we have this very diverse set of research questions. That's again, very different, a much broader take on the role of policing. But again, not something that's not informed or completely thought up in academia, but something that's very well rooted in sort of the historical narratives in sort of touring events as well, in terms of thinking about what do police do and actually taking that seriously and measuring those impacts on things other than the crime. Another thing that'll emerge, and you can just see this from... So the list of names in these articles, as well as me presenting this to you is that, not all of these authors are white and they're not all men, there's a much broader, diverse, author set here, asking a much broader set of research questions about the role of police. And so we do think that this is gonna be really important for when we actually have these conversations, exactly the conversations that we're having today about, "What's the optimal level of policing?" It's not just when we think about that, it's not just calibrating, you know, how police affects crime. It's thinking about the sum of how police affects communities across these different dimensions and weighing them against each other, but then come up with the optimal policing solution. And so this is very much where my work lies. Just thinking about whether the social impacts of police violence on different aspects of community wellbeing. Specifically in my work, I've looked at how police killings affect the emotional academic wellbeing of children, how civilians and citizens engaged with government, their trust in government, as well as how they engage with law enforcement, whether or not they're willing to report crimes, and their willingness to commit violence. And so to do this... And I'll go over this very shortly, and happy that I've decided to show you my results. We'll just be leveraging really, really detailed, granular data on where students were voters, where 911 calls live or emerged from, emerging that with very detailed written data on the time and location of where police killings occur. And so, again, getting to what we actually find. So in that first paper, we saw the effect of the police killing on police violence on inner city students. What I do there is I'm gonna combine very detailed data for 15 years from Los Angeles, where there were over 600 police killings during this 15 year period. Again, we have the exact time and location of these events, which then I can merge and map onto where all the public high school students in Los Angeles lived as well, as each student's various academic outcomes. And so just to show you what we're actually doing here. Here, what we're showing you is just absenteeism rates for students based on how far they live from where a police killing was going to occur in the week before the police killing occur. And so you can see police killings occur in neighborhoods where absenteeism rates are relatively similar to neighborhoods that are further away from where the police killings occur. But now we can look at these same trends in terms of absenteeism rates, in the week after a police killing occurred, which is this line. And what you see here is that for students that lived very close to where the police killing occurred, there's this large increase in the likelihood of missing school the following week, okay? So we can do the same thing. We can look at students who live close to where police killing occurred, before and after the police killing occurred, compare their outcomes as soon as we go slightly further away, or in all of the dimensions very similar, and see how that affects their outcomes over a longer period of time. So here we can look at their effects on GPA. How the police killing... How does exposure to a single police killing effect student's GPA? And here we're defining, again, exposure as living close within half a mile of the police killing. You'll see that students who are exposed to this police killing experience significant decreases in their GPA, lasting several semesters. And these aren't small effects. So these GPA effects... If you benchmark them, they're as negative as some very high intensity tutoring programs are positive. But on average, each police killing in Los Angeles where things are relatively dense is gonna impact 300 students. So these effects you're seeing here is the average effect across 300 students who happened to live close to each police killing, and again, there's over 600 police killings in Los Angeles during this time period. These students also experienced large decreases in their mental wellbeing and psychological wellbeing. So here, what we're showing you is that, students who are exposed to police violence are significantly more likely to be diagnosed as emotionally disturbed, which is a special education designation. That's highly correlated with PTSD. And these effects are gonna serve as exacerbate persistent racial disparities in education. So it's not only that black and Hispanic students are more likely to live in neighborhoods where police killings tend to occur, but conditional living close to a police killing, black and Hispanic students are much more negatively impact than white or Asian students. That's what we're seeing on the left graph here. White and Asian students actually are not affected average by police killings, black and Hispanic students experience significant negative effects on their GPA. And this is coming, as you can see in the right panel, predominantly from police killings of other minority individuals. So police killings of black and Hispanic individuals lead to these specifically dire effects for the educational wellbeing of black and Hispanic students nearby. And the long run costs of this are enormous. So here, what we're looking at is exposure, the police killings, what's the average effect for exposure to the police killing during high school on students and GPA. When they left high school, their cumulative GPA when they left high school. Their likely hood of graduating from high school and their likelihood that actually enrolled in college. And so black dots correspondent effects for black and Hispanic students, white triangles correspond to effects on white or Asian students. You'll see, again, these large impacts on students academic achievement, that's gonna cause them to be much less likely to graduate high school and ultimately, less likely to actually enroll in college. And so if we were to back this out, what does this mean? Over this 15 year time period, I'm looking in Los Angeles, police killing has caused 2000 students of color to dropout of high school. And this is again, just in L.A. And so we can clearly see there's this large educational, and economic cost of police violence. And we definitely wanna be thinking about this. If we think about what's optimal law enforcement strategy, we wanna be benchmarking this to whatever gains we get in crime reduction, right? And so these are margins on which it's really important to measure these things so that we can then come up with the optimal law enforcement strategy, as opposed to just narrowly focusing on one specific outcome like crime. We can also look at other outcomes. So here we're looking at how police killings, how living in the same block as a police killing effects civic engagement, in terms of how likely someone is to register the vote, how likely someone is to actually vote. What we find is that police killings actually increase voter registration and turnout among citizens nearby. And this is driven entirely by black and Hispanic citizens. And it's not so much that these individuals are just responding to violence. What we find is that the largest effects, the largest genes and turnouts and registrations are coming from police killings in which the person who they killed is unarmed, versus the person who they killed was armed. That's what we see on the left is unarmed killing it's... On the right on the is unarmed fellows, on the left is armed killings. And we further see if you look at referendum voting that individuals who are exposed to police dealings are then much more likely to vote for criminal justice reforms that seek to soften and loosen the criminal justice system and penalties. And so all of this, again, points to this idea that these police killings have some impact on how individuals feel about the criminal justice system, whether they think that it's just, whether they think that there's discrimination, et cetera, we can see this boring out in actual voting active. Even from a criminological perspective, there's evidence that police violence might have the perverse and the adverse effect that we want it to, as opposed to reducing crime it could actually increase crime. So here we're looking at just raw data from a number of different cities, how does gun violence and 911 calls change in the wake of George Floyd's murder. That's what this left. That's what this red vertical line on these graphs indicate. In the left graph, what you can see is that, right after George Floyd's murdered there's a huge spike in actual gun violence, casualties, and injuries in this country. There's actually a large decrease in 911 calls, right? And so what you're finding is that George Floyd's murder by police, that's actually gonna increase societal violence, and it's gonna decrease the likelihood of civilians actually calling the police, reporting crimes to police, et cetera, which we might think is a really important input into how effective policing actually is, and ultimately how safe cities are. So all of these things that we're looking at here are things that were outside the purview, outside the scope of the questions that people had asked historically about what's the role of police, how the police affect communities. But I would argue, these are things that are really fundamental to know, if we're gonna act actually sort of, do some cost benefit analysis, to think about what's the optimal amount of policing we should have. And so again, there's significant research and this... You know, if you look at this research that exists, it's very well done, it's very valid, that more police does equals less crime. But then the question becomes, you know, "At what, and who's cost?" Are these crime benefits that we get from having more police, more intensive policing, are those outweighed or not by other consequences of having more police in different forms of policing? And there's increasing evidence that police violence can cause significant social harm across a number of different dimensions, especially for racial minorities. So I think there's two sort of, main takeaways from this talk, that I think are important to emphasize, from a policy perspective. Obviously, I think this highlights potential role of alternative methods of fostering public safety. This could mean a number of different things. Obviously, it could just mean more police officers doing different things. It could mean less police officers, it could mean... You know, again, different ways of approaching policing. It can also mean different stakeholders getting involved in communities, thinking about community investment, in terms of early education, in terms of social safety net benefits, things that we know in the long run have some mitigating effect on crime. But I think there's also a really important takeaway from a research perspective, from sort of like a meta, thinking about academia perspective and the research institution is, just emphasizing the points of diversity in academia. So I think it's often easy to think of diversity as this zero sum game, right? So if we're thinking of affirmative action in schooling, affirmative action in hiring, we're thinking these very narrow terms of who is getting this job and who is... You know, whether or not this person is better than the other person at completing a given task that's been assigned to them. Well, especially in something like academia, especially something like research, there's a lot of portfolio value potentially, of having diversity. And making sure that we're asking all the questions that we need to, to get to the right policy decisions. And I think this is obviously just one case in which you can clearly see there's this large difference in terms of the questions that have been asked historically and presently as well as... That's reflected into the diversity of authors asking questions. So that's it for my part slides, and obviously happy to answer questions or can hear your comments as well. Thank you so much.

- Great. Thanks very much. So we're now gonna open the session for your questions. To ask a question, please use the virtual hand-raising feature on Zoom. And please, in true Kennedy School fashion, keep your question brief and end it with a question mark. You'll be notified via Zoom's chat feature when it's your turn to speak, you may experience a short lag time. So make sure to unmute yourself from here from the staff. If you could also just let us know your Kennedy School affiliation, that would be great. So I'm gonna start things out with a first question that was submitted earlier by Evine Malloy, MPP 1996. And that question is, "What is the role of instant access to social media and videos of these acts of violence? Have they helped or hindered the process of justice?"

- Yeah, that's a great question, Yvonne, and it's a complicated one. In my own work, I find that... And especially if you think about these effects of police violence and them being very sort of geographically concentrated, a lot of these effects come from knowing that these events occur, right? And so there's a little bit of this idea of like, ignorance is bliss. Like, if we didn't publicize these events, a lot of people wouldn't have been exposed to this very traumatic video of George Floyd being killed. Maybe that would have actually be embedded in some sense, in terms of your short run, mental health, emotional health type standpoint. On the other hand, I think that, especially in this country, there's so much segregation in terms of where people live across racial groups, as well as very large differences across those neighborhoods in terms of how policing is done, that these videos really provide a window, especially for individuals who don't live in the types of communities where police violence is common, to understand and empathize with what police violence and policing looks like in different neighborhoods. And that's a really hard concept to sort of like drafts, just looking at numbers. I think looking at some of the social media videos, give this other sort of more extreme take on, sort of really getting a sense of what these things look like. And if we look at survey data in terms of how people trust police before and after Georgia Floyd, what you find is that across all different communities and across all different cities in this country, even white individuals, Asian individuals, those communities who weren't necessarily have direct exposure to the police violence and direct exposure to the police brutality, or intensive policing, there is a large decrease in their likelihood of reporting that they trust police and that they think police are doing a good job, et cetera. In a way that I think, again, highlights the potential benefits of social media of just sort of like diffusing this knowledge around these large disparities in policing across this country. Specifically for communities that don't necessarily have direct exposure to those things.

- [Mari] Great. Thanks very much. Dwight, you are up there.

- [Dwight] Well, I guess we... Can these school grads off and go white? Cause that was gonna to be my question as well on video, and its effect on what happens next. One question you answered a bit earlier, but I'd like to hear more thoughts on, and that is, again, the racial makeup of the police force vis-a-vis, the communities that are being policed. And again, in particularly around the incidents of use of force.

- Yeah. That's a great questions, Dwight. And it's something that's very much at the frontier of the literature is, people asking, you know, "If we have more representative police departments, does that lead to changes and help we see it's done, at the various stream margin, does that lead to reductions in police violence, et cetera? And there's research that it really does. And so in Chicago, there's work showing at a very detailed level that black and white police officers police very differently. They police very differently specifically with regards to minority communities. And that black officers are less likely to make an arrest, to resolve a dispute. They're much less likely to use force to resolve a dispute. And they're much less likely to generate a civilian complaint at the end of the day. There's other work looking in the Southwest U.S, where there are circumstances in which, is essentially random whether or not, a black or white or Hispanic officer response to a certain incident. And they find that in incidents in which the officer responding to an incident, again, ran the races, it's actually randomized. That having a same race officer respond to an incident relative to a different race officer is gonna lead to a lower likelihood that, that forces use the lethal force, or it's gonna be used, the officer's gonna be firing their guns. So I think that there's increasingly evidence that having a more representative. Police force could be really valuable for addressing some of these concerns. And again, I think it gets to this idea of thinking about diversity at like a pretty broad level, right? So oftentimes when we think about diversity, we're thinking about diversity and hiring, and it's more about who's getting this one spot, this talented, you know, job, and it becomes very much this sort of zero sum game. But if you think of diversity and impacts of having representation in different professions, having these spillover effects onto actually, how those professions are done, what research is conducted, who's stopped, who's not stopped, how much police use of force has done. Then it becomes much easier to get a sense of like, way to aggregate effects of diversifying across these different domains.

- [Mari] Great. Thank you very much. If you would like to ask a question, please raise your hand using the Zoom feature. So we had a question that came in from Peter Holmes, MPA 1990. "If you could please talk a little bit about police governance, say by, independent civilian review boards and that how that might affect the social impact of policing."

- Yeah, I think right now there's not a ton of research on that, that I'm aware of. I think you can imagine though, that these things might matter in the sense that, a lot of concerns around minority communities with policing and the extent to which communities are willing to call the police, they are willing to comply when police officers ask them to do certain things, et cetera. All of this, a lot of this boils down to trust, right? And that trust is gonna be built on a number of different things. Some of which is politics, some of which is the extent to which you feel like you have a say in your local government in how policing is done, et cetera. And as you can imagine, the role that, you know, the very fact that there is a civilian review board that we know that this thing was implemented as a way of fostering trust, as a way of fostering oversight, that would have a pretty big impact, even sort of like psychologically or sociologically on how people view police in the way that potentially could foster, improve policing as well as improve just the policing. But I think the actual research on that, I think is a little... We're waiting for that, I think a little more.

- [Mari] Got it. Thanks very much. So Marc Pomper, you are up, please ask your question.

- [Marc] And thanks so much for the call. Just wanted to check and see you had a slide talking about the incarceration rates. You know, in the U.S relative to other countries. That sort of seems to overlap with kind of, the lack of beds for a seriously mental ill, mysterious mental. Can you give me a sense of what work is being done or you may have done, kind of looking at the burden that the seriously mental ill and the lack of hospital beds has posed on police forces around the country.

- Yeah, that's a great question, Marc. And so over the longer time period there has been, sort of the deinstitutionalization of the mentally, I think that's exactly what you're getting to. I think that there's been work done by Steve Raphael, who's a economist at Berkeley, to sort of untangle to what extent are changes in carceration rates, due to sort of the institutionalizing mental health, such as these individuals then just being put into to carceral systems. And they find that it managed, but that is far from sort of the leading explanation or the main reason why there have been increases in carceration rates. And so, you know, maybe this accounts for like 5% of the growth in the carceration rates, but far from that huge growth of 500% increase in carceration rates that we saw. Much more of that is around some of the punitiveness of our criminal justice system, around policies we have in terms of, who should be incarcerated and for how long, et cetera, and what constitutes different.

- [Mari] Great. Thank you. Our next questioner is Lena Benson. If you could please let us know your affiliation and ask your question.

- [Lena] Thank you. Yes, I'm Lena Benson. I'm a graduate from the MPB program class of 2010 and I'm also a 11 year resident of Los Angeles. So thank you very much for your presentation. It really hit close to home. I'm also an African-American woman. My question is, what nations do you recommend that we look to that seem to be doing this well. I mean, in America, we look to, you know, what's going on in other states, but in your research, have you found successful models in other countries that maybe we should be paying attention to?

- That's a great question. You know, I think... So, I don't have, you know, empirical research on this. I think just sort of my sense of my understand that it's sort of like broader informal sense on this, is that I think looking at other countries can be informative and it can certainly be informative in terms of thinking about, you know, whether or not we're an outlier, how our policies will respond to other countries policies, but the U.S is very unique, in sort of the evolution of our law enforcement system, and we're also just at a very different place with regards to race relations, with regards to how policing is done, that it becomes a little less obvious, like, how to take something, for example, that's working in Australia and import it over to how it works in the U.S, right? And you can imagine in a situation like the United States where there's a lot of gun carrying. There's a lot, sort of like illegal gun carrying as well. And there's a lot of distrust between police and communities that, for example, like disarming police, in a way that has been done in Australia, could have very different effects here than it could in those other countries. And so I think, a lot of where we're at, is somewhat path dependent and specific to the United States. And I think that makes it a little more complicated, thinking about how we need to be addressing these issues. And again, that's not to say that those policies, like at the end of the day, I think a lot of people argue that, you know, "We just don't want police to have guns period," but I think that the reality of, given how much discrimination that has been historically, given how much that has led to distrust, et cetera, it gets a little harder to think about, how we would actually get there in the longer run.

- Thanks for that question and answer. Our next person up is Cynthia Hodge, go ahead and ask your question.

- [Cynthia] Hi, I'm Cynthia Hodge, a graduate of the MPA program in 2002, and very much interested in the topic of the impact of violence on communities. So I first wanna just thank you very much for the work you're doing. You brought forth a lot of information that I was not aware of, that I had not seen before. Can you share with us what motivated you to look at this topic, to the extent that you're looking at it?

- Yeah, that's a great question, Cynthia. You know, this comes from somewhat of a personal place, not in terms of like what the findings would actually show, but just in terms of my interest in the topics. And so I grew up in a part of Virginia, in this country where, you know, there just weren't very many people that looked like me, for whatever reason my parents decided to settle there. And so I think a lot about this idea that race is going to be a lens through which people view events, to which they interpret events and that our interactions with different institutions are really gonna color how we feel about our own aspirations, how we feel that we can or can't engage, with these same institutions and other institutions as well. And so in that lens, thinking about the role of race, the role of like perceptions of discrimination, perceptions of injustice on how we actually conduct our lives in some sense, and police killings are obviously a very important topic in their own, but they're also sort of a very sort of salient event in which there's a lot of evidence to suggest that these feelings are very much heightened in the wake of police killings. And so this is, in a very narrow sense, this is just a way for us to quantify, you know, what's the impact the police killings and how should we be weighing that against the time benefits of more aggressive policing or more police in general. Within a broader sense, I think it gives us a window into thinking about what is it like to be sort of, a racial minority growing up in the area where you might have concerns about justice and discrimination and how does it affect your long run outcomes in terms of whether you're going to school, how you're engaging with the government, how you're engaging law enforcement, et cetera. And so that ladder bent is very much how I began approaching this topic. But I think again, you know, I think both of these aspects are very much central to what we actually find.

- [Cynthia] Thank you.

- [Mari] Yes. Thank you very much for that question and answer. So our next question came in from the chat from Jeff Ashford, who is MPP 1982. And he was wondering, "Given recent efforts at policing reform legislation, such as the George Floyd Act, which was passed by the house but was languishing in the Senate, and resistance to higher standards for accountability and accreditation by police associations and unions, is research looking at the cultural barriers to improving police community relations and the use of force? What is the role of lobbying and public relations efforts by police unions and other organizations?

- Well, that's a great question. It's a lot to unpack. You know, I think the history of policing, the current moment that we're in, there's a lot of politics involved and there's a lot of very different and very entrenched stakeholders. All of which I think tie in with what you're asking. I think some of this is very hard to tease apart from the type of research that I do and other economists do, which is a bit more narrow in terms of being able to isolate a specific impact of a specific thing. And in culture is something where there's not so specific shocks to filter that allow us to sort of like, isolate how do these changes in culture impact other outcomes. And so, unfortunately I don't have a ton to say on that, but I do think in some broader sense, these things are vital to how we think about policing, how we think about policing reform and the sort of blow back that we have seen in terms of, you know, calls for reform in the recent past.

- [Mari] Great. As a reminder, if you'd like to ask a question live, please use the Zoom hand-raising feature. We have another question that came in from Gary Grim, and that is, "The selection of officers and their training depends on what we define as their role, have you or anyone else looked at the impact of the selection and training of officers on the social impact of policing?"

- That is an excellent question Gary. And I think you're exactly right. Like, there's a huge sort of a set of articles in the "New York Times" recently talking about police violence and how a lot of police killings. And I saw this in my own research emerged from, stops of motor vehicles who like, had no weapon, but, you know, police are sort of like trained to approach that situation as if it was very fraught, leading to decisions in which the driver of the vehicle was killed or the passenger still, et cetera. And so I do think training and selection has a huge role to play in all of this. There is some work coming out from UC Irvine by Emily Owens, who's a very prominent economist who studies crime showing that having sort of training of police officers so that they sort of like take a moment to think about situations as opposed to so immediately escalating them can, so the reduced use of force. I think this is happening... You know, I don't think we should expect that to sort of solve what we find, but it could reduce what we see in terms of use of force rates and certainly, sort of like unwarranted use of force at the margins. And so there's certainly more work thinking about doing that as well. There's other work in other contexts thinking about, for example, racial discrimination among bail judges and other types of actors within the criminal justice system. And I think there's some work that actually shows that more experienced individuals are less likely to rely on racial discrimination and stereotypes to make decisions in such a way that perhaps, you know, more training or selection of more experienced individuals might help to mitigate some of these biases as well. But again, I think these things are happening at a relatively small margin, and we shouldn't expect that. You know, these things are gonna drive the number of police dealings in this country down to zero. They'll just take it down from 1,000 into something a little bit lower, I think.

- [Mari] Great. Thank you very much. We have a question that came in from Wendy B, who is MPP class of 2012, it's a two-part question. "So are there efforts underway to do similar research in other cities?" And secondly, "What other outcome areas are you exploring under the umbrella of social impacts apart from student outcomes and voter turnout?"

- Yeah, that's excellent question. So, I mean, I would love to do similar research in other cities. The main constraint is really getting detailed enough data to be able to identify the effects. So what we saw in L.A setting was that, these events really only impact students that live relatively close to them, right? And so that requires having data at a very dis-aggregated level where you know where the students live is actually where police killings occur. So you can sort of map these things out, for the sort of zoom out and say, what's the average effect on graduation rates across all of L.A, it's somewhat doubtful that we'd be able to find, or even identify some of these effects. And so I think the constraint there, is really just data constraints. And so if you have any leads on that, obviously more than open to hearing them. In terms of thinking about other outcomes, you know, I think all of the work that I do, touches on this idea, that's a little bit more core, right? Like we're looking at a proxies of engagement with schools, engagement with electoral systems, engagement with criminal justice systems. All of that you can imagine has some motivating underlying driver, which is something to do with trust and institutions thinking about like, "How do I resolve the students? Who do I go to as a result of these things? Like, ultimately what are my feelings around law enforcement system, how does that affect who I trust within my community? Whether or not I take part in community organizations, gangs, et cetera," in a way that is really poor, if you think about, so the development of different communities in the United States, there's something really sort of foundational there, that's hard to capture in some of these outcomes, even though they're economically important. And so I think one thing that I'd love to be able to do, is to really hone in on some of these, sort of psychological sociological mechanisms that are happening, like ask people. Like, what's actually happening in the wake of these places. "How does it affect you?" Who you trust, who you care about, you know, how you feel about people of other races, how you feel about people in the same race, how you feel about different institutions, et cetera, in a way that really gets to sort of like, like the belief motivations behind these things. And so I think that's obviously a difficult thing to capture, but I'm hoping that there's some, you know, using some type of survey evidence to get to it.

- [Mari] Great. Thank you very much. Our next questioner is Pala McKeany Rainey. Go ahead and state your affiliation and ask your question. Okay. I his hand raise was unintentional. So we have another question that came in from Patrick Casitas, MPP 2004. And he is also a independent police investigator for Bart. "So you could define the term police killings?"

- Yeah. So police killings are incidents in which police killed somebody. And so that's... Not to say anything about sort of like, legal adjustability and in those cases, almost all of these circumstances, they are illegally justified. So what we're looking at here is, where somebody died at the hands of police. And in most of these circumstances where we're looking at are someone that was killed because the police shot them. But there's other ways in which police can kill individuals. And again, that's very different from legal definitions of whether or not they were, you know, criminally at fault.

- [Mari] Understood, thank you for clarifying that. So our next question is from somebody with the initial A, you could let us know your name and your question. Hello, you are on. Okay, let's go to the next question. I'm not sure what's going on there. "I'm just wondering if you could talk about police service, whether it should be limited to combat criminal behavior specifically and should new services be adopted using people who've been educated and trained, for example, to handle such things as traffic violations?"

- This other question is around...

- [Mari] Yeah, the question is like, you know, the extent of using police just for criminal activity, as opposed to like traffic violations or more minor misdemeanors, is that a trend that you've researched at all or seen any impact of?

- Yeah. So I think there there's some research, I'm not super familiar with it. I think that there are... Sort of, there is certainly research around and thinking around the idea of like, do we need police officers making traffic stops, if it puts them and the driver into the harmful situations where we could be doing something similar with technology using, you know, a radar guns and automatic ticketing, et cetera. I think that's sort of the frontier of the research in terms of thinking about how that reduces use of force, how that, you know, perhaps improves trust in government, et cetera. I think to your point, you know, there's a lot of research suggesting that, not just police violence, but like very aggressive, proactive forms of policing, like stopping friction, et cetera, have just like detrimental on net impacts in communities. And so, I think you're exactly right in thinking like, "Well, what are the ways that we can be achieving these same goals without necessarily involving these face-to-face police context?"

- [Mari] Great. Thank you very much for answering that question. Just see if there's anything else in the chat. So just wondering if you could talk a little bit about kind of the... Again, not only the training of police officers, but the kind of characteristics of police officers who are, you know, maybe getting there... And this question came in earlier, getting their facts from maybe questionable media sources, and basically, you know, just the whole division between, you know, fake news and all of that. Like, what is there... Are there any efforts being made to educate police officers?

- Yeah. I mean, that's a good question. And it's something that, you know, everyone is susceptible to it, right? Like, there's a lot of evidence, including some of my research that media can have these very important role to play in terms of how people view people of other races, actually, how their behavior towards people other races, like racial violence, racial prejudice, et cetera. And, you know, and that this is something that we're all susceptible to, whether or not we're sort of selecting into like listening to various stream radio or watching various streaming TV, but like, but there are these subliminal effects too, of just like watching things that perpetuate certain stereotypes and media, et cetera. And so, you know, I don't... You know, obviously police are important margin in this drive, but, you know, I think we're all sort of like vulnerable to these impacts. And I think there is sort of more work thinking about, "Okay, when does media matter? When does it not matter?" Like, what are ways that we should be thinking about regulating these things, but I'm not super aware of something like this, specifically in the context of law enforcement.

- [Mari] Got it. Thanks. We have time for one more question. We're gonna go to Bob Ferri, go ahead.

- [Bob] Hi. Thanks professor. This is really valuable, I appreciate it. You know, you make the point that more cops equal less crime, but at what and whose cost? Is there any research out there that would project maybe, do more guns, mean more crime? If we had actual gun control... You know, European style, Japanese style, gun control. Is there any data out there or research out there?

- Yeah, let's see. There is research. I think the research is somewhat mixed in terms of, I think what they look at are sort of ventral policies in different states. And, you know, when you put in these policies, does that affect crime? The research has actually surprisingly mixed in that regard. I think that is a little bit of a different question than what you're getting at, which is like, is there some state of the world in which like, there's just no more guns, right? Or there's a lot more guns. Like what does that do? You know, you can imagine that we are in this equilibrium now where there's already a lot of the guns. And so there's a lot already a lot of stuff that's circulating and we don't already have a good sense of how many guns are even circulating. There's not very good that on that. And so it becomes a little harder to think about, you know, this state of the world in which we're having like these drastic changes in gun policy, if that makes sense, to project what would happen there.

- [Bob] Sure.

- [Mari] Great. Thank you very much. And thanks to everyone who called it to listen to this last winter conference call of the calendar year. And a special thank you to Desmond Ang for providing his expertise this morning to our Kennedy School, alumni and friends. Have a great holiday season and see you in the new year.

- Thank you so much everyone.