{"id":219,"date":"2019-09-30T13:48:00","date_gmt":"2019-09-30T13:48:00","guid":{"rendered":"http:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/?post_type=podcast&p=219"},"modified":"2020-11-16T19:45:57","modified_gmt":"2020-11-16T19:45:57","slug":"episode-13-suspect-citizens-with-frank-baumgartner","status":"publish","type":"podcast","link":"https:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/podcast\/episode-13-suspect-citizens-with-frank-baumgartner\/","title":{"rendered":"Episode 12: Suspect Citizens (with Frank Baumgartner)"},"content":{"rendered":"

Brooke, Christine, and EJ talk with Frank Baumgartner about his new book (co-authored with Derek Epp and Kelsey Shoub) Suspect Citizens: What 20 Million Traffic Stops Tell Us About Policing and Race<\/em>.<\/p>\n","protected":false},"excerpt":{"rendered":"Brooke, Christine, and EJ talk with Frank Baumgartner about his new book (co-authored with Derek Epp and Kelsey Shoub) Suspect Citizens: What 20 Million Traffic Stops Tell Us About Policing and Race.","protected":false},"author":13,"featured_media":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"_acf_changed":false,"_genesis_hide_title":false,"_genesis_hide_breadcrumbs":false,"_genesis_hide_singular_image":false,"_genesis_hide_footer_widgets":false,"_genesis_custom_body_class":"","_genesis_custom_post_class":"","_genesis_layout":"","episode_type":"audio","audio_file":"http:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/wp-content\/uploads\/sites\/20\/2019\/07\/The-Policy-Agenda-Episode-13.mp3","podmotor_file_id":"","podmotor_episode_id":"","cover_image":"","cover_image_id":"","duration":"","filesize":"46.6M","filesize_raw":"48862784","date_recorded":"","explicit":"","block":"","itunes_episode_number":"","itunes_title":"","itunes_season_number":"","itunes_episode_type":""},"tags":[48,72,73,9,71,74],"categories":[],"series":[2],"class_list":{"0":"post-219","1":"podcast","2":"type-podcast","3":"status-publish","5":"tag-agenda","6":"tag-police","7":"tag-policing","8":"tag-policy","9":"tag-race","10":"tag-traffic-stops","11":"series-the-policy-agenda","12":"entry"},"acf":{"related_episodes":"","hosts":[{"ID":613,"post_author":"38","post_date":"2020-07-01 18:02:41","post_date_gmt":"2020-07-01 18:02:41","post_content":"\n

Fagan is an Assistant Professor of Political Science at the University of Illinois at Chicago.<\/p>\n\n\n\n

He studies political parties, Congress, think tanks, agenda setting, lobbying, and economic policy. In particular, he is interested in how legislatures process information and solve problems, and how party elites influence policy. He is also a graduate research fellow and former project manager with the Policy Agendas Project<\/a>. He produces and co-hosts the Policy Agendas Podcast<\/a>.<\/p>\n\n\n\n

Before graduate school, Fagan worked for five years in Washington, DC. From 2011-2014, he was Deputy Communications Director for Global Financial Integrity<\/a> (GFI), a research and advocacy organization based in Washington, DC. GFI works to study and curtail illicit financial flows from developing countries. <\/p>\n","post_title":"E. J. Fagan","post_excerpt":"","post_status":"publish","comment_status":"closed","ping_status":"closed","post_password":"","post_name":"e-j-fagan","to_ping":"","pinged":"","post_modified":"2020-10-23 16:53:07","post_modified_gmt":"2020-10-23 16:53:07","post_content_filtered":"","post_parent":0,"guid":"http:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/?post_type=speaker&p=613","menu_order":0,"post_type":"speaker","post_mime_type":"","comment_count":"0","filter":"raw"},{"ID":638,"post_author":"38","post_date":"2020-10-27 15:48:58","post_date_gmt":"2020-10-27 15:48:58","post_content":"\n

Ph.D. candidate and teaching assistant in the department of government at the University of Texas at Austin who specializes in research on local and urban policy, agendas, and Latinx politics.<\/p>\n","post_title":"Brooke Shannon","post_excerpt":"","post_status":"publish","comment_status":"closed","ping_status":"closed","post_password":"","post_name":"brooke-shannon","to_ping":"","pinged":"","post_modified":"2020-10-27 15:48:59","post_modified_gmt":"2020-10-27 15:48:59","post_content_filtered":"","post_parent":0,"guid":"http:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/?post_type=speaker&p=638","menu_order":0,"post_type":"speaker","post_mime_type":"","comment_count":"0","filter":"raw"},{"ID":634,"post_author":"38","post_date":"2020-10-27 15:44:56","post_date_gmt":"2020-10-27 15:44:56","post_content":"\n

Christine\u2019s research lies at the intersection of law and politics. Her dissertation work examines the policy agenda of the Solicitor General and the separation of powers dynamics at play in Supreme Court litigation. She specializes in judicial politics, constitutional law, and how law interacts with public policy. She plans to complete her PhD in Political Science in May of 2021 at the University of Texas at Austin. Christine holds a juris doctorate and a license to practice law in her home state, Oklahoma.<\/p>\n","post_title":"Christine Bird","post_excerpt":"","post_status":"publish","comment_status":"closed","ping_status":"closed","post_password":"","post_name":"christine-bird","to_ping":"","pinged":"","post_modified":"2020-10-27 15:44:56","post_modified_gmt":"2020-10-27 15:44:56","post_content_filtered":"","post_parent":0,"guid":"http:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/?post_type=speaker&p=634","menu_order":0,"post_type":"speaker","post_mime_type":"","comment_count":"0","filter":"raw"}],"guests":[{"ID":604,"post_author":"38","post_date":"2020-07-01 17:32:41","post_date_gmt":"2020-07-01 17:32:41","post_content":"\n

Frank R. Baumgartner, of UNC-Chapel Hill, has written extensively on the topics of agenda-setting, public policy, lobbying, and framing in both US and comparative perspectives. In recent years, he has focused on statistical studies of criminal justice issues, including the death penalty, racial dispartiies in traffic stop outcomes, and other issues, as well. His most recent book is Suspect Citizens<\/em> (Cambridge, 2018), focusing on racial differences in the outcomes of routine traffic stops. In 2019 he was recognized with the C. Herman Pritchett Best Book Award from the Law and Courts Section of the American Political Science Assocation (for Suspect Citizens)<\/em>, and with the Lijphart\/Przeworski\/Verba Dataset Award from the APSA Section on Comparative Politics (for the Comparative Agendas Project).<\/p>\n\n\n\n

In addition to normal teaching and research responsibilities, for 2019-20 he serves as Director of Placement for graduating PhD students in Political Science<\/a>, and Faculty Co-Chair of the UNC-Chapel Hill Campus Safety Commission<\/a>, among other service appointments.<\/p>\n","post_title":"Frank Baumgartner","post_excerpt":"","post_status":"publish","comment_status":"closed","ping_status":"closed","post_password":"","post_name":"frank-baumgartner","to_ping":"","pinged":"","post_modified":"2020-10-29 17:35:24","post_modified_gmt":"2020-10-29 17:35:24","post_content_filtered":"","post_parent":0,"guid":"http:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/?post_type=speaker&p=604","menu_order":0,"post_type":"speaker","post_mime_type":"","comment_count":"0","filter":"raw"}],"transcript":"

Hello. Welcome to episode twelve of the Policy Agendas podcast presented by the Policy Agendas Project. I<\/p>\n

am E.J. Fagan. Today, I’m joined by Brooke Shannon. Hello. And Christine Bird,<\/p>\n

how are you doing? And we are here to talk to Frank Baumgartner. Frank is<\/p>\n

the Richard J. Richardson, distinguished professor of political science at the University of Carolina,<\/p>\n

Chapel Hill. He was the one of the co-founders of the Policy Agendas Project. And he’s the author<\/p>\n

of the book we’re going to discuss today. Suspects, citizens, what? Twenty million traffic stops tell us<\/p>\n

about policing and race, along with his coauthors, Derek App, who has been on this podcast before, and Karslake,<\/p>\n

Kelsey Shupe. Frank, welcome to the podcast. Thanks for having me. So nice to be here. Yeah. So<\/p>\n

we have some questions about this book. This is this book came out in 2017, but part of an ongoing project.<\/p>\n

So I might have Brooke take that away. Sure. So, again, the book is called Suspect Citizens. And<\/p>\n

it’s a really fascinating, fascinating story about street level bureaucrat interaction<\/p>\n

with folks. And so we want to know what the book’s origin story was.<\/p>\n

The origin of this book is a little bit complicated. It goes back to a<\/p>\n

previous book I wrote about the death penalty. And that book came out<\/p>\n

in 2008. And then I moved to North Carolina to the RNC and 2009.<\/p>\n

And I immediately became acquainted with and got to be kind of integrated into a world<\/p>\n

of people who work on capital punishment from the defense side. And this is a<\/p>\n

group of people who are just really interesting, also very connected in with<\/p>\n

civil rights attorneys and people working on social justice and racial disparity issues<\/p>\n

throughout the criminal justice system. And a couple years after I moved, one<\/p>\n

of them asked me if I would volunteer, because I’ve found that they’ve<\/p>\n

discovered that I’m their favorite phrase statistical consultant. What I volunteer to serve<\/p>\n

on a task force to look into racial disparities in the North Carolina criminal justice system.<\/p>\n

And other people were looking at prison populations, drug arrests, school to prison pipeline.<\/p>\n

And they handed me a C.D. and said, would you look at these data? These are all the traffic stops in North<\/p>\n

Carolina. And so that was how it started. It was I think that might<\/p>\n

have been 2011. So seven years later, we were able to write a book about it.<\/p>\n

It took a long time to figure out how to work with this massive data set. And<\/p>\n

then also, I didn’t know anything about policing. And so I had to learn we all had to learn a lot.<\/p>\n

Yeah. So in the book, we learn a lot about these terms, I feel like some are very well acquainted<\/p>\n

with like driving while black. So I’m hoping you can explain<\/p>\n

a little bit of like defining these these terms, things like driving while black and maybe<\/p>\n

touch on something like explicit versus implicit bias, which comes through really well in this book.<\/p>\n

But some folks might not know. Well, the question of driving while black is something<\/p>\n

that white drivers might have concerns or questions about. But most black or Hispanic<\/p>\n

drivers don’t have any questions about whether such a thing. And that’s the idea that<\/p>\n

minority driver would be subject to increased police surveillance compared to a white driver.<\/p>\n

And that was been in the news a lot, starting, I think, in the nineteen<\/p>\n

nineties. And it really has surged to a lot of attention in the late 1990s.<\/p>\n

And then it kind of faded away after the 9\/11 attacks in 2000. One<\/p>\n

and we didn’t hear so much about driving while black anymore because now we<\/p>\n

had a whole different set of concerns about terrorism and policing<\/p>\n

changed a little bit. But in the meantime, North Carolina was the first state in the country<\/p>\n

to mandate collection of traffic stop data, all traffic stops. And that was an effort to either<\/p>\n

debunk the myth of so-called driving while black. Or perhaps<\/p>\n

validate the concerns of minorities, drivers and community leaders who said that there was<\/p>\n

such a thing. And there was an editorial in the<\/p>\n

Raleigh News and Observer when the bill was being debated in the General Assembly in 1999.<\/p>\n

That said, we support the bill and we believe that it will show there is no such thing.<\/p>\n

But if it does, we understand that our police leaders will take immediate action to address it.<\/p>\n

And so anyway, that that’s where that term comes from, I think. And our<\/p>\n

data make it very, very clear that there is indeed such a thing. We showed<\/p>\n

one part of our analysis as necessarily less it’s not<\/p>\n

on firm ground that is comparing who’s in the population to<\/p>\n

who gets pulled over. And we refer to this as the benchmark problem. We don’t really<\/p>\n

know who’s driving. We don’t know who’s speeding or violating the traffic<\/p>\n

laws. But we do know there are good data, you know, who lives in which communities.<\/p>\n

And the best we can estimate is that, roughly speaking, an African-American driver<\/p>\n

has about twice the odds of being pulled over compared to a white driver. Roughly speaking, and that’s a<\/p>\n

probably an underestimate of the true difference. But we know for sure<\/p>\n

that once you get pulled over, then we know exactly what happened to you. Did you get arrested? Did you<\/p>\n

get the car surge? Did you get a ticket? Did you get a warning? And there we can be very<\/p>\n

precise because we know both the numerator. That’s everybody who got pulled over and the denominator.<\/p>\n

What happened? What was the outcome of that traffic stop? So there were on very firm ground and there was show<\/p>\n

that controlling for other factors, the racial factor<\/p>\n

consists of about it, about doubles, the rate of search after controlling<\/p>\n

for why did you get pulled over and all the other factors.<\/p>\n

Yeah. So what you’ve just alluded to is the massive data contribution that<\/p>\n

this book brings to the political science canon.<\/p>\n

There’s 20 million observations here. Can you talk a little bit about the process<\/p>\n

that your team use to cut up this massive data set to do analysis<\/p>\n

with this work? And what did<\/p>\n

your process look like? Well, actually, the first part of the analysis was the most difficult<\/p>\n

that was getting our hands around the data set itself. It came to us from the state in a<\/p>\n

relational database that actually was broken into, I think five or six different<\/p>\n

databases. And so our first job was to use the u._n._c supercomputer cluster to<\/p>\n

merge all these different databases into one what we call a rectangular<\/p>\n

database that we could open on a computer and analyze. So, for example,<\/p>\n

there was a database for the stop and then there was a separate database that<\/p>\n

had information about searches and there was a third database that only had information about<\/p>\n

contraband. If there if contraband was found, then there was another database related to passengers<\/p>\n

and there was another database related to the demographics of the driver. And all those<\/p>\n

had common I.D. numbers so they could be merged together. But I could tell you it<\/p>\n

was a pain in the neck to try to work with the data base at the beginning to just get<\/p>\n

get it organized in a way that we could analyze it once we did. And I feel like this is<\/p>\n

an increasing problem in political science. We have these extremely interesting<\/p>\n

administrative datasets, but they’re really hard to work with. And so I’m working<\/p>\n

with another one right now, which is everybody in North Carolina who’s been arrested for anything ranging<\/p>\n

from taking a fish out of a river without a license or hunting<\/p>\n

and fishing violations to traffic violations all the way up to homicide.<\/p>\n

That’s everyone arrested for any offense at all. And it’s about 8 million records<\/p>\n

and it’s simultaneously a gold mine for research and also<\/p>\n

a nightmare because it’s a poorly organized database. And so we have to<\/p>\n

do a lot of anal just learning about how the database is organized<\/p>\n

and reorganize it. For example, it has multiple rows in the database for<\/p>\n

each arrest to the same person on a given day. It might be arrested for<\/p>\n

five different violations and each of those is a row in the database and<\/p>\n

then they might be indicted for a certain number of crimes on another date.<\/p>\n

And then maybe a year later those violations will be<\/p>\n

resolved. They’ll either plead guilty or they’ll be found not guilty before a judge or whatever the resolution<\/p>\n

is. But those really relate to three different units of analysis.<\/p>\n

And so there’s a lot of work and just understanding database management skills<\/p>\n

and also dealing with large amounts of data that require a lot of computer<\/p>\n

resources and then a lot of skills that we don’t teach and research<\/p>\n

methods, classes, because they’re the preliminary work you have to do to get the database in a form<\/p>\n

that you can analyze. So we had to do a lot of that with those traffic stops project. And<\/p>\n

I feel like learning those skills is just an important part of the job. Now, this important<\/p>\n

thing that I try to make sure my students learn. Even my undergraduates<\/p>\n

at some point. Do you have to stop relying on statistical methods? Raimy, are we. Are we throwing<\/p>\n

out a hypothesis and hypothesis test? We got regression analysis. Is that that time type of framework<\/p>\n

that so much a political science is in at 20 million traffic stops, that most of that<\/p>\n

becomes not useless but difficult to use? Right now it’s perfectly<\/p>\n

easy to use. And the really fun thing about it is everything is this is right. That’s the point. Every<\/p>\n

bump in the data creates a P value of zero def. Everything is significant. So<\/p>\n

then it gets you away from this discussion about is it significant too? Do I care about this?<\/p>\n

And that’s a much better conversation because we have coefficients in our model, which<\/p>\n

are, you know, like we we show that let’s say<\/p>\n

if you’re a black male driver, your odds of being searched after a traffic stop are,<\/p>\n

you know, one hundred and fifty six, you know, one point five, six times greater. Whatever<\/p>\n

number is a very large number. Very great disparity. And we can have a discussion about that disparity<\/p>\n

large enough to be concerned about. But we don’t have to worry about. Is it statistically<\/p>\n

significant? Cause everything is statistically significant. Yeah. So speaking<\/p>\n

of that, who gets pulled over? What are these sort of like compounding?<\/p>\n

Identities are characteristics that you see in the in these data.<\/p>\n

Well, some of the what I’m gonna say will strike certain listeners as<\/p>\n

obvious and other listeners might be surprised by them. But it kind of depends on the<\/p>\n

identity of the listener. So who gets pulled over as<\/p>\n

young black men and also people in certain communities rather than other areas or<\/p>\n

other communities? And it also depends on what time of day it is. And<\/p>\n

I’m also particularly interested in not just who gets pulled over, but once you get pulled over, you get searched.<\/p>\n

One of the bigger predictors is what is what time of day it is. If it’s 2:00 in the morning and the police<\/p>\n

pull you over, there’s a strong chance they’re gonna search your car. But if it’s 8 or 7:30<\/p>\n

a.m. and it’s the morning rush hour, and for whatever reason, you get pulled over. The police<\/p>\n

are concerned about keeping that traffic moving and so they’re unlikely to search the car.<\/p>\n

So some of the things are obvious<\/p>\n

when you think about it. Like, for example, if they pull you over for suspicion of drunk driving,<\/p>\n

they’re gonna search the car with a very high probability. On the other hand, there aren’t that many such traffic<\/p>\n

stops. There’s way more traffic stops for speeding and speeding. Traffic stop is very unlikely<\/p>\n

to lead to a search of the vehicle because most likely it was not a pretext.<\/p>\n

It was probably that you really were speeding and a radar gun said you were going 85 in a 60 zone.<\/p>\n

So you’re gonna get a ticket, but the officer doesn’t necessarily suspect you<\/p>\n

of criminal behavior. On the other hand, if you’re pulled over for expired<\/p>\n

registration tags now, officer might not really big that concerned about the expired tags.<\/p>\n

He might want to use that as a pretext to have a conversation because he’s sees you as<\/p>\n

for whatever reason, a suspect. So that was really<\/p>\n

what we tried to get to the bottom of this in this analysis. You seem to make<\/p>\n

a distinction where you can see the data, the difference between a fruitful and a fruitless<\/p>\n

search. Can you talk a little bit about what you see there? Well, we weren’t quite surprised in the<\/p>\n

in the analysis in general that the math is really bad for<\/p>\n

using traffic stops as a mechanism to find drug offenders.<\/p>\n

For example, we have about 20 million traffic stops in our database anyway. Of those, about<\/p>\n

vehicle searches. Of those, about 30 percent lead to the discovery of contraband.<\/p>\n

So that’s about hundred thousand of those. Only half lead<\/p>\n

to arrest because the so-called contraband was in such small quantities<\/p>\n

or wasn’t actually drugs. It was what might be called drug paraphernalia, or if it was<\/p>\n

drugs, it was, you know, a prescription pill which may or may not have<\/p>\n

been the person may or may not have had it with legal,<\/p>\n

might have been legal or not legal. So anyway, we just<\/p>\n

go through the numbers and show that using the vehicle code and the traffic,<\/p>\n

the vehicle called in the traffic code to search for drug offenders is really<\/p>\n

a needle in a haystack strategy that doesn’t really<\/p>\n

add up mathematically to an efficient policing strategy. And<\/p>\n

the other thing you mentioned about fruitless searches and there I would.<\/p>\n

The other thing we’re trying to do in this research program is to bring people’s attention to the fact<\/p>\n

that if the police do have the authority to pull you over or when you’re driving a car, because if you’re driving<\/p>\n

a car, most likely, almost certainly you’re breaking some law in some way, shape<\/p>\n

or form, even if it’s, you know, that either you’re speeding or if you’re not speeding,<\/p>\n

then the officer could say you’re impeding traffic by not going at the speed limit. So<\/p>\n

either way, you’re breaking the law. And there’s many other laws, such as having a cracked tail<\/p>\n

light, an expired tag, which are technically violations of the law, but which<\/p>\n

may not be very serious violations. But still, once you break any law,<\/p>\n

you open yourself up to a police investigation. Now, if<\/p>\n

that investigation reveals that you were involved in crime, then you simply got caught<\/p>\n

and you might be upset getting caught and having been caught. But<\/p>\n

you’re a criminal. If the investigation yields no evidence<\/p>\n

of criminal activity and the police officer then lets you go with a warning or<\/p>\n

no action at all. And there was no contraband found as a result of the search. Then you understand<\/p>\n

that you were fruitlessly searched and perhaps. She were humiliated and detained<\/p>\n

for potentially an hour. It could have been in a public<\/p>\n

place maybe where friends might have seen you being searched by the police.<\/p>\n

It could have been in front of your school. If you’re a young person, so it can have some really<\/p>\n

bad consequences. And from the legal perspective, the police have<\/p>\n

no responsibility to avoid such things. Of course, they wouldn’t want to engage in too many of them.<\/p>\n

But on the other hand, there’s no penalty. And so we wanted to bring attention to the<\/p>\n

idea that each such fruit fruitless search chips<\/p>\n

away a little bit at our fabric, at the democratic fabric, because it<\/p>\n

alienates a young person, typically a young man of color from agents<\/p>\n

of the state and from the government. So we know from other research that<\/p>\n

that I’ve not been involved in that involvement with the police and the courts and<\/p>\n

such things reduces the likelihood of voting. And it has all kinds of deleterious<\/p>\n

effects on people’s political participation. And so our research, I think,<\/p>\n

pushes in that same direction to demonstrate some of the costs. But the costs<\/p>\n

have been secret. I should point out to middle class Americans<\/p>\n

who aren’t concerned about it because it doesn’t happen to middle class Americans.<\/p>\n

But it happens routinely to individuals who fit certain profiles.<\/p>\n

I like what you say about it, detracting from the democratic fabric of our society. It<\/p>\n

affects all sorts of things that you point out in this book. It could affect your participation,<\/p>\n

your voting, your whether or not you would want to become a police officer, whether or not<\/p>\n

you’re going to engage with any other type of government bureaucrat, how you deal with the court system generally<\/p>\n

and not all interactions with the court are of a criminal nature. It can be whether<\/p>\n

or not you show up for jury duty. Things like that are all compounded by the work<\/p>\n

that is that is in this book. Do you.<\/p>\n

Is there any evidence that a fruitless search or a pattern for the searches<\/p>\n

could make someone more likely become a criminal? Well,<\/p>\n

I think one of the big in our book, we don’t have evidence about individuals<\/p>\n

repeatedly being pulled over because we don’t know the name or there’s no identification of the individual.<\/p>\n

So I can only really speculate. But one of the issues about police saying<\/p>\n

overpolicing certain neighborhoods is that if people come to be alienated from the police<\/p>\n

and then don’t trust the police, then we can see that they<\/p>\n

might be less likely to call the police when there’s a need for law and order.<\/p>\n

And then actually in our analysis, we were able to take advantage of a reform<\/p>\n

in one city in North Carolina where the police chief mandated certain changes<\/p>\n

to police policies to de-emphasize the kinds of traffic stops that we think<\/p>\n

of as more pretextual and to emphasize is really community safety by<\/p>\n

pulling over people for speeding, running through stop signs and things like that, and reducing<\/p>\n

the number of traffic stops for relatively minor infractions or equipment<\/p>\n

failures or regulatory problems. And in that community,<\/p>\n

which is Fayetteville, North Carolina, there was an increase in calls to 9-1-1.<\/p>\n

After these reforms were implemented. So I think that was a sign that there was<\/p>\n

increased trust in the police and understanding that the police were there for everybody and people who<\/p>\n

live in high crime areas need to trust the police more than anybody. And yet<\/p>\n

they don’t. On average, there’s likely to be more alienation from the police.<\/p>\n

Precisely. In those areas where you would hope from a crime fighting perspective that<\/p>\n

people would be most cooperative with the police. But that’s something that we haven’t seen in America for a long time.<\/p>\n

Something Derek told me that was an early criticism of this work is that people would say<\/p>\n

this isn’t about politics, this isn’t about political science. This is criminology. And I thought that was<\/p>\n

really interesting because it’s like I can’t think of a more fundamental question of politics than how much<\/p>\n

interference can the government have in somebody whose life is being searched. It’s a fundamental question of politics<\/p>\n

and of your interaction with the government and of the social contract in all of those things. How do you respond<\/p>\n

when these criticisms pop up? Well, I’ve written a book about the death, a couple of books about the<\/p>\n

death penalty as well. You know, there’s nothing more the government can do than, you know, take your life<\/p>\n

or plan to take your life in a judicial execution. And similarly,<\/p>\n

in traffic stops analysis, the behavior of the police towards its own towards citizens,<\/p>\n

I think is inherently a question of government and should inherently be<\/p>\n

a question of political science. Unfortunately, I think it’s true that<\/p>\n

within our discipline, there’s been a reluctance to understand the<\/p>\n

politics and the importance of understanding these judicial outcomes.<\/p>\n

I mean, we’ve had papers, deaths projected from major political science, reviewed journals<\/p>\n

for the absence of what they called political variables, such as partisanship of the governor<\/p>\n

or something of that nature. And as a political scientist, I’ve found<\/p>\n

that to be quite insulting and misguided. So was really frustrating.<\/p>\n

So with 20 million observations, there’s obviously quite a lot that<\/p>\n

you could do to dig deeper and go to different projects as well, and I know that this book is part<\/p>\n

of a pretty large research project. So we’re wondering about different<\/p>\n

articles that have come after the book. We know that you’ve recently written on characteristics of the officers themselves.<\/p>\n

Can you tell us a little bit more about the ancillary projects to this book? Yeah. After we<\/p>\n

got our hands around the North Carolina data, we realized that there were other states<\/p>\n

that had similar very large data sets in Illinois. I think there’s over 30 million traffic stops<\/p>\n

that have been collected in Illinois. And California has just come online with some new<\/p>\n

traffic stops data. So it’s really a growth industry, I think, for political science, some<\/p>\n

criminology and other scholars. And every time we look<\/p>\n

at a new database, it turns out there’s a couple new variables that aren’t available in some of the old ones,<\/p>\n

such as how old is the car? Does the car have out-of-state plates, et cetera.<\/p>\n

So we’ve been looking at various things. One major question we had was,<\/p>\n

would any of those additional control variables explain away the race effect? So we always<\/p>\n

look at that. And actually we did a paper where we calculated the simplest<\/p>\n

ratio of the black search rate divided by the white search rate with no statistical<\/p>\n

controls at all. And then we calculated the most sophisticated and<\/p>\n

complete logistic regression that we could possibly do, given the data that were collected in that<\/p>\n

state. And this analysis ended up covering, I think, Maryland, Illinois,<\/p>\n

North Carolina, several other states and had almost 60 million observations<\/p>\n

all together. And we found that the correlation between the simple ratio<\/p>\n

of the two search rates and the more sophisticated and statistically complete odds<\/p>\n

ratio between the four, the logistic regression for the race effect<\/p>\n

correlated at zero point nine seven. So what that meant<\/p>\n

was that if an analyst can calculate even the simple percent<\/p>\n

of black drivers in a percent of white drivers should get searched. That’s a robust indicator<\/p>\n

as it doesn’t get explained away by other factors such as time<\/p>\n

of day. How old is your car, your age?<\/p>\n

So any one thing was to check on the robustness of our findings, whether North Carolina<\/p>\n

was different than other states. We found that the findings are unfortunately<\/p>\n

depressingly common all across the United States. That was one.<\/p>\n

And then we’ve been in a number of other projects. One, as you mentioned, we have a database<\/p>\n

for one city that where we know the characteristics of the officer. And it<\/p>\n

turns out that just like the driver, race and gender matter a lot. The race<\/p>\n

and gender of the officer make a big difference. And white male officers have much higher search rates<\/p>\n

controlling for other factors. So that was a characteristic that we weren’t<\/p>\n

able to study in the broader North Carolina study. And we’ve looked at a number<\/p>\n

of other elements, such as minutes of pole reliance on fire and fines and fees.<\/p>\n

And it turns out that the the greater share of the municipal budget comes from fines and fees,<\/p>\n

the more likely that city’s police department is to have a more aggressive traffic stop<\/p>\n

program and also higher racial disparities in those traffic stops.<\/p>\n

As far as. Oh, sorry. As far as the different states go, have you noticed any<\/p>\n

diffusion of these policies or other states looking to North Carolina<\/p>\n

to see the success or maybe something that they don’t want to publicize?<\/p>\n

Well, one of the first things that happened after our first report, I think in 2012<\/p>\n

came out in North Carolina that showed these disparities was legislation was introduced to stop<\/p>\n

collecting the data. And so I think the data are there a challenge.<\/p>\n

And but that was not that was not successful. And North Carolina<\/p>\n

continues to collect the data. And I think other states have come online with similar data collection<\/p>\n

projects north of the state of California recently passed a very extensive legislation requiring<\/p>\n

it. And those data will be rolling out over the next several years.<\/p>\n

And so I think there is a movement in general towards open data and greater access<\/p>\n

to administrative computer, a computerized administrative records. And I think that<\/p>\n

for the next generation of political science and public policy scholars, it’s going to be<\/p>\n

a goldmine. So those do come with some limitations.<\/p>\n

There can be a lot of. Typographical errors can be.<\/p>\n

You really have to understand the nature of the administrative database you’re working with. But I do<\/p>\n

believe that they’re gonna be increasingly common. As<\/p>\n

you know, government agencies of all kinds maintain pretty sophisticated<\/p>\n

databases about their activities.<\/p>\n

Well, I’m sure you’ve also received criticism of like, well, isn’t this just North Carolina? Is isn’t North Carolina<\/p>\n

different than other states? But it seems to me that maybe North Carolina is a great test<\/p>\n

case to start this type of analysis, because it’s not too urban. It’s not too rural, it’s not too southern.<\/p>\n

It’s not too eastern. It’s not you know, it’s kind of a pretty<\/p>\n

good place to start in terms of what we look for in terms of that will be the Goldilocks theory of<\/p>\n

North Carolina. And I don’t really ABB’s to the Goldilocks theory of anything. I think every<\/p>\n

agents, every place is distinct. But there is nothing<\/p>\n

peculiar about North Carolina with regards to racial disparities and policing, because racial disparities<\/p>\n

in policing are apparent throughout the United States.<\/p>\n

For example, one of the reasons we got into this or political scientists have been talking<\/p>\n

about this was the U.S. Department Justice investigation of Ferguson, Missouri.<\/p>\n

And in Ferguson, the Justice Department found that there was a 70 percent<\/p>\n

increased likelihood that a black driver would have their car searched compared to a white driver<\/p>\n

in Ferguson, Missouri. And this was part of the evidence about the extreme racial<\/p>\n

bias in policing in that community. Well, 70 percent is less than the statewide<\/p>\n

average disparity in North Carolina. So if we thought<\/p>\n

that it was an outrage that Ferguson, Missouri, had such a big disparity, actually<\/p>\n

the disparity in North Carolina, I think is over double. So<\/p>\n

and then. And as we expanded our studies onto more and more communities,<\/p>\n

we did find a few outlying communities. And the one metropolitan region<\/p>\n

that we found that had the highest disparities. And of course, this is an incomplete test because many communities don’t<\/p>\n

have any data. But of all the ones that we’ve been able to look at, it was Chicago, Illinois,<\/p>\n

that had the highest racial disparities. Some of the suburbs around Chicago and the city<\/p>\n

itself. And so I think that has to do with something about<\/p>\n

racial segregation in housing patterns so that police can have a certain<\/p>\n

kind of behavior on the south side of the west side of Chicago and a different type of behavior<\/p>\n

that’s much less aggressive in the center part of the city or the northern<\/p>\n

part of Chicago. Frank? Yeah, for our listeners, Frank gave a talk right<\/p>\n

before we recorded this. And so I reference I mean, from that talk, you noted that<\/p>\n

highway patrolman had very different profiles. I mean, I mean, civil police, they’re<\/p>\n

less likely to be racially discriminatory. They’re less likely to have fruitless searches. They’re mostly stopping people for<\/p>\n

traffic violations and safety reasons. I’m wondering, there’s a general principle we can<\/p>\n

we can draw from this. Is it would it be a good thing if we if we,<\/p>\n

I guess, got the municipal police out of the traffic game? Should there be traffic police<\/p>\n

and other police? Is that is that one way to and to to organize our policing system<\/p>\n

to make it more effective? Well, that’s a good question. The fact that I was<\/p>\n

mentioning that you’re referring to just for the listeners, let me describe it turns out that about half of<\/p>\n

all the traffic stops in North Carolina, as in most states, come from the state highway patrol.<\/p>\n

And of course, the state highway patrol typically works on the interstate highways and the<\/p>\n

major three ways, big highways and especially the interstates.<\/p>\n

And then rural areas tend to be policed by county sheriff’s departments<\/p>\n

in urban areas, have municipal police departments. And it turns out that across<\/p>\n

those three different types of police agencies, the search rates are highly different.<\/p>\n

And as you mentioned, the highway patrol search rate is very, very low, like one fifth, the<\/p>\n

statewide average. And then certain municipal police departments have much<\/p>\n

higher search rates, ten times higher than the highway patrol, for example. And those tend<\/p>\n

to be the more urban police departments. So it seems clear that those police departments<\/p>\n

are using the traffic stops as a means to investigate people, whereas the<\/p>\n

highway patrol is giving tickets to speeders and then moving on to give a ticket<\/p>\n

to another speeder. But they’re not using that speeding ticket as a opportunity<\/p>\n

to search the vehicle. At the same rates as other types of agencies.<\/p>\n

So I think the real question is, are we getting a lot of crime-fighting value in these<\/p>\n

urban police departments by using the traffic code as a way to investigate people?<\/p>\n

Our evidence suggests that it’s it’s really not worth the effort. There’s there’s got to be better ways<\/p>\n

to investigate, get intelligence, find out who is<\/p>\n

involved in criminal activity. But the fact is, it’s tempting for a police department<\/p>\n

because. You have the legal justification to pull somebody over if they break any law,<\/p>\n

even if they make a wide right turn or they touch the you know, they<\/p>\n

they change lanes without signaling, they’ve just open themselves up to a police investigation<\/p>\n

at the discretion of the officer. And so it’s part of the police culture,<\/p>\n

too, to take advantage of these vehicle code violations<\/p>\n

and traffic law violations that might not be serious, but which are technically<\/p>\n

legal violations. Once you break the law, you can be investigated.<\/p>\n

So I think that’s the real question. Are we getting a lot of benefit for this? The typical<\/p>\n

middle class white American is unaware of these behaviors<\/p>\n

unaffected by them. But the typical member of a<\/p>\n

minority community living on certain areas, because these policies are very<\/p>\n

focused, targeted on what the police call high crime areas. The typical person who<\/p>\n

lives there might have a really different range of experiences, much more<\/p>\n

much more experience with intrusive police behavior. Right. Well, we need to<\/p>\n

wrap this up. We’re a little bit over our time, Frank. Our last question we like to ask our<\/p>\n

guests if people are interested in this topic, what other work of political science should they read that’s<\/p>\n

not written by you? Oh, there’s a fantastic book called Pulled Over by<\/p>\n

a trio of political scientists at the University of Kansas. So I<\/p>\n

would recommend pulled over and they did a different type of study of the same question.<\/p>\n

It was based on a survey of people who had recent experience with traffic stops.<\/p>\n

And I would also say that it’s interesting to read some history, because the more you learn about<\/p>\n

the history of policing in the United States and its connection with the history of race,<\/p>\n

these things become part of a pattern that’s been going on for hundreds of years in our country. And<\/p>\n

I think it’s important for all of us to understand our own history. And I’ve been reading a lot<\/p>\n

over the last few years about the history of police saying the history of slave<\/p>\n

codes, the history of where police came from. And it’s it’s quite<\/p>\n

shocking. And so I think would be useful for all of us to know a little bit more about that. All right,<\/p>\n

Frank, on behalf of. So Christine has one more point. Yes. We have to add that suspects’<\/p>\n

citizens just won the Hermann’s you predict best book awards from the Land Court section at APSA. So not only<\/p>\n

is this a book you should buy, it’s an award winning book that you should buy it for your mom, too. That’s a fantastic<\/p>\n

point. Thank you. You’re welcome. Thank you. Thank you, Brooke. Frank, thank you for joining us.<\/p>\n

This has been your Policy Agendas podcast.<\/p>\n"},"episode_featured_image":false,"episode_player_image":"https:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/wp-content\/uploads\/sites\/20\/2018\/12\/The-Policy-Agenda-Logo3-e1545063000555.png","download_link":"https:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/podcast-download\/219\/episode-13-suspect-citizens-with-frank-baumgartner.mp3","player_link":"https:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/podcast-player\/219\/episode-13-suspect-citizens-with-frank-baumgartner.mp3","audio_player":"https:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/podcast-player\/219\/episode-13-suspect-citizens-with-frank-baumgartner.mp3<\/a><\/audio>","episode_data":{"playerMode":"dark","subscribeUrls":[],"rssFeedUrl":"https:\/\/podcasts.la.utexas.edu\/thepolicyagenda\/feed\/podcast\/the-policy-agenda","embedCode":"

Episode 12: Suspect Citizens (with Frank Baumgartner)<\/a><\/blockquote>