S4 E7: The High Cost of Pretrial Detention

                   [THEME MUSIC IN, THEN UNDER]

Risa Goluboff: Today on Common Law, the high cost of pretrial detention, with UVA Law's Megan Stevenson.

Megan Stevenson: There hasn't been any real careful thought about what amount of risk, what type of harm, would justify taking away someone's freedom.

                   [THEME MUSIC UP, THEN UNDER AND OUT]

Risa Goluboff: Welcome back to Common Law, a podcast of the University of Virginia School of Law. I'm Risa Goluboff, the dean. Today we're welcoming back co-host Danielle Citron, a UVA law professor and director of the school's LawTech Center, which focuses on pressing issues in law and technology. Danielle's book “Hate Crimes in Cyberspace” is considered a landmark work linking cyberstalking and civil rights. Her new book, out this summer, is "The Fight for Privacy: Protecting Dignity, Identity and Love in the Digital Age." Danielle, I am so happy you could be here again.

Danielle Citron: It's so great to be back.

Risa Goluboff: So tell me, who are we talking to today?

Danielle Citron: So UVA law professor Megan Stevenson is coming on. Megan is an economist and criminal justice scholar who conducts empirical research on various criminal justice reform issues, including bail, algorithmic risk assessment and juvenile justice. She's going to talk with us today about a new paper she co-authored with University of Georgia law professor Sandra Mayson, looking at how people value their liberty in light of the ubiquity of pretrial detention.

Risa Goluboff: I can't wait to dive in. We will be right back with Professor Megan Stevenson.

                           [THEME MUSIC UP, THEN UNDER AND OUT]

Risa Goluboff: Hey, Megan.

Megan Stevenson: Hey!

Danielle Citron: Megan, we so appreciate your coming to talk with us today.

Megan Stevenson: It's so nice to see you guys.

Risa Goluboff: So this research paper we're going to talk about is called "Pretrial Detention and the Value of Liberty," and it was based on a really interesting study that you did. Can you tell us a bit about it and where the idea for that study came from?

Megan Stevenson: Yeah. So that was my first job out of my Ph.D. program. I was a fellow at U Penn with Sandy Mayson and we shared an office and we were both working on similar stuff at that time, working on bail and working on risk assessments. And so this was just, you know, like a late-in-the-day conversation about risk assessments, about pretrial detention and about how we would figure out the question of how dangerous somebody needs to be in order to justify pretrial detention.

Danielle Citron: Yeah.

Megan Stevenson: So we started off with just a straightforward conversation amongst ourselves thinking, okay, well, if we had to choose between these two bad options, spending time in jail or being the victim of crime, what would we choose? And we kept kind of going back and forth, until we kind of figured out our own balance points, our own equivalency points between these two very different and yet not entirely uncomparable types of harm.

Risa Goluboff: You start with your own intuitions, but then you said, well, let's test them against lots of other people's intuitions. Right? So you created this online survey. What did you ask in that survey?

Megan Stevenson: Think about the experience of being incarcerated. Make a list of things that would be challenging or difficult aspects of being incarcerated and the same thing for crime victimization. The idea behind that is just like, let's get people thinking about it, you know, like let's, let's make sure this is salient and present in their mind what these experiences would be like. And then we asked them a series of open-ended questions. So if you had to choose between a month in jail and being the victim of a burglary, which would you choose? If they chose a month in jail, we raised it. How about a year in jail? Until we had a few binary choice options between these two. We ended with just an open answer question. How much time in jail is equally as bad as being the victim of a burglary? We did this for three separate offense categories — burglary, robbery and serious assault.

Danielle Citron: And what did you discover?

Megan Stevenson: There's a lot of agreement about how bad jail is. People were incredibly averse to spending time in jail. And the median respondent said that a day in jail was as bad as being the victim of a burglary. Three days were as bad as being the victim of a robbery. And a month is as bad as being the victim of a serious assault. If you take these numbers as kind of an estimate of these costs and benefits, you know, on the one side of the ledger, the benefit of pretrial detention is the harm that it averts, the crimes that it averts. The cost is the harms that it creates by locking someone up. What you find is that you've got enormous amount of costs from locking people up and relatively small benefits. And the reason I'm calling these relatively small benefits, I'm not trying to say that crime is not bad. It is bad. It does impose serious harms. But we as a society are very bad at predicting who is going to be the one to commit these harms, particularly when it comes to the more serious, the more violent crimes, like the ones we're talking about here. And so, what the takeaway of the paper is, is that under the framework that is used to justify preventive pretrial detention, almost nobody should be detained pretrial. Which stands in really, really stark contrast to our reality in which pre-trial detention is really the norm.

Risa Goluboff: Pretrial detention is exceedingly common here in the United States. I mean, half a million people are being held in jail, awaiting trial on any given day.

Megan Stevenson: Yeah. 90% of the growth of jail populations over the last 20 to 30 years, millions of people that cycle in and out of pretrial detention every year. I mean, it’s just very, very commonplace, de facto part of our criminal legal system right now. You know, in many jurisdictions, the large majority of arrested are detained pretrial, including misdemeanors. Clearly, practice doesn't match theory. At all. So you're left with this question of, okay, well, what's going on? What's going on in our world that we are somehow comfortable locking up all these people, imposing these massive harms on presumed innocent people, harms that vastly outweigh the benefits. Ultimately, I think it comes down to discounting the well-being of the people that are detained pretrial relative to the potential crime victims. It, it comes down in some way to prejudging them as somehow less worthy either because they were alleged to have committed a crime in the past, because of this risk that they pose in the future, because they are disproportionately Black people, brown people, people of color, the system values the harms they experience way, way less than they value the harms that potentially could have occurred if they were released.

Risa Goluboff: It strikes me that one of the things that's so important about your article is the insight that we're systematically discounting the harms to the person who's being incarcerated. But the second thing that I think is so important about the study as you designed it is: you found a way to get the survey participant to internalize those costs, right? That's what was so important about the survey design was that you wanted them to think: what's the cost to me if I go to jail versus the cost to me, if I'm the victim of a crime.

Megan Stevenson: We felt like if we asked it in a third person — you know, how much harm is created by sending George to jail for a month versus having Maria being the victim of a burglary — what is implicitly going to happen is people are going to look at the person that is facing jail time and think, ohhh, they probably did something bad. You know, we don't care about their suffering, cause they're probably bad, they did something bad. Either based on the crime or based on the type of, you know, othering that happens when somebody who's in a more privileged place in society thinks about the type of person that is facing the threat of jail time. And we wanted to extract away from that. You know, we thought that the cleanest and easiest way is like, think about it for yourself. Think about how you would feel experiencing these different types of harms.

Risa Goluboff: Obviously a big part of what you're doing is saying, well, the costs have to include the cost to the perpetrator or likely perpetrator of spending time in jail. So you have a, a name for your approach to this issue. Um, tell us what you've called it and what it's meant to capture.

Megan Stevenson: So we've been calling it relative harm valuation. Ah, I mean it's pretty straight forward. There's two types of harm. There's crime victimization, and there's jail and we want to compare them. It's a spin on a method that has been used for a long time to provide monetary estimates of non-monetary goods.  It's called contingent valuation. So, you know, if you ever read in the literature, you know, somebody estimated that the cost of a burglary is, you know, $40,000 or whatever it is, or the cost of pollution is X thousand dollars. A lot of these estimates come from a method that's actually very similar to what we used. It's really just asking people, conducting a survey, saying how much would you pay to lower crime rates in your area by 10%? How much would you pay to reduce pollution in your area by 10%? And using that answer to back out the monetary costs of these things that are typically hard to put a dollar sign on because they're not, you know, bought and sold on the open market.

Danielle Citron: I just had an I love you moment, Megan...

[LAUGHING]

Danielle Citron: ... when you talk in the paper about intangible harms that coming up with a money assessment is filled with all sorts of noise and it's complicated. It's so true. That is, we're always trying to put a money value on things. It doesn't mean we can't do it, but because it's difficult to do and feel those harms, unless you've experienced them, the money evaluation you were explaining for all sorts of reasons can be just inaccurate in all sorts of ways.

Megan Stevenson: Yeah. Converting these into monetary estimates introduces noise and bias. People are really comfortable with making direct comparisons of large, bad things. Basically nobody said that they'd spend more than a year in jail to avoid being the victim of a burglary. You know, like, that's by far in the outer realm. But when you convert it to dollar signs, people just get a little bit, you know, there's just like a lot of noise in the estimates partly due to people's preferences, but partly due to people's positions. You know, if you, make $40,000 a year, the might – the amount you might be willing to spend to avoid a particular harm is very different than if you're, you know, Warren Buffet for instance.

Danielle Citron: So you tie all of this to the question of pretrial detention, being put in jail before you've had the benefit of a trial, whether you can't make bail or you're deemed dangerous.

Megan Stevenson: Yeah. The only justification that gets offered for pretrial detention is this forward-looking idea of we are detaining people in order to prevent something bad from happening, from committing a new crime, failing to appear in court, from tampering with the evidence or something like that. It's based on speculation of what somebody might do in the future. Theoretically, that's the idea behind it. That's the legal rationale. That's what people talk about when they say, well, wait, why are we locking up people that are presumed innocent and have not yet been convicted of any crime? This sounds kind of dystopian. It sounds kind of science fiction. But the idea is this consequentialist idea of harm prevention. In practice, there hasn't been any real careful thought about what amount of risk, what type of harm, would justify taking away someone's freedom, putting them in a concrete cage for an undetermined amount of time. And so this was really the motivation behind our research.

Danielle Citron: Your case for the cost to detainees is so profound. And in fact, it sounds like detainment could also create criminals of people. They wouldn't have offended had they not been sort of detained pretrial.

Megan Stevenson: Yeah.

Danielle Citron: Have we lost our sense of mooring? Like, there was some sense that, of course the Constitution limits pretrial detention, because it's so much like the “Minority Report.”

MINORITY REPORT MOVIE TRAILER

TRAILER: I'm placing you under arrest for the future murder of Sarah Marks. Give the man his hat. The future can be seen...

Danielle Citron: Now it seems like we've flipped the presumption against pretrial detention, cause it's dystopian, to having the assumption  it's okay. How did we get there?

Megan Stevenson: Pretrial detention has been a part of our system since the founding. However, originally, there was always this idea of you can be released onto proper amounts of bail. Originally being released on to bail meant being released to a family member who vouched for you.

Risa Goluboff: Just to clarify, that was about making sure they'd come to court rather than about incapacitating them in the meantime, from committing other crimes or harming other people, right?

Megan Stevenson: Exactly. It was very much focused on the idea of appearance. Then in, uh, you know, in the mid-19th century in the gold rush in San Francisco, all of a sudden there were all these people, all these men out there, getting in trouble, but they didn't have family members. You know, they had come out — prospectors, you know — to make their fortune and they didn't have anybody to step forward for them. And so the bankers and the lawyers got together and said, hey, let's, you know, I've got this great idea, we'll come up with this scheme of loaning the money, offering to be their surety, be their bail bondsmen and we'll just charge them some money for doing so. And so that was the origins of the monetary bail system. That slowly evolved over the years. And this idea of incapacitation or preventing crime started to play a larger role. Although it was always very contentious. And it wasn't until the 1980s, when the Supreme Court finally made it clear that they believe that the Constitution allowed to lock somebody up based on the speculation of a future crime. So not just making sure they appear in court, but to prevent the crime that might occur if they were released.

OYEZ: UNITED STATES V. SALERNO ORAL ARGUMENT - JAN. 21, 1987

Justice Rehnquist: We'll hear argument next in number 86 87 United States against Anthony Salerno and Vincent Cafaro.

Risa Goluboff: Right, so this was the 1987 Supreme Court case United States versus Salerno. So tell us about that.

Megan Stevenson: So Salerno was this gangster who, who had a really awful reputation …

Danielle Citron:   Oh yeah.

Megan Stevenson: … was alleged to have masterminded all sorts of awful crimes. And so he was a really great poster child in many regards for this idea that you should be able to detain somebody on the basis of dangerousness, that it was important for the government to be able to step in and prevent whatever crimes would have occurred if this notorious gangster had been released.

Danielle Citron: I have some backstory on the Salerno case.

Risa Goluboff: I want to hear.

          Anthony Cardinale: My client, Mr. Salerno not Mr. Cafaro, had a year earlier been indicted...

Danielle Citron: Tony Salerno was before my judge, the honorable Mary Johnson Lowe in the SDNY, and she issued an order finding that he should be detained before trial.

Anthony Cardinale: I walk into court, I'm handed an 88 page indictment the government tells me they're moving for detention. I asked them why. Well, we don't have to tell you is basically the answer. It's you'll find out, in essence.

Danielle Citron: The reason why he was detained before trial is because there was a white hearse in front of her house. It was a warning, a threat to Judge Lowe's life. I'm on trial before you, watch out.

Risa Goluboff: That's really scary.

Danielle Citron: The Second Circuit apparently overturned her finding, right — said it's a violation of the Constitution to hold someone even though they pose a real, serious risk of danger to the public and including the judge sitting in the case.

Charles Fried: It is said by the court of appeals below that a mere prediction or concern for dangerousness cannot consistently with substantive due process, justify detention of an adult charged with crime.

Danielle Citron: The Supreme Court overturned the Second Circuit. So I just got tingles reading your paper, thinking about what dangerousness meant in the Salerno case and what it meant for my – the judge who I clerked for.

Risa Goluboff: Wow.

Megan Stevenson: That's so interesting.

Risa Goluboff: So with Salerno, it's legal to have pretrial detention. It's not a constitutional violation. But as you say, right, there aren't actually great standards for figuring out when someone should be detained or not. And we have to then, you know, have these conversations that I think are the ones that you and Sandy were having about, you know, how much liberty should be sacrificed in order to prevent crime. You're an economist, right? So this is a kind of cost-benefit analysis that you're using to determine whether it's worthwhile to detain someone. So can you talk a little bit more about cost-benefit analysis and why it's appropriate here and, you know, what kinds of critiques could be made of using it in this way?

Megan Stevenson: So let me be clear: we chose this framework, not because I'm an economist ...

[LAUGHING]

Megan Stevenson: … not because I'm particularly sympathetic or have an allegiance to consequentialist points of view. We think that's the legal framework. We think that's the governing legal rationale that's proffered for pretrial detention. In Salerno, you know, they make it clear this is not punishment. This is not detaining someone on the basis of what they've done in the past. This is purely forward-looking like an ex ante evaluation of what they might do in the future. And, you know, the Salerno court says that you can detain someone when the government's regulatory interest in community safety outweighs the individual's interest in liberty.

Risa Goluboff: It's a cost-benefit analysis. Yep.

Megan Stevenson: When the benefits outweigh the costs, that is when preventive detention is justified. Salerno's by far the most prominent case, but when these cases are discussed, this is the language that people use. Pretty much nobody proffers a “we can detain them because they've met the probable cause standard. They committed a crime in the past. We can detain them because they're clearly bad, or there's something wrongful about this kind of, this risk.” It's just this kind of consequentialist framework that underlies the legal arguments.

Risa Goluboff: Got it.

Danielle Citron: You talk about how we have a hard time assessing risk of dangerousness, like judges and machines or algorithms attempt to do it. Tell us a bit more about that and why it is so hard to assess the risk of dangerousness.

Megan Stevenson: You know, up until fairly recently, these evaluations of the likelihood that somebody was going to commit crime in the future was mostly just based on a, kind of an intuitive or anecdotal basis. And so over recent years, there's been a big shift in the criminal legal system to adopting algorithms, these kind of computer-generated assessments of the likelihood that people are going to commit crime in the future. They're called risk assessment tools. They include inputs such as the criminal record, demographics, age, gender, sometimes socioeconomic factors, employment status, housing status, marital status, things like that. All of these different factors are kind of weighted in different ways based on their correlation with measures of future offending, like re-arrest. And it spits out a number and that number is then translated. Low-risk people can be released. Moderate-risk people need cash bail or electronic monitoring. And high-risk people should be detained. And so, this gets back to the original motivation of the paper. Where do you draw the line in the sand between low and moderate, between moderate and high? That's ultimately the most important question. You know, that's the question that is at the basis of determining when pretrial detention is warranted.

Risa Goluboff: You've written elsewhere about problems with using these kinds of methods for risk assessment. So one of them is the problem of actually getting people on board with using this approach. Right? So can you talk about that? Why won't they use it? How do they use it?

Megan Stevenson: I don't know that they should be using it. They're pretty blunt tools. And some of the factors that go into our risk assessment, things like age, some of the socioeconomic factors, you don't necessarily want to incarcerate somebody just because they're 19 and live in a poor neighborhood. Among these risk assessment tools, which are believed to be the most accurate measures of predicting re-offending, people in the highest risk category really only have 1%, 2%, 3% chance of being rearrested for a violent crime within, you know, a month or two. Now in order to justify imposing the harms of incarceration on a presumed innocent person for a month under the standard consequentialist framework, the legal framework that is used to justify pretrial detention, you would need to avert so many crimes, you would need to avert 30 burglaries. You would need to avert, 10 robberies. You'd need to avert one serious assault in order to justify locking somebody up for a month. And we just don't have that kind of accuracy. We don't know who is going to be committing these crimes.

Risa Goluboff: Last season we had Debbie Hellman, uh, another one of our colleagues on the show and she talked a lot about algorithms and algorithmic discrimination and if people are interested in a deeper dive on that, we've got an episode for them.

Danielle Citron: It was terrific.

Risa Goluboff: Yes.

Danielle Citron: Can you talk a bit about how judges are often risk averse in the sense that weighing all things, that they'd rather incarcerate someone, it seems, pretrial because if they get associated with future crime that then happens if you let someone go, that then they’re just inevitably going to err on the side of caution and detain someone. Tell us a bit about that.

Megan Stevenson: If a judge releases somebody who goes on to re-offend, particularly if they go on to re-offend in a, in a serious manner with a serious crime, that's bad. It doesn't just feel bad, it can result in some real negative press, loss of your job in some of the more extreme circumstances. This type of error is visible, it's very salient, and judges, you know, when you speak to them, they talk about being very sensitive to it. Now, of course, there's lots of other types of errors as well. You know, there's incarcerating people who would not have gone on to commit any serious type of crime or any type of crime at all. And yet you never know who these people are. You don't know what people would have done if they were released. Moreover, there's, you know, there's no media paying attention to these people by and large. They become much more invisible. And so you have these kind of asymmetric errors that might be part of the reason why judges tend to err on the side of, I don't want to say err on the side of caution because this not caution. It's err on the side of one type of harm.

Risa Goluboff: Do you have hope for this paper and what it's going to do to the law? You know, how it affects the way we talk about pretrial detention or the way the law currently handles pretrial detention? What are your hopes and dreams for the intervention that this paper is going to make?

Megan Stevenson: I got an email the other day from a colleague, somebody I don't even know that well, just saying, you know, I read your paper and it stuck in my mind like as a refrain. So every time I teach bail or talk about bail, I think about this idea of what would you choose? You know, how much time in jail is equally as bad as, as this particular crime? What's the likelihood of a person committing that sort of crime? I think that's about as much as a scholar can hope for — that in some way that some idea that they generate or some particular framing of a question kind of forms a little groove in a person's brain and that little groove changes how they process information in the future and hopefully this accumulates to real-world change.

Risa Goluboff: Megan, that was just so fascinating. And I really enjoyed the conversation.

Megan Stevenson: Well, it was great to chat with you guys.

Danielle Citron: Thank you so much.

[THEME MUSIC IN, THEN UP, THEN OUT]

Risa Goluboff: That was just such a terrific conversation.

Danielle Citron: I couldn't help but feel like there's just so much harm that we don't see and particularly for vulnerable people, whether they're detainees or they're female or women of color, you know, as victims. It seemed like a punch to the gut that this was yet more evidence that, you know, we just willy-nilly detain people, we don't think hard about the consequences for the detainees and their families that Megan helped us see. It's as if it doesn't happen. It didn't happen to me, you know, it didn't happen. So to some extent, I thought the insight of the paper was like Derrick Bell's interest convergence theory that like, if you care, if you can put yourselves in the shoes of a detainee, you're really going to kind of get at least more right than you're getting it now, judges who don't even care or see that harm.

Risa Goluboff: I totally agree, and I was thinking about similar things and particularly in my work on vagrancy, I, I wrote about the role of empathy versus the role of sympathy. And so it's not just caring. It's also caring in a certain way, right? And so you can have sympathy for someone, but not really get what it is to be them or not really put yourself in their shoes and not really understand what the harms are. It's so different from empathy, right? What it would be like for this to be me, right? And for it to happen to me. And I think that what's so provocative and stimulating about their study was their choice really, to say, we want to ask you about how you feel about the harm to you if you were to be jailed versus the harm to you, if you were to be the victim of a crime and whether it's easier for you to imagine yourself in the one role or the other, you know, I think that could have some effects on how you value them. And they talk about that critique in the paper, but it's still a given person trying to put the price on both types of eventualities.

Danielle Citron: So making people empathize, right? Versus sympathize.

Risa Goluboff: Yeah.

Danielle Citron: Yeah.

Risa Goluboff: I so appreciated Megan's answer to my question about you know, what, what are your hopes for what this paper does? And I think that's such an important part of the work of a scholar is to, you know, affect just how people frame an issue, how they see an issue. And, you know, I think sometimes people think, well, either as a scholar, you're irrelevant or you're speaking directly to a judge or directly to a legislature, and you're trying to tell them exactly what to do. And I think she really illustrated so nicely that the process by which the creation of new knowledge that happens at an academic institution gets disseminated kind of organically and iteratively and slowly. But it eventually filters out into the way various kinds of decision-makers are making decisions.

Danielle Citron: Absolutely right. The journalist who writes about Megan's work and highlights it on The Markup or, you know, ProPublica and then some legislative aid or judge's clerk starts a conversation.

Risa Goluboff: Yeah.

Danielle Citron: It's not that framing is everything. It's not, but it's really important, and it can be really important as an accelerator for change.

Risa Goluboff: Right. Thank you so much for inviting Megan to join us, Danielle.

Danielle Citron: Thank you so much for having us. That was a treat.

 [THEME MUSIC IN, THEN UNDER]

Danielle Citron: That does it for this episode of Common Law. If you'd like more information on Megan Stevenson's work on the costs of pretrial detention, please visit our website Common Law podcast dot com. There you'll find all of our previous episodes, links to our Twitter feed and more.

Risa Goluboff: In two weeks, co-host Greg Mitchell and I will be speaking to University of California, Irvine's Elizabeth Loftus on the fallibility of eyewitness testimony.

Elizabeth Loftus: So many variables affect eyewitness testimony. All of these can provide an opportunity for new information, sometimes misinformation, to contaminate memory.

Risa Goluboff: We can't wait to share that with you. I'm Risa Goluboff.

Danielle Citron: And I'm Danielle Citron. Thanks for listening.

[THEME MUSIC UP THEN UNDER]

Credits: Do you enjoy Common Law? If so, please leave us a review on Apple Podcasts, Stitcher, or wherever you listen to the show. That helps other listeners find us. Common Law is a production of the University of Virginia School of Law and is produced by Emily Richardson-Lorente and Mary Wood.

[THEME MUSIC UP THEN OUT]