Probable Causation

View Original

Episode 4: Megan Stevenson

Jennifer Doleac

Megan Stevenson

Megan Stevenson is an Assistant Professor of Law at George Mason University.

Date: May 28, 2019

A transcript of this episode is available here.

See this content in the original post

Episode Details:

In this episode we discuss Professor Stevenson’s work on cash bail:

“Evaluating the Impacts of Eliminating Prosecutorial Requests for Cash Bail” by Aurelie Ouss and Megan T. Stevenson

OTHER RESEARCH WE DISCUSS IN THIS EPISODE:

Transcript of this episode:

 

Jennifer [00:00:07] Hello and welcome to Probable Causation, a show about law, economics and crime. I'm your host, Jennifer Doleac of Texas A&M University, where I'm an Economics Professor and the Director of the Justice Tech Lab.

 

Jennifer [00:00:17] My guest this week is Megan Stevenson. Megan is an economist and Assistant Professor of Law at George Mason University. Megan, welcome to the show.

 

Megan [00:00:26] Thank you so much. It is wonderful to be here chatting with you today.

 

Jennifer [00:00:30] So today we're going to talk about a new paper you just put out with Aurelie Ouss about the effects of recent bail reform efforts in Philadelphia. But first, why don't you tell us a bit about your research expertise and how you came to study this topic.

 

Megan [00:00:45] So I have been studying issues related to bail for about four years now, and I became interested in the topic because I started thinking about money bail, which honestly I hadn't thought about a lot before. It's just one of those institutional features of our criminal justice system. It's always been that way. I've known about it, hadn't really given it a second glance. And then I was speaking with a friend and colleague all about it. And all of a sudden, like in this flash, I was like, this is absurd. I can't believe we have a criminal justice system in which you have to pay for freedom that feels anathema to our Constitution, for one thing, and many people argue that it is in fact anathema to our Constitution and also feels very unjust. So that that's part of what got me interested. The other thing is just looking at some basic statistics, people have been very concerned about the scale of incarceration in our country, and I am no exception there. But many people don't realize how much pretrial detention contributes to the numbers of people incarcerated. It actually constitutes one in five of the entire incarcerated population in the United States. That's almost five hundred thousand people, which is a massive number. And according to some back of the envelope calculations that I did once, it's actually more people than is serving time for a drug sentence. So you hear about the War on Drugs and yes, that's certainly a contributor to our incarceration problem. But in terms of numbers, pretrial detention is is actually greater. And just to clarify, I assume most of your listeners out there know this, but pretrial detention refers to the time that somebody is required to stay in jail in between when they are arrested and when their case is disposed, meaning whether they're either convicted or charges dropped or any of the other ways that a case could could complete itself.

 

Jennifer [00:02:58] And what's the rationale? I mean, this this system didn't come from nowhere, right? Why do we have this system to begin with?

 

Megan [00:03:05] We have this system to begin with because people believed and people still believe that it is necessary to ask people to put some monetary collateral down in order to get them to show back up in court. So in order to ensure that people will will come and and face their charges, the belief is either you need to to hold them in jail or you need to have them pay something that they would be at risk of losing if they fail to appear. And this may have been logical back in, I don't know, the 1940s or something like that where people could just hop a train out of town or disappear into nowhere's land. It's actually really hard to truly flee anything nowadays in this, you know, the Internet era and the era of no privacy and shared information. So, the rationale is it was perhaps more justified in a different era than it is now. And I'm not saying it is entirely unjustified. I am, you know, an economist, I believe that people respond to incentives. I personally respond to incentives if I am at risk of having to pay a twenty five dollar charge or something, if I forget to cancel my dentist appointment. You know, I will be more likely to do so. So I I think that there are people who do respond to money and many people respond to monetary incentives. But it certainly is a different world we live in now than in years when monetary bail became more dominant.

 

Jennifer [00:04:43] Yeah, and it's a great point about having a really hard time now to just fall off the radar with our phones being tracked and all that sort of stuff. But, of course, you know, it could be somewhat costly for the jurisdiction to track you down, if you if you don't show up for trial. But so let's dig into a little bit more background just on how cash bail is used. So in general, you've got a judge or a magistrate that sets bail and then you can either pay it or not. So just dig into that a little bit more for us. What's the process?

 

Megan [00:05:12] Exactly. So after arrest, a person is booked into the police station and then they usually sit in a holding cell for a period of time. Hopefully it's 24 hours or less, which I will say is still a long time to sit in a police holding cell. But let's say hopefully it's 24 hours or less. Sometimes it's not, particularly if you're arrested on a Friday and the judge isn't coming back in til a Monday or in a smaller town or over a holiday or something like that. But at some point, you will be brought in front of a judge or a magistrate. Magistrates, by the way, may not even be lawyers. They may be just a person whose cousin is a judge and got them the job, or something like that. And you will have your bail hearing. Now, the bail hearing is I mean, "hearing" is really not the right word because a lot of the time these things last 30 seconds to one minute. They're extremely brief affairs. They often even occur over video conference so the instead of bringing the person into the courthouse where the judge or the magistrate is sitting, there'll be a video feed that connects them from the police holding cell to the courthouse. There is very often no defense counsel present. I would say that sometimes there is some jurisdictions do have public defenders at the bail hearing, but I would say that the dominant model is not to have defense counsel. And in these 30 seconds, the judge will take a look at your- the police report, they'll inform you of your charges, and they will set a bail amount. And so for people that are facing very minor charges and who have no criminal history, they might get they might be allowed to be released without any monetary bail, and that's called release on recognizance or ROR. For other people, the judge might decide that monetary bail is required and set bail amounts that range from a couple hundred to most of them range from a few hundred to a few thousand. But they can skyrocket up to, you know, millions of dollars in exceptional circumstances. And then in some cases, the judge may also require you to agree to certain conditions, like meeting with a pretrial supervision officer, doing drug testing, or something like that. And if you agree to the conditions and pay the bail, then you are released. If not, you are brought from the police holding cell into the jail where you will then sit until you either plead guilty - you know, if you are offered probation, you might plead guilty to go home - or you may sit until your case is is disposed of in some other fashion.

 

Jennifer [00:08:00] And I imagine the judges and magistrates might be thinking about lots of things while they're making this decision in the very short time they have. But what are they supposed to be trying to guess about the person in front of them when they're setting bail?

 

Megan [00:08:12] So it depends on the jurisdiction. But in most jurisdictions, they're trying to figure out whether the person will show up in court without any monetary or other supervisory conditions. And if not, what conditions would be appropriate to get them to show up in court. In many jurisdictions, they're also allowed to consider dangerousness, so the likelihood that somebody is going to commit crime in the future. But the rationale for using a monetary bail as a method of preventing crime is kind of convoluted and shaky. So preventive detention is what it's called when you detain someone pretrial for the purposes of preventing some future crime that that person might commit. So pretrial detention on the basis of dangerousness. And in some jurisdictions, judges have fairly wide latitude to order preventive detention for anybody that they think is appropriate. In other jurisdictions, due to the state constitution, judges have less latitude to simply order someone detained pretrial. And so what they wind up doing is they kind of effectively order them detained pretrial by setting their bail at amounts that are high enough that it's very unlikely that they'll be able to pay. And so, you know, this is something that is generally frowned upon as bad practice. I mean, for one, it's it's a sneaky way of getting around your state constitution if your state constitution has broad rights to bail, which means right to be bailed, right to freedom. In other words, this is certainly not in keeping with the spirit of that constitution. And it's also putting monetary bail in this sort of peculiar role where it is kind of used to keep people that might be more dangerous in jail. But it's also kind of used to incentivize people that are released to come back to court. It's supposed to be used for the latter as an incentivisation method, but in practice, it's used in both ways.

 

Jennifer [00:10:12] Yeah, I agree it is a funny tool to use to try to increase public safety directly and basically you've got the judge trying to, one, predict how dangerous you are and then to predict how much money you have access to in a short period of time. And the likelihood that they're going to get both of those things right seems low or lower than we'd like. So what had we previously known about the effects of this system, both cash bail in general and pretrial detention generally? You've got a bunch of work in this area. This this new paper is very cool, but not the first paper on this topic. So kind of set the stage for us in terms of what answers we'd had going into the study that you and Aurelie wrote.

 

Megan [00:10:52] Well, we know, first of all, that monetary bail can very often lead to pretrial detention and that pretrial detention rates are kind of surprisingly high, even at fairly low bail amounts. And this is because a lot of the people that get arrested are very poor and so they don't necessarily have access to five hundred dollars in order to pay their bail or pay the deposit to either a bail bondsman or to the courts in order to ensure their freedom. So that's that's one thing. We we also know that being detained pretrial has significant downstream consequences for a person. We know that somebody who is detained pretrial is more likely to plead guilty in their case. And that could be, as I was mentioning earlier, if the prosecutor comes to them and says, listen, I'll make you a deal. If you plead guilty, you can go home tomorrow. Yes, you have to do five years of probation. You'll have five years of suspended sentence in which if you fail to keep your nose squeaky clean, if you miss, you know, one of your appointments with your probation officer, we could throw you back in jail in jail or in this case, it would be in prison for up to five years. And you have to agree to all sorts of limitations on your freedom. But if you agree to this, you can go home right now. Now, that's an offer that's kind of hard to say no to. Even if you didn't commit the crime, even if it is a punishment that is inappropriate for the level of crime that you committed. So we know that pretrial detention is is coercive in terms of pleading guilty and in the terms of case outcomes. There is some evidence suggesting that it's actually criminogenic. And that's a kind of fancy word for saying that being incarcerated makes you more likely to commit crime in the future. And this could be because lives get destabilized by losing your job, losing your housing, so on and so forth. Could be for a variety of reasons, could be because of the peer effects of the people that you meet while you're while you're in jail, opening up opportunities for future criminal activity that that you might not have known about beforehand. And we also, you know, connected to this- we know that it can be destabilizing on future labor market outcomes. So there is a bit of a nasty cycle that can happen with pretrial detention where you are detained pretrial if you can't afford to pay your bail amount, you're then more likely to be convicted, you have more serious sentences, you have labor market destabilization. And all of these could potentially lead you to committing more crime in the future, being less likely to be able to afford your bail the next time around. And you're kind of in this revolving door kind of cycle of where each time around poverty disadvantages you further in the process. So this is all stuff that we've known.

 

Jennifer [00:13:42] Can I stop you right there, though? I mean, I think some people listening would say, you know, we've known this for a very long time. We've seen statistics. We can say that people who are detained pretrial have worse outcomes on a variety of dimensions than people who aren't. So just people who are able to pay their bail or not have different outcomes. One of the things that's been really interesting to me as I've watched this literature from the outside for a while is we really have had this just huge influx of new really cool research evidence in the last couple years  trying to to identify the causal effect of these policies on people's outcomes. Right? Because we might worry that people who are detained pretrial are just different than people who are who are not. They're poorer, for the, you know, is the most the most significant difference. So talk a little bit about what the specific papers that that you're referring to here, how are they measuring this effect on people who are detained and distinguishing that from just the fact that on average, they're poorer than people who aren't?

 

Megan [00:14:43] Yeah, so, so sure. So as you mentioned, this is something that people have been talking about for a long time. And there have been a number of papers that have shown that people who are detained pretrial have worse outcomes than people who are not detained pretrial. Now, the debate has been is this because they were detained pretrial or does the causation work the other way around? Because they were at higher propensity of committing crime or more likely to be convicted because they were different in some way, were they then more likely to be detained pretrial? And what looks like a causal relationship where pretrial detention leads to worse outcomes is not actually a causal relationship in that direction, at least. And the recent round of papers have finally answered this in in a matter that I think is fairly definitive. And they they did so using a different technique of identifying the impacts of pretrial detention. Most of these papers used a research methodology that is known as random assignment to judges. Now, random assignment is a term that you hear a lot in RCTs, randomized controlled trials, where you have a bunch of researchers that say, let's randomly assign people to a treatment, other people to a control, and then you can see what the effects of this treatment are on the outcome. Now, in real life, there would be huge ethical violations in randomly assigning certain people to pretrial detention and others, others not. And I would never propose doing something like that. But already in our system, something kind of similar is happening, which is that people are randomly assigned to different bail judges and the bail judges due to their own personality, their own idiosyncratic experiences or a view on justice are more or less lenient. So imagine this. You have 100 people. You split the group randomly in half and you send 50 of them to a strict bail judge and 50 of them to a lenient bail judge. Now, on average, these groups of people are going to be very you know, statistically very identical because we we just randomly split them into two groups in all ways, except the people that had the more strict judge are going to be more likely to be detained pretrial than the people that had the more lenient judge. So if at the end of the day, the group of people that had the more strict judge are also more likely to be convicted, have longer sentences, you know, more likely to be charged and convicted for crimes in the future or are less likely to be gainfully employed in a legal manner, then we can infer that this is due to differences in pretrial detention and not some other underlying differences in characteristics.

 

Jennifer [00:17:42] Right. And so the studies you're referring to find just that, that defendants who are randomly assigned to harsher bail judges are more likely to be detained pretrial, more likely to be found guilty mostly through plea bargains and more likely to commit more crime in the future. And they're less likely to be employed going forward. And we should clarify that this is all for those on the margin, right. So defendants for whom the judge they're assigned to actually matters. So in other words, where the harsh judge and the lenient judge might disagree about whether you're a flight risk and assign you different levels of bail. So if you're someone who would never be assigned cash bail or would always be assigned high cash bail, regardless of which judge you're in front of, then these findings might not apply to you. But I agree that at this point it's clear that being assigned high cash bail has important negative consequences for at least a meaningful subset of defendants. So let's dig into your new paper. Tell us about the policy reform that you are studying here.

 

Megan [00:18:40] All right. So a little over a year ago, Philadelphia elected a new district attorney named Larry Krasner. Now, Larry Krasner is a very unique district attorney. Usually district attorneys, you know, they were prosecutors, they were judges. They're often fairly tough on crime. This is not Larry Krasner at all. Larry Krasner had never prosecuted a case in his life. He's a former defense attorney. He is a radical, one would say. He has sued the Philadelphia Police Department, something like 70 different times for abuses. He has been very involved in Philadelphia's criminal justice system. But as an outsider, pushing to change what he what he saw as as corrupt and unjust practices within the system. So he gets elected, he takes office. And one of the first things that he does is he declares that he is he's going to instruct his his line prosecutors to no longer ask for cash bail for defendants charged with 25 different offenses. Now, these are nonviolent offenses, but they're not all trivial offenses. They're not all misdemeanors, they do include some felonies. And numbers wise, they constitute about two thirds of all of the cases that go through Philadelphia each year. So it's a pretty big intervention. It's a pretty bold move to declare no longer asking for cash bail for these people. But to begin with, it's not immediately clear that this would change bail setting practices at all, because in Philadelphia, the magistrates that set the bail, they're not employees of Larry Krasner. They do not work for the district attorney's office, they work for the judiciary. And they are the ones making the decision. The prosecutor's role is only advisory. They can make requests. They can make suggestions. But they have no direct influence over the bail amount. And, you know, Krasner ran on a pretty radical platform of change. And it's not immediately obvious that all of the different criminal justice agencies, the courts, the police and so forth would get on board with his vision. So the first thing that we tested was, does this have any effect? Progressive prosecutors are kind of a thing right now. They're being elected in big cities all over the country. And we have one in Chicago, in Houston, Texas, in San Francisco now in Boston. Are they going to be able to make a difference in this one dimension, the bail reform movement that many people care about so much? So the first thing that we find is that it did actually have an effect. It led to an increase in the likelihood that defendants would get released on recognizance. That went once again, that means released without being required to pay any monetary bail or agree to any sort of supervisory conditions. There's about a 12 percentage point increase in really it's on recognizance, which is mostly coming from a decline of low monetary bail, a decline of monetary bail that's under five thousand dollars.

 

Jennifer [00:21:52] Tell us a little bit more about how you're measuring this. So you have the reform, you have Larry Krasner is elected, he issues this new instruction to the prosecutors and that starts on a particular date, I gather. And then so how are you using that timeline then in your actual empirical analysis to figure out what the impact is on the eligible population?

 

Megan [00:22:15] Yeah. Thank you. So he ran on a platform of bail reform, among other things, so people knew this was coming, but they didn't know exactly when. And he announced this bail reform platform on February- he announced that he was going to implement these changes on February 21st. And the very next day, those changes were in practice. Those those changes were in place. And so what we look at is we look at abrupt, discontinuous changes in bail amounts, in pretrial detention, in court appearance rates and in rearrest rates that occur for defendants that were arrested right before this reform was put into place and right after. What we're leveraging is the fact that people that were arrested right before should be pretty similar to people that were arrested right after in all ways, except for their exposure to this new policy change. Formally, what we're doing is we're doing a regression discontinuity with time as the running variable.

 

Jennifer [00:23:26] Right. So basically, you're exploiting this natural experiment where you have someone who is arrested or charged the day before he makes this announcement, who looks very similar to someone who's arrested or charged for the exact same crime the day after. But the day the guy who's arrested the day after suddenly has no cash bail.

 

Megan [00:23:44] Exactly.

 

Jennifer [00:23:45] And so so you can be you can then see what's the impact of this kind of reform. So what what data do you have at your disposal to be able to dig into this?

 

Megan [00:23:56] So we used data from Philadelphia courts. So they keep detailed records on each case, describing the time that bail was set. The amount of bail, even the magistrate that set the bail, they record if the defendant fails to appear for future court dates. They record if they have new charges filed against them in the future. They record whether a person has charges filed against them in the past. And this is data that is web scraped. It's publicly available data, but they don't make it super easy to get at. You can't just, like, download a big Excel file, which is probably a good thing. It's contains sensitive information like name and birth date and stuff. But we web scraped with with the help of a wonderful computer science friend who did some of this work for us. We web scraped a large number of court records and did the analysis using that data.

 

Jennifer [00:24:52] So what you find in the first set of results is that a bigger share of people are released on their own recognizance, but that's due mostly from the removal of the collateral requirement. Is that right? And so talk more about the first level effect and what that means for what the mechanism is that we should have in mind for why this reform might affect subsequent behavior.

 

Megan [00:25:17] Yeah, so there's a big increase of release on recognizance, and it comes largely from a decline in this low monetary bail. And I hesitate to even say low because five thousand dollars is not a trivial amount of money. You do in Philadelphia, they have a deposit system. So if your bail is five thousand, you actually only need to come up with 10 percent of that to get your freedom. So that's five hundred dollars that you'd have to pay, which is somewhat more within reach, but still a a not insignificant sum for people that are living hand-to-mouth. And there's also a decline in being released on what they call non-monetary conditions, which can include a variety of things, including being under supervision of a pretrial officer or or curfews or things like that.

 

Jennifer [00:26:11] What's the change in the number of people who are actually released, pretrial versus detained?

 

Megan [00:26:18] So it led to about a five percentage point decline in the number of people that needed to spend a night in jail. But it didn't affect longer jail stays in a manner that is statistically significant. Mostly what it seems like it has done is targeted people that would have been able to come up with the money sooner or later and just got them out of jail a little bit faster. And I think you could argue about whether this is a, you know, a success or a disappointment. I think it probably depends about what your expectations are. But I would say it's a success that there are now thousands of people that don't need to spend the night in jail and just jump ahead to the rest of the results. We don't find any real adverse consequences, but it certainly isn't it certainly isn't a reform that is having a huge effect on pretrial detention rates beyond just that first night.

 

Jennifer [00:27:16] Yeah and I think if I'm remembering right from the paper, you also mentioned that in Philadelphia they do have this deposit system. But you also only get like 70 percent of the money back if you show up to court. So it's still like there's a cost, there's an actual financial cost here to having the bail set in your case. So, yeah, that's an interesting additional benefit for this this group. That would be fascinating to see added up over time how much money is this policy saving defendants?

 

Megan [00:27:44] Absolutely. And even the money that they get back you that's still coming up with a large sum of money. And tying that money up for the couple of months or whatever it takes until your case is disposed, is it's it's a non-trivial burden.

 

Jennifer [00:27:59] Right, and probably very stressful for a lot of families. OK. So I agree with you that some people might not think of this as being a, you know, the dramatic change that they had hoped for. Which is probably going to be the story of a lot of criminal justice reform as things are getting pushed through. So change is hard sometimes, but it's really nice, I think, for research purposes, because it seems like you guys have done it. You have an a situation where you've you've really almost entirely isolated this effect of the monetary incentive. Right? So usually you'd wonder, like if you if you eliminate cash bail, you don't have the monetary incentive, but you also- people aren't locked up anymore. So both of those things could have big impacts on their lives. And and understanding the impact of just the monetary incentive to show up show up back in court, I think is really important. And I agree that that's that's a piece where especially for economists, we we tend to assume that everyone responds to incentives. And the question here is just, you know, what's the magnitude of the effect and and who's going to respond more than others? Right? If you don't have any money, then, like, there's- it's not- maybe going to have less of an effect on on your decision making than if you have a ton of money or vice versa. I guess it's kind of hard to tell which way that would go. But but it is really I think it is a really cool aspect of the study that you can you can speak directly to the effect of this monetary incentive.

 

Megan [00:29:26] Yeah, exactly, it's it's a it's a unique opportunity since since the main change is just releasing people without monetary bail or other conditions of release and not on the number of people that are sitting in jail until their case is disposed. You can really test this theory that has been driving our entire bail system, you know, for really for centuries, to be honest. You know, this theory that it's uh monetary bail is the is pivotal in getting people to show up to court. And what we find is that there was no there was no statistically significant increase in the failure to appear rate. So that's failure to appear is a it's a it's a phrase people use for not appearing in court during one of their appearances. So no statistically significant change in failure to appear, in recidivism, or in serious recidivism, which is being charged for a new graded felony, which are the slightly more serious types of felonies within the four months after the bail hearing.

 

Jennifer [00:30:37] And so, I mean, I think we'd expect the monetary incentive to have the biggest effect on the FTA, right? I mean, there's not a clear channel through which you would expect an effect on recidivism unless people are actually being locked up, but-.

 

Megan [00:30:49] Exactly but at the same time, I mean, there is really a lot of disinformation campaign that is going on around monetary bail. There is literally there's been quotes in the Philadelphia Inquirer where anonymous sources that the police department have blamed Krasner's bail reform for the spike in homicides that are going on. The big California bail reform bill, just to give a little background information, California last year passed a very major comprehensive reform of their bail system where they outlawed monetary bail. They banned monetary bail. Now, the bail industry got busy and they got they got a bunch of people out there on the street. They put a bunch of money into gathering petitions, gathering signatures for a petition, and they were able to get enough signatures that this reform bill is now put on hold. It's being taken to direct vote by the voters this fall and have really put the brakes on what could be a serious reform using, among other things, the idea that getting rid of monetary bail will lead to a huge spike in serious crime. So. While, as an academic, I would think, yeah, I don't really expect recidivism effects. If anything, I expect, you know, an increase in failures to appear in court. In terms of the public policy, the kind of conversation that's going on, I think it's it's worth noting. It's worth noting that there was no effect on recidivism.

 

Jennifer [00:32:25] Yeah. And well, I mean, I think the broader conversation people people probably are assuming that this reform led to a big reduction in detention, and certainly in the broader reform pushes I think that's that's the end goal is is to release more people and not not detain as many people pretrial. And then, you know, I mean, there there is an incapacitation effect that we might expect where if we're locking people up, they don't have the opportunity to commit more crime. We can talk about, you know, the whether that's ethical or legal to lock people up based on like whether we think they might be dangerous or not. But but I think as as academics and people who study this sort of data for a living, we would expect sort of mechanically for crime to go up at least a little bit just because people are out on the street. But, yeah, in this case, it does seem like FTAs are where we'd expect all the action to be. And as you said, you don't you don't find any significant effects and I think in terms of the coefficients you find, it looks like around a 25 percent increase, it's just the standard errors are huge. And so I guess I wanted to get your thoughts on just to what extent do you think this is just a statistical power issue? To what extent, you know, even if you took those numbers at face value, how many people are we talking about here? What's the cost to- the cost could still be super low, right, even that that number. But yeah. But 25 percent isn't nothing. So I just wanted to get your sense on, you know, to what extent are you pretty confident there's nothing going on here versus a statistical power issue?

 

Megan [00:33:56] So so the the actual coefficient suggests that right after this new policy was put into place, the failure to appear rate increased by 1.9 percentage points. So, you say 25 percent. It's a little less, 20 percent.

 

Jennifer [00:34:14] I mean the baseline is low.

 

Megan [00:34:15] That is true. The baseline is- the baseline is low. So the the average failure to appear rate is 10 percent. And so, you know, 1.9 is you know, it's 9-, 19 percent. Um, the- I have a couple of things to say about that. First of all, I as I started out this conversation, I assume that monetary bail incentivizes at least some people out there to show up for court. I just, that- I hold those priors very strongly. And so I would not argue that what we are showing is that monetary bail has no effect whatsoever. When I talk about this, the language that I use is that we fail to detect an increase in failure to appear. And what I mean by that is a couple things. One is obviously it's not a statistically significant effect. It's a it's, you know, what what we call a null effect. But also that in terms of practical importance, a 1.9 percentage point increase, yeah, it's large relative to the baseline, but it's it's small enough that I think most people would look at it and say, even if that number was statistically significant, it doesn't matter that much. Not that there's no cost. There are costs every time somebody fails to appear, you know, it's- there are hassles in the criminal justice system. But it's it's not you know, it's certainly not earth shattering.

 

Jennifer [00:35:46] Yeah, well, I think especially in relation to the other papers that you that both you and others have written, showing these massive costs to especially pretrial detention. But the big question mark, I think, in all those studies well is well what are the benefits, though, that we'd be foregoing if we got rid of this? And so I think one contri- big contribution of your paper here is starting to get at what what the benefits of cash bail are. And I agree, they seem to be pretty small.

 

Megan [00:36:12] Yeah. The other the other thing that I want to mention is so we're doing there are we look at the impacts on defendants who are eligible for this reform. And these are the group of people that are charged with one of the 25 offense categories for which Krasner said, we're no longer going to ask for monetary bail for these people. Now, we also do the same experiment with defendants who are charged with ineligible offenses. Mostly these are violent offenses, but they're also some nonviolent but more serious offenses. And this is what we call a sort of placebo test. So we don't really expect to see any changes for this group because they there was no policy out there that that was targeting them. And along with this idea, we'd- we don't see any change in bail practices or in pretrial detention rates. When you compare people that were arrested just before the reform to people that were arrested just after the reform. You do, however, see, you know, the coefficient on failure to appear is also not statistically significant, but it's actually almost identical to the coefficient among the ineligible offenders. It's it's 1.3 percentage points instead of 1.9 percentage points. So it's entirely possible that there was just, I don't know, uh, you know, just something going on that you just random noise, you know, some reason that for, you know, they. Maybe they, you know, maybe they all got assigned bail during a time that was, you know, inconvenient or I don't know. I don't know, maybe it's like right before the holidays or spring break. I mean, I have no idea. I have no idea what it is. But I'm just saying that, like the-

 

Jennifer [00:38:06] It's not like a flat zero for everybody else and then a 1.9 for for the eligible cases.

 

Megan [00:38:10] Yeah, yeah. These these types of specification choices can, you know, they can fluct- you know, they're a little bit sensitive to noise, they can fluctuate. I don't want to interpret too much into that 1.9 percent.

 

Jennifer [00:38:25] So let's talk a little bit about Philadelphia relative to other places and to what extent it seems that we want to extrapolate from Philadelphia's experience. So I was at the Public Choice Society meetings last week and saw Josh Page present some qualitative work he's been doing based on having worked as a bail bondsman for some period of time, which was really cool and never- and the sort of thing that economists never do. And he talks about how under the current system where, you know, public defenders are so under-resourced, bail bondsmen often take on something of a social work role, providing information about the system, as well as making sure defendants get to court. And so, you know, Philly doesn't have much of a bail bondsman system now or even before Larry Krasner came along, essentially because they handle all those financial duties in-house. And so this just as I was reading your paper, it made me wonder if the effects of moving away from cash bail might be more detrimental in other places if we don't first find a way to provide that that social support in some other way. I just wanted to pick your brain about that. What are your thoughts on that that issue?

 

Megan [00:39:32] I think that sounds I think that sounds plausible. I mean, of course, we would have to, you know, have to do the research to know for sure, but. You know, money bail most of the time when people talk about it, they talk about it as this collateral financial incentive mechanism to get people to show up. But as you mentioned, in many jurisdictions, a lot of people get their money bail from a bail bondsman. And these bail bondsman act not only as lenders, but they act as kind of like pretrial supervision officers. They're involved in a variety of different ways in the defendant's life, providing information to them, hunting them down if they fail to appear in court and so forth. So Philadelphia is a little bit different because they have this deposit system where the court essentially acts as a bail bondsman by saying instead of having to pay the whole bail amount, you just pay 10 percent. If you fail to appear, you are liable for the you owe us the rest of it. But if you show up, we will return most of your deposit. As I said, as we were talking about previously, they still hold on to 30 percent, but they return most of it. So we are really testing the kind of classical iconic role of monetary bail in in financial incentives. In practice, the way it works in other jurisdictions is that people don't really have financial incentive in the same way. They uh, they gave a non-refundable deposit, most of the time nonrefundable deposit, to a bail bondsman in the form of either cash or collateral and the bail bondsman puts up the rest of the money for them. Now, once they put this money up, if a defendant fails to appear in court, it's the bail bondsman that loses money. It's not the person that loses money. So in a way, in a in a a a jurisdiction where bail bondsmen are really prevalent, the mechanism is entirely different. The defendant has no personal financial incentive once they've coughed up that nonrefundable deposit. But they do have this bail bondsman that is going to, you know, either optimistically make it help them navigate the system or, you know, or in the in the bounty hunter context, you know really hunt them down and drag them, drag them in. So, you know, it could be true. It depends on on how important that latter role is. You know, it could be true that in other jurisdictions that if you get rid of monetary bail, what you're essentially doing is getting rid of this- what is I think the best way of looking at it is kind of a privatized pretrial supervision service. And if that is if it's effective, you could see problems. But, another possibility is that in other jurisdictions, you could have even, you know, less adverse consequences because if the the other services that the bail bondsman provides are not that important or that helpful, then they're not really providing anything of value.

 

Jennifer [00:43:10] Or if they they could be actively detrimental even.

 

Megan [00:43:13] It could be actively detrimental. And you don't have the you don't have the financial incentive tool that monetary bail is supposed to be, so.

 

Jennifer [00:43:23] Yeah, I agree. We need, this is- this would be a great topic for someone to study. We need another experiment in a different city and then we can figure out the answer. OK. So based on on your research, the current paper, your previous research and then other work that's in this area. What's the takeaway here for policymakers who are thinking about how to reform their policies around cash bail, pretrial detention, the pretrial context more broadly. There's a lot of, you know, talk in across the political spectrum these days about reforming the system. When people come and ask your advice, what do you tell them?

 

Megan [00:43:59] Well, the takeaway- the policy takeaway from this particular paper is that you can stop asking for monetary bail from a sizable fraction of your defendants without any real adverse consequences. And you know the the study is limited to defendants that are eligible for the reform. But that's not a huge limit because as I mentioned previously, it's actually two thirds of all defendants. So, you know, it seems like a no brainer. Based on this Philadelphia research, you know, if this Philadelphia research holds true in other jurisdictions, it would be a no brainer to to wind back the use of monetary bail, at least somewhat. Now, it's possible that if you take a more radical reform like California is proposing, that there might be more adverse consequences. You know, if you extend it to all defendants, you might start seeing failure to appear rates creep up. You might start seeing pretrial crime rates creeping up.

 

Megan [00:45:26] But I actually just wanted to say something about the pretrial crime thing. That's what people get most concerned about. That's the real that's the real fear. And it's, and I, you know, obviously, crime is important, crime victimization is is, you know, is horrible. But the pretrial context is a little bit different because if these people get convicted, you know, many of whom do get convicted, they're they're going to be doing the time anyway. You know, they if they get if they're detained pretrial, they get credit for time served. And so you're sort of just shifting the you know, if you if you reduce the pretrial detention rate to a certain extent, a lot of what you're doing is you're just shifting the incarceration period forward in time. So to the extent that getting rid of monetary bail entirely might make crime in the month or two after arrest more common, it might make it in the future less common because people are going to be incarcerated further in time.

 

Jennifer [00:46:38] Of course, I mean, the scenario people have in mind is someone who's on a murderous rampage or is actively threatening people out in public. And now we have these these reformers who are saying, no, no, wait, we'll give them their day in court. And in the meantime, you know, 10 more people get killed. And presumably, none of these reforms are set up in such a way where the judges would not have the discretion to lock that person up. And the question is just how many people are actively like massive public safety threats in such a way that would be difficult for a judge to see during that that initial hearing? Is that right?

 

Megan [00:47:21] Yeah, I mean, there are certainly concerns about correlated time, crime being correlated across time. So in what you're saying, you know, if they were arrested for something now, that might be because there is something is hot in a certain way. You know, like a gang activity is hot. And so you need to incarcerate them right now and prevent crime right now. But I think I raise the other point just because I think it's underappreciated that in the pretrial context in particular, it's not like you are if you're reducing pretrial detention, some of the time that you reduce during the pretrial detention period, you are tacking on into the later the later era. And so it's not quite as- the crime effects when you look just at pretrial crime would overstate the total effects on crime.

 

Jennifer [00:48:02] Yeah, I think that's I think that's a really important point. I think the you know, it's kind of a broader philosophical point where like even if even if we did see that it it mattered a lot. We kind of have this ethical question about whether whether it's the right thing to do to lock people up based on crime that we think they're going to commit. And that's that's something that, you know, lawyers and ethicists can argue about. It is not really something where economists' toolkit can can provide much can shed much light beyond just telling people what the costs and benefits are. But but, yeah, it's definitely another aspect of this debate that's really interesting. Any other any other thoughts on kind of advice about reforming pretrial context or anything else that comes up in the in this debate?

 

Megan [00:48:51] Just evaluate, you know, any I think jurisdictions, any time they adopt something, you know, they hope certain things would happen. You got to you got to evaluate and see if if what you hoped to achieve actually did. And if you if you didn't achieve it, go back to the drawing board and and keep working. Because I think being able to predict the impacts of a policy are always a little bit challenging and the more research we have, the better.

 

Jennifer [00:49:17] And along those lines, so, I mean, it's one way to do this and people like us do this a lot is sort of following what policies are passed and then trying to convince someone to to roll out that policy in a way that allows us to to study what the impact is. But the other potential route is for is for researchers to sort of think about what the big open questions are and then beg someone to go run that experiment. So so in your mind, like in this context, what are the big open questions that we need to be answering here?

 

Megan [00:49:50] We certainly need to be looking at what are the impacts of broader reaching reforms, you know, like I was talking about with California, if that if that goes through, what what are the impacts of that? And I don't just mean in terms of these potential adverse consequences, like failure to appear and crime. I mean, are the reforms going to affect the fundamental inequities that I brought up in the beginning of this conversation that brought me to do research on bail and has fired this this big bail reform movement that's going that's going on. One concern is that there are enough loopholes in the reform legislation that gets passed that you could see really no dent in the pretrial detention rate, which means that we're gonna continue to incarcerate just massive numbers of people with no due process on the basis just of, you know, a police officer thought that it was likely enough that something happened to justify an arrest. Are we going to still see the the inequities in which people who are poor, people who are brown skinned, get detained at at massively higher rates? Or is the is the reform going to lead to what we hope to see, which is is deeper and more fundamental changes?

 

Jennifer [00:51:15] My guest today has been Megan Stevenson from George Mason University. Megan, thanks so much for joining me.

 

Megan [00:51:20] Thank you.

 

Jennifer [00:51:26] You can find links to all the research we discussed today on our website, probablecausation.com. You can also subscribe to the show there or wherever you get your podcasts to make sure you don't miss a single episode. Big thanks to Emergent Ventures for supporting the show. Our sound engineer is Caroline Hockenbury. Our music is by Werner and our logo is designed by Carrie Throckmorton. Thanks for listening and I'll talk to you in two weeks.