Episode Transcript
[00:00:00] Speaker A: Imagine a justice system built on rigorous evidence, not gut instincts or educated guesses about what works and what doesn't.
More people could access the civil justice they deserve.
The criminal justice system could be smaller, more effective, and more humane.
The Access to Justice Lab here at Harvard Law School is producing that needed evidence. And this podcast is about the challenge of transforming law into an evidence based field.
I'm your host, Jim Greiner and this is Proof Over Precedent.
Welcome to another edition of Proof Over Precedent, the Access to Justice Labs podcast. In our effort to try to transform the legal system into an evidence based field. I'm Jim Greiner. I'm your host for today's podcast, and I am the faculty director of the Access to Justice Lab. And we're going to be talking today about this will be the first of several sessions that we're going to record or podcasts that we're going to record about our expungement study, which is a randomized field experiment that took place in Kansas and will, as we'll hear, like almost all of our field research that was long lasting, took several years to plan, to plan and implement and randomize, and then several more years to collect the data and to analyze it. And so we'll be walking through that timeline. I have the basic research team that has been on this study since it's near its inception or since its inception. And so I'm going to go around to each of you and ask you to introduce yourself. So we'll start with you, Renee. Who are you?
[00:01:36] Speaker B: Hello. Welcome. Thank you for having me on again. My name is Renee Dancer. I'm a researcher at the Access to Justice Lab.
[00:01:44] Speaker A: And Matthew, who are you?
[00:01:46] Speaker C: Yeah, thank you for having me on as well. I am Matthew. I am A2L at Columbia Law and I'm a research assistant for the Access to Justice Lab.
[00:01:54] Speaker A: And when you started this project, you were where?
[00:01:56] Speaker C: I was an undergraduate at Harvard. I started during my sophomore year back in 2022.
[00:02:02] Speaker A: Super. And Patricia, who are you?
[00:02:06] Speaker D: Hey, great to be here. I'm Patricia Gantzert. I'm a data analyst at the lab.
[00:02:10] Speaker A: And Ryan, who are you?
[00:02:12] Speaker E: Hi, I'm Ryan Halen. I am also a data analyst here at the lab.
[00:02:17] Speaker A: Terrific. There was a lot of data and a lot of data processing, and so we had both of you on board for the duration of this study. So. So as we continue to discuss this study, I'd like to just mention who funded the Access to Justice Lab. We received generous funding from the National Institute of Justice.
So the scientific arm of The Department, United States Department of Justice. And we're most grateful for that. For the Chan Zuckerberg Institute, czi, which is no longer with us. The Chan Zuckerberg folks reorganized and became a different set of foundations, but we're most grateful for their funding and for the from the Charles Koch foundation, who have funded this study. In particular, our ability to investigate not just outcomes after two years, but outcomes after five. So this study will be consisting of a longer term analysis as well as a short term analysis, and we're most grateful to them as well. One of the more interesting things about this study was the fact that it was a coalition of folks who represent interested people all across the political spectrum. And so we're especially grateful for the funding that we received and for the interest that folks across the political spectrum have shown in evidence based thinking and attempting to transform the legal system into an evidence based field. Let's just start with the basics. Renee, can you give us the broad outline? What is this study about? What were we trying to find out?
[00:03:45] Speaker B: With this study, we wanted to understand the effect of expungement or record clearing on housing and employment stability.
But to do that, we couldn't really randomly assign whether or not someone received a record clearing remedy.
So we needed to identify a way that we thought individuals would be more likely to receive a record clearing remedy and randomly assign that. And so what we did was work with a legal services provider, Kansas Legal Services, and we randomly assigned either full scope representation, which we thought would be the condition in which people would be most likely to receive record clearing, or materials, which was the condition that we thought folks would be less likely to be successful in their record clearing attempts.
[00:04:44] Speaker A: Superb. Just to unpackage some of those concepts for those who are fully nerded up and understand some econometrics. Ryan, we call this an instrumental variable study, is that correct?
[00:04:55] Speaker E: Yeah. Yeah. We could add the term headache inducing if you wanted to. It makes it more complicated to be sure, but yes, this is in this sense. We're using assignment to legal representation as an instrument for expungement or record clearing itself.
[00:05:11] Speaker A: Super.
And Renee, going back to you, just to unpackage some of these terms, there's a broad category of criminal justice record clearing. There's more than one type of criminal justice record clearing, and we were investigating only one of those types. Can you unpackage 2 broad categories of criminal justice record clearing?
[00:05:30] Speaker B: Sure, I can try. Because you're right, there's lots of different jargon and also different remedies. This record clearing that's available by legislation in Kansas is suppression of a record public view and it is applies to convictions. So meaning you may have or you an individual was found guilty and they served their whatever sentence was passed down to them and they finished that and then after a period of time they were eligible for record clearing. There is also and suppression from public view means that it's no longer supposed to be searchable or findable by the public either in hard copy records or in docket form.
But it might be available to those to a subset of individuals dictated in the legislation which are usually some law enforcement or licensing bodies to for future purposes of determining different things in whether or not you're eligible for some sort of additional program in the criminal justice context or eligible to continue with a certain license.
The other type of record clearing is complete destruction.
And that means that the record is completely destroyed, just like you might just like it sounds that it's no longer available to anyone. I'm actually not aware of a jurisdiction that actually does that. Fine, say they do that. But I'm not aware of a jurisdiction that actually does that. There's also different within the record clearing space we have Kansas who suppresses convictions. Many other record clearing remedies simply suppress non convictions. Meaning your case was dismissed, it was never prosecuted. You were actually went to did the full trial and found not guilty.
Those you were arrested, but nothing more happened after that.
Those are all convictions. Most record clearing remedies apply to non convictions. And then there's a to complicate things even further, there is petition based clearing and automatic clearing. Petition based clearing means you as the defendant must take some action with a written piece of paper and ask the court to clear your record. Automatic record clearing is the process in theory whereby the court knows what the law is and identifies cases that are eligible for record clearing based on the law and clears them as a matter of course without an individual defendant needing to do anything or take any action.
[00:08:28] Speaker A: There were a whole lot of different distinctions built in that you discussed as is necessary in this type of discussion, destruction versus suppression. And in this one, this study, it was a suppression remedy, not a destruction remedy. Then there's the scope of the records that get that get. And so in this study it was convictions that we were focused on primarily as opposed to records of arrest without charges or charges without convictions.
And then you mentioned perhaps the most important one of all, which is the automated record clearing remedy where the person who has the record doesn't have to do anything versus the one that they have to petition For. And am I correct that this one was the petition based?
[00:09:09] Speaker B: This petition based, yes.
[00:09:11] Speaker A: Which allows us to use the legal aid in order to. As an instrument to try to figure out the effect of expungement. Renee, sticking with you for another question here. So you mentioned that there's a petition that. That. That somebody has to file. Is that a simple process that you have to initiate that we're actually going to do a podcast on this separately on the process itself with a different set of co authors. But just to briefly summarize, is this an easy thing for somebody without a lawyer to get through?
[00:09:40] Speaker B: It is not. And to preview that. That podcast, we actually did write an article about this. So we can add that to the notes of this podcast and maybe. And we'll certainly talk more in depth about it at the other episode that you alluded to. But it is not. There are a number of separate documents. It's not. It's. To call it petition based as if that's a singular document would be wrong. It is. That's a nice name, but it's not the reality.
And then there's those petitions ask for a lot of information that an individual working on this on their own would struggle to find. But then, but in reality, the practitioner who works in this space all the time might recognize that some of this is unnecessary. So for what I find to be one of the more egregious examples is asking for the arresting officer's badge number.
And what is so egregious about that is that you are not even eligible for record clearing until after you have completed your sentence. And then also a waiting period has passed. And that waiting period is dictated within the law based on the type of offense you're attempting to clear. And so it's been a number of years, so you probably didn't remember that officer's badge number if you ever took it down the next day, let alone a number of years later. And so it's items like that and then that those are just the. That's just the petition itself. The multiple documents themselves are difficult to navigate. But there's also a filing fee and there are things like needing to create proposed orders both for your petition to be granted and also one for your petition to be denied. And while that's commonplace and common practice, if you are a lawyer, if you are an individual attempting to win your case, it seems really strange that you should give the judge an order that actually denies your case.
[00:11:33] Speaker A: Super. So a complicated process. And then the end goal is, in Kansas, the suppression of the criminal record.
And this is a phrase we hear a lot in public, just public parlance, the criminal record or my criminal record or something like that.
But I wonder whether there is actually such a thing as a person's criminal record or my criminal record. And Patricia, I want to go to you and Matthew to you shortly after Patricia, starting with you, is there such a thing as a person's criminal record, singular? I have my criminal record. And we can go to one location in the government where it's held and fiddle with it or suppress certain parts of it, not others.
[00:12:16] Speaker D: Yeah, I would say it's better to think of it as criminal record. There are records of arrests, there are records of confinement. There are records of all kinds of things, but they're not stored in one place. They're not all under the jurisdiction of the same people. And some of them are accessible to the public, some of them aren't.
They might be under different names if you've been married, if you've changed your name. So there are records out there. But finding those records and doing something with them is not necessarily a straightforward process.
[00:12:43] Speaker A: And when you say that they're not all accessible to the public, some of them are semi accessible. Right. In other words, some of them you could look up just by hopping on a website, but others you had to take other steps for.
[00:12:55] Speaker D: Yeah, and obviously this is also. We work in Kansas.
[00:12:59] Speaker A: Right.
[00:13:00] Speaker D: So this is specific just to one state. So you can imagine the complexity of this as you're expanding out to the nation. But we download some criminal records from a Kansas database where they just cost originally $20, the repealing cost be increased, the cost of $30 per search, where you can search by students, name your birth date if you know the Social Security number. And that's pretty publicly accessible. But not every record is necessarily going to be in that database. So our partners with KLS also did their own searches through part and case information systems, which is not accessible to the public and which sometimes turns up different bases that may not be on kbi. And sometimes that transpiring investigations database have cases that weren't in court.
So you have a lot of different sources that don't all agree with each other and not everyone is going to have access to them.
[00:13:49] Speaker A: And Matthew, let me ask you to speak a little bit about this. Patricia just mentioned that not all of the records agree with each other and that sometimes you find something in one and not the other. Can you speak a little bit about that? Did we encounter some of that with respect to this study?
[00:14:05] Speaker C: Yeah, definitely. And even for cases that pertain to the same matter for a specific person, there were sometimes conflicting dates, particularly on sentencing date and date of offense or two that largely could differ there. There weren't necessarily major differences across the multiple instances in which they would appear, but there were discrepancies, which certainly adds to confusion. And then beyond blatant contradictions, there were occasionally omissions for certain types of documents. There were, there was much more specific information about the nature of the offense, about the specific, like duration of the punishment and the nature of the punishment. For other documents, you had much more bare bones level of information. And I think a lot of that. I don't know whether that's information that's been lost due to archiving issues from older documents or whether or not it wasn't recorded in the first place. But in terms of the many documents that I analyzed, there certainly were both discrepancies and omissions of key information for certain matters.
[00:15:02] Speaker A: And this led to some of my favorite, in a kind of cringing and laughing in order to avoid crying type examples of efforts to, to clear criminal records. My favorites. Here are the zombie records. Ryan and Patricia, let's start with you, Ryan, can you maybe speak to this? There were instances in which there, there was apparently insufficient information to clear a record. Even in, when an agency that had charge of those records had received a court order to clear records for this individual, they couldn't do so because they had insufficient information in certain instances to figure out whether the record was in fact within the scope of the court order.
And so we found, for example, some instances of, after a court had issued an order, we found some instances of records of convictions that didn't have. Excuse me, that had. Let me, let me rephrase. We had, we found records of prison stays that didn't have any records of convictions to justify those prison stays. Is that right?
[00:16:08] Speaker E: Yeah. So I mean, to give a bit of context of what exactly we're looking at, for one of our data sources, they separated out court case instances of a record from, from prison instances of a record. So you can imagine in the, in a normal course of having a case brought against you that resulted in a prison sentence, you would nominally have the initial court case record. And then in a separate section on this piece of data, you would have a second prison record stay. And those are two separate things.
But after going through the process of getting some of these individuals to, through, through their expungement process, the court case would disappear, but the prison stay record would remain as this Kind of just floating instance of having been detained despite the fact that they had zero court case records for which they would have been detained for.
[00:17:04] Speaker A: Just to clarify, in case anyone's wondering, it still is the law in the United States that it's illegal to put somebody in prison without a legal basis for doing so. But from reviewing records, you wouldn't know that necessarily because in fact, prison stays with absolutely no explanation for why somebody was, was, was incarcerated in prison. Is that's what I'm calling the zombie records, Is that. Is that. And we found a few instances of those, right?
[00:17:28] Speaker E: Yeah. I would love it if one of the study participants decided to sue for unjustly being imprisoned since the case disappeared.
[00:17:36] Speaker A: The case disappeared, exactly. And Matthew, was there. Were there any other trends that you noticed with respect to trying to go through these records that you thought might have been particularly troublesome, say, for someone, not a lawyer, that were encountering these? I'm not suggesting. I don't have any particular in mind. I'm just curious of whether there were other things that you observed.
[00:17:56] Speaker C: Yeah, a lot of the difficulty in parsing through documents was definitely the poor digital record keeping management.
A lot of them that I reviewed were files from the 90s or the 2000s that have not been archived very well, either poor scanning or oftentimes poor handwriting that can be difficult to parse.
So there were definitely difficulties in terms of the archive management and preservation of documents. In terms of newer documents, I think many of them were organized very unclearly. Through my document review process, I ended up creating a sort of categorization of four different types of documents that I worked with Patricia on, being able to analyze the trends in those different types of documents. And across those four types of documents, there were ones that were definitely much easier to read. Some of them, for example, one of them was just a table of the critical information, which on one hand is easy to read. On the other hand, you could argue that it lacks crucial details. And then meanwhile there were ones that maybe included more details of a given arrest, but might be harder to find, let's just say the sentencing information for. So there's definitely trade offs in how you present the information.
But overall, I think that the two things that really stand out to me are first being the sort of just archive management, particularly of older documents, and the second being just the layout of the documents, either lending itself to being easily parsable but maybe lacking some details, or on the other hand, having an appropriate level of details but maybe being harder to navigate.
[00:19:24] Speaker A: And as a final point on this. And I think, Renee, you may be best suited to answer this question. But if other folks have thoughts, please chime in.
How many different agencies would generate criminal records that we tried to find? And so if the, if a court is to issue an order, for example, to say, clear the records at as to this particular event where allegedly somebody did something wrong and was maybe arrested, tried, convicted and incarcerated for that event, and now after a certain period of waiting time, it's eligible for record clearing. How many different agencies within the government are we talking about? We're probably not going to be able to name them all here, but let's name at least some of the big ones.
[00:20:04] Speaker B: Sure. This is actually indicated in legislation because it does say who the court is supposed to notify.
But essentially the court needs to notify itself.
So it needs to direct itself to suppress the records. And then that includes during the life of our during the course of our enrollment period in this study, Kansas converted to only hard copy records in some of its counties to a online docket system for the entire state. And so we got to see that process, if that sounds exciting to anyone.
So the court has to direct itself, it also has to direct then that statewide system to suppress those records. And then if there are any records in any of the lower courts, it needs to direct those lower courts. So that may be what some folks think of as magistrate courts or Fritz or traffic courts or something like that.
It also needs to then direct any of the agents, law enforcement agencies. That is that sounds simple, but that is not simple. That means any police agencies that might have had a hand in this record.
It could be the jail, it could be the it could be probation or a agency. And then it also needs to send direct the Kansas Bureau of Investigations, which Patricia mentioned as one of her record holders, the kdi, to suppress its records. And then the KBI directs the FBI to suppress its records because everything rolls up to the KBI and then rolls up to the FBI. Again, in theory, that is in contradiction to some of what we just heard Ryan and Matthew talk about
[00:21:54] Speaker A: was compliance. We got to observe some of this in real time in the sense that we saw court orders, we knew about court orders being issued because, of course, it would be unfortunate for a court order directing an expungement to itself be public. That would defeat the purpose. So the court has to remember to suppress the records of the expungement proceeding. In addition to the criminal convictions. We saw some instances of this not working the way it was supposed to put it that way. And in the least one of the cases of the agencies is this, right. The Kansas Bureau of Prisons or the Kansas Department of Corrections. I forget what the name of it is.
They initially took the position that they weren't covered, is that correct?
[00:22:37] Speaker B: That they didn't need to suppress records, you mean?
[00:22:40] Speaker A: Yes.
[00:22:41] Speaker B: Yeah, we, we were fortunate enough working with Kansas Legal Services, had their own relationships so that we could have conversations about oh no, this actually is part of the requirement. And that that quickly I think was bumped. But that wouldn't be the case for that was because we were working with Kansas Legal Services who all who serves the entire state of Kansas and has relationships with these other state based offices. And so that wouldn't be the case for a self represented litigant who might be saying I think my sponged and it's still visible on your website. Right. It'd be hard to even know who to go to for that.
[00:23:19] Speaker A: And we didn't make a systemic effort on this, but we did have some folks that were with the Access to Justice Lab over the summer periodically try to find records online just to see if they appeared visible. And did we, did we ever find any instances of records that we thought should have been cleared given the issuance of a court order that were in fact later visible and publicly available?
[00:23:43] Speaker B: Yeah, we definitely did. And Kansas Legal Services also then we began doing that on a regular basis and Kansas Legal Services took up that mantle as you might imagine, because that would be appropriate for them to do as with their what is commonly termed now aftercare, Expungement. Aftercare. Looking up the case to make sure it actually has been suppressed where it's supposed to be suppressed.
[00:24:07] Speaker A: And in another podcast we're going to ask Marilyn to join us because she was a co author on the paper about the difficulties of getting an expungement, in other words the expungement process, but can until then just pending for now. Renee, can you briefly remind folks who Kansas Legal Services is and what did they generally do?
[00:24:28] Speaker B: Sure. Kansas Legal Services is the legal, the civil legal aid provider in and Kansas. And so some states have regional based civil legal aid providers. Kansas serves the entire state, so they have multiple offices throughout the state, primarily providing civil legal services for low income individuals that meet their eligibility requirements, which is typically income based and then some other eligibility requirements surrounding citizenship status typically and case type. And so the interesting thing about expungements is you are attempting to suppress a criminal record, but it is a civil legal service. And so you mentioned Jim, about the idea of potentially needing to create A new case to suppress a seminal criminal record. And there is differences of how that works in reality among states because it is a civilization case type. But the you are affecting a criminal case. So some states will add this record, they sign to the criminal record and then suppress it all at once. And some states will create a new civil case and then need to suppress both cases at the.
Upon the successful expungement.
[00:25:47] Speaker A: Which one of those did Kansas do?
[00:25:50] Speaker B: They did the latter.
[00:25:52] Speaker A: Yep. And so let's just then get the basic basics of the study down. The idea is that Kansas Legal Services, Renee, had folks calling into its system or walking in, or maybe it had an online enrollment system and it's asking people, and people are asking for legal assistance on to get record clearing remedies, to get help in filing petitions, take it from there. What was the base, what were the basics of the study? Walking forward.
[00:26:22] Speaker B: Sure. So folks would, as you said, they'd reach out to Kansas Legal Services, go through their typical intake process. And in that process, Kansas Legal Services is assessing whether or not this person is eligible for services from Kansas Legal Services on those dimensions that I mentioned. Right. Income based citizenship status and case type, making sure it's a case type that they have lawyers to serve.
If somebody is eligible, that person then. And their case type is expungement.
That person then is funneled to a paralegal that exclusively worked on intaking for the expungement work. And then, and also this evaluation, that paralegal then would have another phone, converse, conversation with a potential client where they would go through the details of the study, invite the person to participate and explain to them like all that meant. Right. So it meant talking to them about what data we would be collecting, how long we'd be collecting it, explaining to them the idea that they would be randomly assigned to either receive full scope legal services or receive the self help materials and then ask for their agreement to participate. That agreement included also. So we, it included not only signing a consent form but also signing some releases to receive some of the data that we were receiving. And so that included releases to receive data from the Department of Labor release, excuse me, a limited power of attorney to allow us to obtain their tax records from the Kansas Department of Revenue.
It included a release to allow us to obtain us to obtain their public utility records as well.
Uh, and then we also explained that part of participation was completing a survey every three weeks for a period of two years. So this was actually a much more frequent survey case than we have talked about in other studies. On the podcast.
And that would include compensation and how to receive that compensation.
So for folks who said yes, they would do that, we, we did all the forms, we did that baseline survey, the paralegal did the baseline survey with them, and then we did the random assessment. And so the random assignment is a little more robust than how I described it. So the full scope Legal Services doesn't get more robust. Right. That is a lawyer who is with you then from the start of your case to the end of your case, also doing that follow up that we just talked about, the Aftercare piece, the self help materials, was a kind of online guide and file type of program that Kansas Legal Services had already had in place before we started the study. So when they couldn't, when they didn't have capacity to serve someone, they would refer them to this, or if somebody couldn't meet with them or whatever, they'd refer them to this. It's an online system that would help you complete your petition and then it would be your responsibility to file it. Kansas Legal Services in this instance, added on the ability to, once you had completed your forms, to have a 30 minute consultation with one of the lawyers to discuss next steps, how to file, where to file, what that process might look like.
So once you completed your forms, they knew they could see it in their system and then someone would reach out to you to schedule this 30 minute consultation.
[00:29:50] Speaker A: Super. And you mentioned some surveys and Patricia Ryan, I forget actually, who was Patricia? Was it you that was administering these surveys every three weeks or so? No, it wasn't. It was Ryan that was sending them out or who was it that was initiating them?
I've lost track. Regardless. Regardless of who it was with respect to the surveys, they went out every three weeks and there were five different surveys that we cycled through, asking each set of questions each time, and then we would repeat them.
But Ryan, I think it was you that calculated response rates to the surveys, Is that correct?
Was this one where we were scraping in terms of credibility for the response rates or one where we had a lot of success?
[00:30:32] Speaker E: No, this was a lot of success. This was astoundingly high response rates in terms, especially for a population that was at risk for a lot of things that would generally lead to not responding to surveys. The average we got over.
So you think of a successful response being a person responding to a particular survey.
We had a success rate on average of 85%, I believe, which is just very high.
[00:31:01] Speaker A: It's very high for a social science survey. And then I guess the other Success, successful thing. Often when you have a more active intervention and a less active intervention. With a survey, people that received the more active intervention respond at higher rates was noticeably higher rates, 10, 15 percentage points higher. Was that true here?
[00:31:22] Speaker E: No, the. The rates were very similar to one another. So we didn't see. Despite the fact that the interventions differed fairly drastically, the response rates from both groups were remarkably similar.
[00:31:34] Speaker A: And so we. The whole idea then is we're going to randomize legal services to obtain an expungement, expecting that there will be different rates of success.
And so by randomizing the legal services, we are effectively randomizing the expungement itself.
And then we can compare outcomes. And Renee, could you just talk us at a very broad level because we're going to get a lot more specific. What were the various classes of outcome variables that we were studying? You mentioned earlier, there was housing, which we measured in various ways, and we'll talk about that in a future episode. You mentioned earlier there was employment and income, economic effects of that time, which we'll talk about in a future episode as well. But there were some other things that we were measuring. What were some of those other things?
[00:32:20] Speaker B: We were also measuring rates of recidivism and then happiness and agency.
[00:32:28] Speaker A: And so the recidivism we could do from criminal records and the happiness and agency we did from one of the surveys. Is that right?
[00:32:37] Speaker B: That is accurate?
[00:32:38] Speaker D: Yeah.
[00:32:39] Speaker A: Okay, super. We're going to be talking about these results in future episodes. I want to talk that. We're going to just close this one with a discussion of whether in fact it was true that legal services improved expungement rates. Different levels of legal services saw different rates of success, and full legal services were much more successful than limited legal services. We're going to talk about that and then we're going to call it a day with this episode. Before we get there, though, just one final thing, and I think this is for Renee, but I'm curious about what everybody thinks. Why wouldn't anyone just know already that this works? In other words, how could it possibly be that suppressing a criminal record from public view. Patricia, you just raised your hand there. Why don't we. Why don't we start from you? I'm curious.
Yeah.
[00:33:26] Speaker D: This is the view of a non lawyer. I have never even been inside a courtroom and was called for jury duty once and I didn't get past the first stage where we were just in a random conference room. So I have a lease about court here.
If you asked me prior to working in the study. How do you think that getting a criminal record expunged work? I would have heard the word, but I would have said I have absolutely no idea. And I probably would have assumed you probably just applied to the court and it just happened. There's a database somewhere that someone is in charge of these things. I'm sure they have a record. They do it. So I think it's a question, why don't people know how this works?
If you've never come into contact with it, how would you have learned it? When would you have ever been told on you. To most people, this is not a part of their daily life.
[00:34:08] Speaker A: And that's a major point here in terms of how hard it is to obtain an expungement.
Talked about. We're going to talk about that again in a future episode. Let me ask you, though, just to focus on a slightly different question. Why is it possible that we would need to. Why would we need to have a study on whether expungement improves people's lives?
Isn't it so obvious that suppressing criminal records from public view would allow people to get better jobs, would allow people to get better apartments, would keep them out of recidivism trouble, or make them happier?
Why is this even worth studying?
And I think the more at least I think about it, the more skeptical that I am that it would have any effect. But most people think the opposite. Most people think that it must be that having the clearing criminal records would be.
Would automatically be useful, would automatically improve people's lives. So actually, let me start with you, Renee, again. What are some of the theories that suggest that this may not be an effective remedy for people, that it may not actually improve their lives, and therefore it is actually worth testing to see whether there is an improvement?
[00:35:17] Speaker B: Yeah. Here we're seeing a long time between the event that becomes that we're seeking to suppress and the time by which you actually seek to suppress that. A long period of time. And so if we're thinking that folks are able to, after suppression, are able to obtain better housing or better employment, or all of a sudden they feel like they need to participate in any criminal activity, you've asked them to wait 10 years before they do that. Right. And so they likely have figured out how to live in between those that period of time. And so to see a large change might be difficult. Right. Because they've already figured out how to obtain housing and employment. Now, I do think that it's likely that we might see improved housing and employment, but I. But it would be but folks have already figured out how to live in in this period of time.
We we also like the legal space, does not look at outcomes that are beyond what happened in the courtroom. So we don't know. Right. We don't have any research to tell us that that our suspicions about whether or not this works, it is true. Are true. Whether or not we think it must work is a good hypothesis. But we know because we as a legal field have been reluctant to care about outcomes beyond the courthouse steps.
[00:36:47] Speaker A: And there's a variation of that theory. It's actually the same thing. But there's a variation of that theory that suggests that people, whatever effects there are of being involved in the criminal justice system, have occurred already and are built into people's lives in a way that it is virtually impossible to reverse by the time they become eligible for records.
There's an excellent paper that's still, to my knowledge, in draft form on SSRN by Amanda Agun, Andrew Guerin, Dimitri Kostas, Alexandra Moss, and Crystal Yang called can you erase the mark of a criminal record? And it calls this effect scarring. That there's basically the people are scarred by their interaction with the criminal justice system and that erasing the records at the end of the process does not remove the scar and can't remove the scar.
In fact, there may be very little to do to remove the scar. There's also again, the suggestion that maybe whatever jobs are available to people after waiting period, they've already got them, or that employers and landlords are already discounting criminal justice involvement that is old. And so the waiting periods, if they're long enough, just result in criminal justice involvement that is so old that the employers, landlords, and other socioeconomic actors are saying, I would have discounted that anyway, even if I'd known about it. There's certainly some things where people are disqualified from belonging to certain professions and from even seeking certain jobs. So they're written in law and there's tens of thousands of those. But the question is whether folks with criminal record, with criminal justice system involvement would actually apply and obtain those jobs anyway, even if you remove the criminal records or allow people to become eligible to that end. Rene, can you just briefly say what it what were the waiting periods here? How long were they for? Kansas?
[00:38:44] Speaker B: Sure. There were three, five and ten year waiting periods after and the the clerk doesn't start on those waiting periods until you've completed all of the requirements of your sentence. So if that included a period of incarceration and probation or parole, you had to wait until you discharged that probation or parole. The 10 year waiting period is not, is for just one or two crimes. Most fall within the five year, but some within the three year.
[00:39:14] Speaker A: But that five year is after the last day of your probation. For example, if you were sentenced to probation, which might have followed your probation might have followed a period of incarceration, which followed a period of court adjudication, et cetera, et cetera. So you might be decade or decades plural away from the events that that caused you to be to have the criminal justice involvement by the time you're even eligible to seek the expungement.
[00:39:39] Speaker B: And often that that clock restarts if you do recidivate in a certain way. Right. If you have another crime specific setup crime.
[00:39:52] Speaker A: Super. So Ryan, we're going to turn to you then for kind of the last word here, which is did the basic process premise of this study turn out to be true? The basic premise of the study that basically everything else depends on was the idea that people would have differential rates of success based on whether they were randomized to the full representation from Kansas legal services versus the self help plus one half hour, the opportunity for one half hour consult. So this is certainly something that legal services providers and access to justice community will be intensely interested in. Does a lawyer make a difference? Right. This is the classic thing that we keep studying over and over again with good reason.
And in this case, despite the fact that there were no hearings at all that people have to, you know, and the adversarial hearings were vanishingly rare in this context. There was no trial they had to prepare for.
It was finding a lot of paperwork and filling it out and then filing it all at the right time.
Were lawyers actually? Did lawyer make a, did a lawyer having a lawyer make a big difference as opposed to having self help materials plus a half hour consulting?
[00:40:59] Speaker E: Sure. So like in a word? Yes, in three words. Yes, a lot. Having a lawyer definitely helps. Just so to contextualize it a bit, for the entire study population, it's roughly equal. Not exactly, but roughly equal were self help and legal representation. 50% of the self help group ended up getting an expungement. Now relevant number for the legal representation group was 90% which is a stark kind of difference in outcomes. And that's over the entire length of the study. You can get, we can, you know, get more specific than that. For the 50% of the self help group who did end up getting an expungement, it took them on average over a thousand days from when they were enrolled to when the record actually got cleared for the 90% of the legal representation group that ended up getting an expungement. And it took them slightly more than a year from enrollment to record clear. So both in terms of overall success as well as in like how long it takes. Yes. Having a lawyer, hands down super helpful
[00:42:07] Speaker A: for getting an expungement for both success and for speed. And the only reason that we know that some of the folks got in the, the self help group got expungements a thousand days later was because we randomized people over a lengthy period of time. Because that's just how Kansas Legal Services worked. Renee, back to you. When did we, when did we open randomization in this study? And so give give people an idea of a timeline. When did we begin randomization and when did it close?
[00:42:37] Speaker B: We began randomization in 2020. In July of 2020.
And we closed in July of 2022.
[00:42:44] Speaker D: Yeah.
[00:42:45] Speaker A: So around three years. And just to clarify for people that will mean that when we do analyses on these, on these economic outcomes, particularly the labor market outcomes, we're going to have to take into account time effects due to pandemic which we were able to do with some modeling. Ryan, going back to you, the figures that you just cited are reflective of the fact that we following people over different periods, different lengths of time. So people that entered the study early, we had more information about them than people who entered the study late.
If we limited everybody to a single to a two year window after, after enrollment. So we're observing everyone for the same period of time and that period of time is two years.
What did the comparisons look like in terms of what fraction in the full representation group versus the self help group got record clearing.
[00:43:36] Speaker E: Yeah. So as you might intuit, if the average amount of time for the legal representation group was about a year, it doesn't restricting our window to only two years of follow up, it doesn't move that number that much. It goes from 90% of the legal representation getting it overall to 83% within two years. Or for self help, that's a very different story.
If you're looking at what proportion of the self help group got an expungement within two years, it is 9%.
[00:44:04] Speaker A: Right.
So we're talking really because the comparison of all information that we have for the self help group versus all information of the lawyer group is a little bit misleading because again, that represents different follow up period lengths for different people. If you entered early in the study, we followed you for longer. If you entered later in the study, then we followed you for a shorter period of time. And so it's possible you would have gotten an expungement later, but for the two year follow up period that we had for absolutely Everybody, it's almost a 75 percentage point difference in success rate. Was that accurate?
[00:44:40] Speaker E: Yeah, yeah. It's you're looking at almost an order of magnitude difference.
[00:44:44] Speaker A: It's almost the point of your flipping a light switch. It's either on or off. Right. Just from the fact of legal services. So I guess from the legal services point of view, that's a big win for it to demonstrate the effectiveness of legal services. And this is a theme that I've talked about before with a Philadelphia divorce study.
For the rest of us in the access to justice community, this is a big moment to go cry in our beers because it suggests that the that the process of obtaining expungements, a legal process, something the law requires people to go to court systems to do, a process that does not require anything close to an adversarial court hearing, in fact does not require a court hearing at all. There is no live hearing here, is still inaccessible. And that the court system has not succeeded in simplifying a process that is inherently by itself relatively simple and involves obtaining information that the court system already itself possesses and feeding it back to the court system.
It has not succeeded in simplifying that system down to the point where people could do it without a lawyer. So again, big win for legal services in terms of obtaining the legal remedy for the rest of the world. Usually a win for legal services means a sad moment for those of us who care about access to justice in terms of measuring the approach accessibility of the system.
We're going to leave it there for this episode with thanks to Matthew Gross, Patricia Gansert, Ryan Halen, and Renee Dancer. I'm your host Jim Greiner and we will be reconvening with this group and perhaps some other folks, or as much of this group that wants to get together, put it that way, with some other folks to discuss the question of whether obtaining an expungement itself improves socioeconomic outcomes. It's particularly employment outcomes that are of intense interest, housing outcomes and happiness outcomes. And we are going to have at least two more podcasts discussing those findings as well as potential reasons for them. So my thanks to everyone for taking this time to record this episode and I will be inviting each of you to reconvene as a group to see if you'd like to participate in the future. So thanks so much everybody. Proof Over Precedent is a production of the Access to Justice Lab at Harvard Law School.
Views expressed in student podcasts are not necessarily those of the A J Lab.
Thanks for listening. If we piqued your interest, please subscribe wherever you get your podcasts. Even better, leave us a rating or share an episode with a friend or on social media.
Here's a sneak preview of what we'll bring you next week.
And so, in part one of our podcast series, which you may have listened to already, we discussed the logistics of the Randomized Control trial, the rct, and the partnership that Kansas Legal Services was kind enough to create with the Access to Justice Lab to implement this study in Kansas.
And today, what we're going to be discussing is the results of the RCT as they relate to employment outcomes. Because, as we'll discuss, that is one of the primary rationales that folks who are seeking expungement cite for seeking expungement. This is one of the things that they want to accomplish is improved employment, improved income.
