Safety in Numbers?

The Atlantic profiled the work of University of Pennsylvania statistician Richard Berk last month.

Berk uses statistical prediction models to conduct risk assessment of criminal offenders.

Berk’s expertise is being sought at nearly every stage of the criminal-justice process. Maryland is running an algorithm like Philadelphia’s that predicts who under supervision will kill—or be killed. The state has asked Berk to develop a similar algorithm for juveniles. He is also mining data from the Occupational Safety and Health Administration to forecast which businesses nationwide are most likely to be breaking OSHA rules. Back in Philadelphia, he is introducing statistics to the district attorney’s office, helping prosecutors decide which charges to pursue and whether to ask for bail. He may also work with the Pennsylvania sentencing commission to help determine whether and how long to incarcerate those convicted of crimes.

Data-based decisions are often less flawed than decisions based on human judgment alone–even (and in some cases, especially) where the decision maker is an experienced professional. We blogged about this awhile back in “Finding the Fat Catchers of Criminal Justice.” There, we argued that data collection and evaluation are essential to properly implement criminal justice reforms and avoid misleading results.

But decisions based upon past behavior alone have the troubling potential to perpetuate those past behaviors. University of Chicago Law Professor Bernard Harcourt explains:

“When you live in a world in which juveniles are much more likely to be stopped—or, if stopped, be arrested, or, if arrested, be adjudicated—if they are black, then all of the indicators associated with prior criminal history are going to be serving effectively as a proxy for race,” said Bernard Harcourt, a law and political-science professor at the University of Chicago, who wrote Against Prediction: Profiling, Policing, and Punishing in an Actuarial Age. By using prior record to predict dangerousness, he insisted, “you just inscribe the racial discrimination you have today into the future.”

This is a risk to be reckoned with. But it is no greater concern than the risk for bias inherent in more discretionary decision-making.

Policy Perspectives: Dr. Shari Diamond on Bond Court

Dr. Shari Diamond

Recently, Chicago Appleseed sat down with Northwestern University Professor of Law and social psychologist Dr. Shari Diamond to discuss her remarkable recent work on courts and juries.

In 2010, Dr. Diamond published a study (pdf) of Cook County Bond Court establishing a causal relationship between videoconference hearings and judges ordering higher bail amounts. She is currently working on a project which gives her unequaled access to observe jury deliberations.

You can read Dr. Diamond’s complete bio here.

Chicago Appleseed: You are part of a growing number of researchers applying empirical practices to the legal field. How does this approach differ from traditional legal research, and why is it important?

Dr. Shari Diamond: Traditional legal research is extremely important. There’s no question that doctrinal analysis gives a sense of what the law is and expects. But many times what you need to know is how the law behaves and how people respond to it. So, the only way to know how the law and legal institutions behave requires a look at actual behavior. We look at the behavior that the law attempts to regulate, the procedures it puts in place, the legal institutions that are supposed to accomplish certain goals–to see whether in fact they do accomplish those goals. It’s almost hard for me with a modern sensibility to imagine any other way of sensibly approaching our assessment of the legal system.

Do you think it’s appropriate for practitioners to take the empirical research to apply directly practice, or is there an intermediate step to be taken? Can you make a direct application of research to policy? 

It depends on whether direct application is easily done. Sometimes a researcher will ask a question that isn’t particularly interesting to the policy matter that is on the table. Or it’s only partially relevant, so there’s some extrapolation or adjustment or modification that needs to be done. This is an academic’s answer: it depends.

Has empirical research made it’s way onto policy makers desks? 

I think it’s happening more frequently. For example, there has been a movement of late to introduce very small changes in the way trials have been  conducted. And that is to give jurors a copy of the written jury instructions. In other words, incremental reforms have been implemented using evidence-based practices, but there are also times when research is ignored.

In 2010, you published an illuminating study of the impact of closed circuit television hearings on Cook County bond amounts. How did you come to work on the study with Locke Bowman and the McArthur Justice Center? 

I knew Locke, and he talked to me about the [bond court litigation] that he was involved in. We sat and talked about it, and it became clear that it should be possible to do an evaluation of whether the closed circuit television arrangement had caused any change in bond decisions. That was essentially the lurking empirical question in Locke’s case. Since the bond court change had taken place in 1999 that meant that, if the records were reasonably good, it would be possible to look at behavior before the change, and a long time afterward. It was an ideal situation for making this kind of “interrupted time series” design to see whether the interruption had caused a change.

And what did you conclude? 

The results were quite dramatic. For those offenses that were affected by the change–that is, those offenses for which a cctv hearing was used after 1999–the amount set for the bond went up substantially. We had a kind of natural control group in offenses for which live in-person hearings continued to be conducted over the same time period. We did not see a change in that group’s bond amounts. So the graphs [see Figure 1 below] kind of popped out to show the change.

What do you think drove that change? The CCTV used in Cook County was the cause, but why did it cause the uptick? 

That’s really the question. There were a number of things about the particular set-up for those hearings that were troubling and potentially influential that are not inherent to all cctv hearings. And there were some that are or may be inherent.

In the course of the project, we went down and watch the hearings. I thought it was important that I had a feel for what it actually looked like in place. The cctv they had probably dated back to 1999: tiny screen, black and white, and a lack of contrast that meant it was particularly difficult to see the faces of the defendants who were dark-skinned. That it was difficult to see them.

The set-up also meant that the defendants’ attorneys were in the courtroom, so their representation was kind of split. If they wanted to say something to their attorney during the hearing, they had no way of doing that. So their representation opportunities were impaired under that arrangement. Overcoming that without an in-person hearing would be difficult to accomplish.

Do you have any thoughts on what might have impacted the judges’s decision-making? 

Honestly, we just don’t know. This is where there’s an imperfect match between the existing research and the study we did. There is some may be evidence that people may behave differently in response to live and in-person versus video presentations, but we certainly don’t know if that affected the judges in this situation. It was certainly easier for the judges to pay less attention to the particular defendant in the set-up that existed in Cook County than if the defendant had been standing in front of them., but whether you could say that that attention would’ve been different from the live hearing.

There’s been a lot of discussion lately in the public sphere in the newspaper–CCBP President has really taken up the mantle of bond court reform. Do you have any thoughts on changes that you think could be simple or incremental to make to the bond court system? 

One of the disadvantages judges have in making these decisions is the limited amount of information they have. The question is whether you can increase that amount of information efficiently, and whether it would affect decision-making. As you know, bond hearings are generally very brief and court typically results in very little time spent per defendant. That usually means that they use very little information is presented. And it may be that judges spend so little time because they simply don’t have much information to inform their decisions.

I have to say that on the day that I watched, it appeared–and we have no records on this so we couldn’t test it directly–that there was substantially more time spent with defendants represented by private attorneys than those represented by public defenders. The vast majority of defendants are represented by public defenders. Whether that impacted the decisions that were made, I have no data on that.

I understand you’re currently working on a book about juries which includes analysis of recorded jury deliberations. That sounds fascinating. Could you give us a preview of that work? 

That project, which has been going on for some time and is reaching fruition (we’ve published a fair number of articles but the book is in process) is on civil juries. I was ecstatic at the opportunity and it’s not something I expect to come again. I refer to it as my “Camelot” project.

There were some judges in Arizona who decided that in the course of evaluating a jury reform that they were implementing that it would be good to know how juries were responding. So they let us do an experiment in which some juries were told that they could discuss the case during the breaks of the trial, and others were told that they could not. And we were allowed to videotape the juries during their breaks in the course of the trial as well as all of their deliberations in order to study their behavior.

Can you share any interesting findings so far? 

The juries work very hard to get the right answer. Their definition of the “right answer” may not completely jibe with every dot and jot of the law, but in large measure it’s quite consistent with it. I think that the way they correct each other and put together information has made it clearer to me than it ever was how valuable it is to have the group as a decision maker. Jurors bring a lot of personal experience and different amounts of attention to specifics in the evidence to bear as they talk about the evidence.

And do they discuss damages awards? Anything interesting come out of that? 

These are very complicated judgements that we ask them to make and the jurors have to work hard, with surprisingly little guidance on how they should go about making them. A couple of overriding themes emerge: 1) Jurors are very skeptical about claims for damages. So they scrutinize them very closely looking for “the greedy plaintiff” or the claimed injury that might not actually be there, or the extra doctor visits that aren’t warranted. Particularly in the low-impact ordinary personal injury cases.

Second, they also struggle in understandable ways to arrive at appropriate future damages.  For example, one juror commented that it would be helpful to have a crystal ball. Future damages in an odd sort of way requires them to have a crystal ball. Because this is the one time when an award may be made, and if the claim is that the injury lasts a lifetime and further medical treatment may be needed, the jurors have to predict what the future will hold. Then they have to make that decision.

What the jury deliberations reveal overall is that jurors pool their different experiences and bring considerable care and effort to a challenging task.

We’re interested in how to bring more humanity and efficiency to the criminal justice processing system. Why do you think it’s important that that stage in the process be more fair and yet efficient? 

Well, I wonder–and this is pure speculation–whether there wouldn’t be a potential triage system that could be established in reaching bond decisions. We know that, for example, somebody who has no prior record and is being charged with a particular kind of offense is typically given a recognizance bond. Perhaps we could do what hospitals do in emergency rooms and triage people out, so as to and leave more attention for the more difficult and questionable kinds of cases. I don’t know enough to be able to set that up, but it does seem to not make sense to treat all the cases the same in this situation. Which is what it looks like we do, except for those represented by private attorneys.

Could you use actuarial or statistical models to make these decisions? 

Absolutely, but I’m arguing for a more mixed model sort of approach. People resist when you get some kind of bad decision made for you based upon actuarial tables, even though they can in fact be more equitable if the criteria you include are legally legitimate. But if you were to say, okay, there are two pots here: one says you will get a complete hearing, and one says, we don’t even have to put you through that process (because you are clearly eligible for release on recognizance). Then, you’re really not penalizing those in the first pot. You’re just saying, we have to look at this more closely.

There’s an interesting challenge with relying upon statistical predictions within the criminal justice system, which is the issue of prosecutorial and judicial discretion. But their caseloads are massive. Do you know where policy has balanced those issues? 

The classic example is the federal sentencing guidelines. They were originally constructed as primarily based upon a summary of what had happened in patterns of sentencing before, to regularize them, and then adjust them from a normative standard of what was thought should be done. So, it was a mixture.

I was a strong advocate of sentencing guidelines before the federal guidelines went into effect. But what happened was a ratcheting up in the setting of those guidelines over time. The public argument always seemed to be in favor of greater severity: oh, yes, we’re going to really punish these things.  So, until fairly recently judges felt completely bound by those guidelines. The guidelines replaced discretion with harshness rather than replacing discretion with regularity and rationality. 

That’s the balance between the values of discretion and the equal treatment of a non-discretionary system. That’s a constant tension. 

Texas Appleseed Report Profiles Small Dollar Lenders

This post originally appeared on the Illinois Asset Building Group (IABG) blog. Chicago Appleseed is an IABG partner.

A report released this week by Texas Appleseed, “Reshaping the Future of Small Dollar Lending,” makes the policy case for small dollar loans, and then profiles six market newcomers offering affordable alternatives to high-cost payday and auto title loans in Texas. They conclude with a number of suggestions for how Texas policy makers and financial institutions can nurture a non-predatory small dollar credit market.

Like many states, Texas has a bundle of legislation aimed to reduce predatory lending. However, payday and title lenders have found ways around the laws, and have in fact exploded their business. In Texas alone, 2,000 new storefront lenders opened in the last six years alone. While mainstream financial institutions have not kept pace, a handful have explored this product, for which there is high consumer demand and tremendous opportunity within unbanked and underbanked communities.

For their report, Texas Appleseed studied four institutions. They also drew information on two other institutions from recent reports on national small-dollar lending programs, the FDIC Small-Dollar Loan Pilot Program and the National Credit Union Foundation’s REAL Solutions program.

For each institution, they identify strengths and challenges to the program, though they do not discuss profitability in detail. Overall, the report issued these recommendations to promote a more robust small dollar lending industry in Texas:

• Texas needs basic standards for affordable credit to support fair competition in the small-dollar lending market.
• Texas banks and credit unions should consider investing in positive lending models to promote the availability of affordable small-dollar loans in Texas.
• Explore affordable small-dollar loan products offered by or in partnership with nonprofits as a strategy to serve nonprofit clients and other consumers who may otherwise fall through the cracks.
• Texas would benefit from enhanced outreach and education about affordable small- dollar lending that targets both financial institutions and consumers.

The IABG recently studied a Chicago credit union’s small dollar loan program. We will be providing insights from that study and other research in the form of a toolkit for Illinois lenders interested in offering these important products.

Links of Interest

What We Read, February 6-8, 2012
 

Immigration

• The National Immigrant Youth Alliance has created UndocuHealth, a resource for undocumented immigrants with mental health issues (via Feministing ).

• The merit’s brief for Petitioner Arizona in the federal immigration case U.S. v. Arizona is now available at SCOTUSBlog.

 

Community Justice:

• The Manhattan Institute has released a report on racial segregation in neighborhoods, looking at trends over the last century. The ACLU critiques the report here.

 

Criminal Justice:

• The ACLU discusses the current state of politics and the federal sentencing guidelines.

• The 7^th Circuit issued a ruling (.pdf) discussing what meaningful consideration of sentencing recommendations is. The decision also concerns not only how judges must meet their obligations for evaluating competing sentencing request, but how they must explain why one recommendation was accepted and the other rejected.

 

Reducing Urban Crime and Violence Forum Discusses Alternatives to Incarceration

The University of Chicago Crime Lab and Urban Education Institute co-hosted a panel discussion on criminal justice and education responses to Chicago violence. The first panel just concluded, with the following participants:

• Moderator: James Warren, Columnist | Chicago News Cooperative
• The Honorable Paul Biebel, Presiding Judge | Circuit Court of Cook County Criminal Division
• Patrick Fitzgerald, United States Attorney | Northern District of Illinois 
• Garry McCarthy, Superintendent | Chicago Police Department
• Frank Zimring, William G. Simon Professor of Law | University of California at Berkley
  Chair of the Criminal Justice Research Program | University of California at Berkley

The discussion centered on evidence-based practices, Chicago’s uniquely entrenched gang structure, as well as noted successes in New York. At Chicago Appleseed, of course, our focus is on court reform.

To that end, Judge Biebel made a powerful statement in favor of alternatives to incarceration:

“The Wall Street Journal has written more than one article about the fact that there’s agreement across the political spectrum that there’s too many people in prison and in jails. Some people think there are too many sentences for nonviolent or drug crimes. Some people think it’s too expensive. Regardless, the incarceration rate in Illinois is the highest it’s ever been. So we need to ask, Can we take some monies and put them into treatment? It’s a lot less expensive to do a drug or mental health program in the community than to lock people up. That’s the challenge today.”

We couldn’t agree more. In fact, several months ago, we blogged about the fact that the NAACP and Americans for Tax Reform share the position articulated by Judge Biebel.

Earlier this week, Public Safety was the subject of the latest Chicago Forward Forum, where panelists including President Preckwinkle, Chicago Ceasfire Interrupter Ameena Matthews, and Police Superintendent McCarthy agreed that the war on drugs has failed on many levels. There, as here, reducing the incarceration rate was a primary topic.

Cook County Bond Court Overlooks Critical Information

This editorial originally appeared at the Huffington Post. 

In less than the time it takes to make an ATM withdrawal, Cook County Bond Court judges make decisions affecting individual liberty and the public safety. This way of transacting justice exacts a dear convenience fee — defendants’ liberty, the community’s safety, public funds — all contrary to long-standing legislation.

It’s not as if judges are provided too little information about defendants.

As mandated by the Illinois Pre-Trial Services Act, Cook County’s pre-trial services department produces a report for the purpose of identifying eligible defendants for release without bond.  County staff interviews each defendant, assessing flight risk and public safety threats. When completed fully, the report includes employment, family, health, criminal history and substance abuse information.

But bond court judges are not utilizing this information in making critical decisions that affect defendants’ liberty as well as public safety.

They couldn’t be. According to many reports, including a 2011 Chicago Appleseed court-watching initiative, bond court judges spend an average of 30 seconds contemplating each defendant’s case. Court watchers reported that judges rarely looked at the paperwork provided to them, relying instead upon brief statements by the prosecutor and defender assigned to each case. Proceedings were often inaudible and defenders and prosecutors sometimes mistook the defendants’ identity.

On the whole, court watchers found bond court personnel to be respectful and considerate. But no amount of courtesy can make up for the system’s inherent shortcomings: too little time and preparation spent on decisions affecting defendants’ liberty and public safety.

Cook County Board President Toni Preckwinkle criticizes the bond court system for sending too many defendants to the county’s overflowing jail. The jail is so crowded that the United States 7th Circuit Court of Appeals last year ordered the sheriff to release 2500 defendants, bypassing bond court altogether.

Preckwinkle points out that even seemingly “low” bail amounts are often unaffordable. Unable to meet the 10% bond requirement of even a $2,000 bail, defendants are jailed for weeks, at an estimated expense of $143 per person, per day.

There is simply no demonstrable benefit to setting low bail rather than release on recognizance or electronic monitoring — a form of non-monetary bail where the defendant wears a GPS bracelet during the pendency of his case. An informative history of bail explains that in the 1960s, a number of municipalities experimented with requiring nominal ($1) or no monetary bail. In one study of 56 municipalities adopting these practices, the failure-to-appear rate was historically low: less than 1 percent.

The crime wave of the 70s and 80s led to a ratcheting-up of bail — and, in turn, pre-trial detention. The failure-to-appear rate is no better, but the costs to taxpayers, society, and defendants are massive.

One of the leading authorities on criminal justice practices and procedures, the American Bar Association, unequivocally prefers release on recognizance and non-monetary bail in its report on Criminal Justice Standards for Pre-Trial Release (pdf).

“[E]ach jurisdiction should adopt procedures designed to promote the release of defendants on their own recognizance or, when necessary, unsecured bond… Release on financial conditions should be used only when no other conditions will ensure appearance.” (emphasis added)

The rationale for this position is straightforward and is grounded in history: defendants in bail hearings are presumed to be innocent. Unattainable monetary bail results in the jailing of unconvicted individuals for an average of several weeks (Cook County estimates range from 24 to 48 days). A great majority of these individuals either subsequently have their cases dismissed or are found to be innocent.

Surely this is not what the Illinois legislature had in mind when, in 1986 upon the passage of the Illinois Pre-Trial Services Act, it declared, “The functions of the [pretrial services] agency would be to interview detained person concerning their background and verify that background, and make recommendations to the court on possible conditions that can be imposed on a defendant prior to trial.”

We’ve asked that Cook County Bond Court judges take each defendant’s pre-trial services assessment into account when setting bail. We also ask the Cook County Judicial Advisory Council to urge this reform in its forthcoming bond court study.The state legislature has required it for over 15 years. Meanwhile, the public pays for the delay.

Chicago Appleseed Fund for Justice

Links of Interest

We hope everyone is enjoying this very mild winter weather! Here is some of what we read this week.

 Immigration

• Sen. Shumer (D-NY) has sent a letter to DHS and ICE, urging that some immigrants be given a temporary “deferred action” legal status which would allow them to work while the deportation of higher priority criminal cases are processed first, under new immigration initiatives.

 Criminal Justice Reform

• The Springfield Journal Register reports on efforts to reduce the prison population in Illinois.

 

Community Justice:

• The Wall Street Journal blogs about impending lay-offs in legal services. Last fall, Congress slashed the Legal Services Corporation budget by nearly 14%, seriously affecting LSC’s ability to fund local legal services. LSC, a 501(c)(3) nonprofit corporation, funds three programs in Illinois: Legal Assistance Foundation of Metropolitan Chicago, Land of Lincoln Legal Assistance Foundation, Inc., and Prairie State Legal Services, Inc., and is the largest funding source for civil legal-aid for low-income Americans.

 

 

Other Links of Interest:

• A graduate student in architecture at IIT has made a jigsaw puzzle inspired by the new ward map in Chicago. The City Council approved the new ward map—41 yes votes to 8 no votes—January 19^th.

 

Cameras in the Courtrooms

On Tuesday, January 24, the Illinois Supreme Court announced (.pdf) approval of a pilot program to allow cameras in court rooms in Illinois. Cook County Circuit Court Chief Judge Timothy Evans is enthusiastic about the idea and plans to petition for inclusion in the program. Judge Evans said he has “no doubt” that the program will be a “big plus” for the courts and citizens because justice is served when the public has a clear view of how the courts work. Chief Justice Kilbride of the Illinois Supreme Court expressed a similar opinion, stating the move will “. . . bring more transparency and more accountability to the Illinois court system”.

Prosecutors in DuPage and Cook Counties, as well as Cook County Public Defenders, have expressed reservations. Cameras will not be allowed in sensitive proceedings, such as custody, adoption, divorce and juvenile hearings, and judges will have the final word in whether or not a hearing or trial may be filmed.

The Chicago Appleseed Fund for Justice supports public oversight of the judiciary with its research on behalf of the Judicial Performance Commission. Appleseed also routinely places court-watchers in courtrooms to assess their functioning. According to a Minnesota CourtWatch report (.pdf), when the public routinely view court proceedings it promotes a higher quality of work among judges, attorneys and other courtroom personnel. When public viewing of proceedings becomes routine, court-watchers report that everyone appears to be treated with greater respect than in courtrooms which operate without public scrutiny.

The Supreme Court’s move toward making the judiciary open, accessible and transparent is good for everyone. Confidence in the courts is necessary for efficacy in the judicial system because the rule of law does not function without the trust of ordinary citizens.

Progress Report: City of Chicago Crowdsources Ethics Reform

At https://webapps.cityofchicago.org/EthicsTaskForce/ you’ll find a growing debate among Chicagoans about how to address ethics problems that have plagued the city for years.

The discussion board style site currently lists three topical forums:

• What ethics related requirement or reform would strengthen your trust in local government? 
• How can city government be more transparent? 
• How can Chicagoans be more involved in holding their government accountable?

Among the debated suggestions? City Council term limits, public official pension reform, replacing many officials’ vehicles with shared vehicles such as those offered by Chicago-based I-GO, and expedite the business and building permit process.

This democratic style of problem-solving, sometimes called “crowdsourcing,” is based upon the idea that a variety of people produce better solutions than small group.

This is a unique opportunity to discuss ethics reform with your fellow Chicagoans. Be part of the solution by contributing to this forum!

Links of Interest

What We Read, January 16-20 2012 

Criminal Justice:

• A Chicago Reporter investigation reveals that a strategy to crack down on gun crimes is disproportionately affecting 15-16 year olds.

 

• Cook County Board President Preckwinkle announced that the Judicial Advisory Council will conduct a 6-month review of Cook County bond court.

 

• Former Mississippi Governor Haley Barbour explains in the Washington Post why he released 26 prisoners just before leaving office this month.

 

• The ACLU released a report detailing strategies for reducing the prison population in a cost-effective way.

 

Immigration Reform:

• The ACLU provides more information about Scott Douglas, plaintiff in the ACLU’s lawsuit challenging Alabama’s immigration law, who appeared on the Colbert Report this week.

 

Community Justice:

• Chicago City Council approved a new district ward map. Check to see whether your ward has changed!

 

• We were sad to learn that Hull House, a Chicago institution serving  communities in need, with an emphasis on assisting immigrant communities, will close. Hull House has been in operation since 1889 and its board is in talks with Metropolitan Family Services and city agencies to ensure its clients continue to receive necessary services and support.