An Examination of Justice Reinvestment and Its Impact on Two States

Marie Garcia: Good morning, and thank you for coming to our panel, An Examination of Justice Reinvestment and Its Impact on Two States.

My name is Marie Garcia, and I'm a social science analyst with NIJ, and I want to welcome you to NIJ's annual conference.

Just last week, policymakers in the state of Oregon proposed a series of budget cuts in an effort to, quote, “rebalance an out-of-whack budget.” Two of the cuts included the closing of three state prisons and the freeing of a thousand inmates back into their community. And though these cuts are unlikely, the situation that Oregon is currently facing has become more and more common. Departments of corrections around the country are being forced to do more with less.

Several states, including Texas, Kansas, Arizona, Michigan and New Hampshire, have taken a new approach in developing new strategies for increasing public safety.

The justice reinvestment aims to promote effective data-driven practices to affect public safety issues. Today, you will hear from representatives of two states who are currently involved in justice reinvestment: John Lazet, chief of staff for the Office of Senator Alan Cropsey in Lansing, Michigan, and Anne Rice, associate attorney general from New Hampshire. Marshall Clement is the project director for the Justice Center, and he has worked with both of these states in developing new policy options for public safety. And our first speaker is Jake Horowitz, project manager for the Public Safety Performance Project.

Jake Horowitz: Morning. I'm talking about an issue that I've been working on now for about five years at the Pew Center on the States. Back in 2006, the Pew Charitable Trust launched the Public Safety Performance Project, which is geared at helping states to get a better public safety return on their taxpayer investments, looking for fiscally sound, data-driven sentencing and corrections policies that improve public safety, hold offenders accountable and contain corrections spending.

Along with the Bureau of Justice Assistance, which is a sister agency to NIJ at the Office of Justice Programs, we support both some of the work of The Council of State Governments as well as our own research, research briefs, policymaker convenings across multiple states, as well as direct technical assistance in several states. To the list that was previously mentioned, you can now add South Carolina, and we've got a few more states in the wings.

So just two things I want to address sort of to kick this off: The first is looking at the national trends that we think explain increasing state interest in justice reinvestment processes, and the second is some evidence that states can have less crime and more accountability with fewer taxpayer dollars.

And then, after I go, I'll kick it over to John and Anne, who will tell you about what they actually are working on and accomplishing in the states, which is the meat of this work as opposed to just the sort of a quick national overview.

So, back in 2008, we released a report called One in 100, detailing that 1 percent of all adults in this country are now behind bars. It's unprecedented statistics in terms of national history. The United States is the world leader in this category, and while One in 100 is an important touchstone and, I guess, a benchmark in some sense, it also hides extreme concentration by gender; youth, or age; race; educational levels; and geography. So one in 100 is a powerful statistic, but it becomes much more powerful when you dive into some of those demographic subgroups. We'll be releasing our report hopefully in July which does this, and I encourage you to take a look.

Also, on another note in this, we're releasing a national justice atlas, which will be an interactive online tool where you can look at the concentration of correctional populations down to the census track level for about 22 states.

So there's one in 100 adults behind bars, but one in 31 adults are under correctional control in the country, so one in 100 behind bars, one in 45 under probation or parole supervision. You add those together, you get one in 31. This is an important thing because people tend to focus on who's behind bars, but two out of three correctionally supervised individuals are in the community and not behind bars.

And what this means for states, among other things, is a big drain on taxpayer dollars. So states spent about $10 billion on corrections 25 years ago. They now spend about $50 billion a year. Even in inflation-adjusted terms, this is a massive increase. And one out of every 15 state general fund dollars now goes to fund correctional spending.

Obviously, prisons dominate the spending. They're not the entirety of it, but prisons, due to their capital and labor-intensive nature, tend to really dominate states' correctional spending.

So, again, you know, if two out of three correctionally supervised offenders are in the community, what we're actually finding is that nine out of 10 correctional dollars go to prisons, and there's obvious reasons for this, right? I mean, prisons do cost more than correctional supervision, but the question I think we're asking with some but not all justice reinvestment projects is “Could we reallocate some of those dollars and improve community supervision practices in a way that gives us more public safety with fewer dollars?”

And, finally, you know, sometimes we get a little immune to this hanging out in D.C. where we don't always have to balance budgets, but states do, and so, when you spend a dollar in one place, you don't have it to spend somewhere else. So, again, just looking at general fund spending in two primary categories, you see corrections consuming an ever-larger share of state spending, which is leaving fewer dollars for other priorities of policymakers in those states. So, again, the justice reinvestment idea is can we reallocate those dollars.

Now, don't read too much in this slide. We're not saying “Go fund your state's university system,” but there might be other areas where you can shift your money to get better returns.

There's another reason that we see state chambers of commerce and local chambers becoming more and more interested in this work. You know, business taxes are on their mind as they struggle in the recession, and they're saying, you know, “Are there areas where we can get better return on the investment?” These are sort of the words and the concepts that resonate well with business leaders.

So these last couple of slides I want to throw up here get at this idea of can you have … is there a win-win here or actually a win-win-win where you get more public safety, more accountability at less cost.

So the idea here is a lot of states experienced crime declines in the last … well, since the early '90s, but these states did vastly different things with their sentencing and corrections system. So you see three states from very different parts of the country experiencing the exact same amount of crime decline over the — this is the period '97 to '07, but I think this would probably hold today as well. But, you know, Nevada decreased its prison population, while Louisiana and Minnesota substantially expanded theirs. There's a lot of reasons for this, and we could dive into this if it's an issue people want to follow up with, but I think at base, this is not a hard argument to make.

Unless you think that every correctional dollar is currently allocated to the most optimum program or policy, then there's always ways that we can reallocate dollars and get more public safety and more accountability.

This next slide, I want to prep a little because it's the messiest slide we ever used. It's sort of the faux pas of PowerPoint presentations, but imagine you took these three states and gave them each a coordinate. So you have a crime decline coordinate and a prison population or incarceration rate coordinate, and you'd map them on a little Cartesian grid. So the lower you are on this grid, that means you averted more crime, and the farther you are to the right on this next grid, that's the more you increased your prison population. So, with a little trepidation, this is what you get when you map all 50 states over 10 years.

First, there's massive variation, and there's a big blob. So the big blob, I think, indicates that there's not a whole lot, there's not a very robust relationship between these two things, your prison population and your crime rate.

The other thing is I'd call your attention to the states on the left-hand of that blue line. So those are states that got a crime decline that looks a lot like the national average or, in cases like New York and New Jersey, much better than the national average, but they didn't increase their prison population. So they either averted some expected prison construction or they're actually able to reduce their prison population, but I don't think we can go so far as to say, you know, they actually reduced their correctional spending. There's a whole 'nother discussion on that end.

The two things I want to throw out here is that, really, this is about reallocating dollars to get a better return on investment. I'm reading this book right now called Money Ball. I don't know if anyone's read this. It's about the Oakland Athletics baseball team, and they have … they had the lowest payroll in baseball for a long time and consistently outperformed teams that had a much larger payroll. So the question is how were they allocating their dollars across scouts, you know, prospective players and finding the ones that other teams are missing. In a weird way, it's a perfect analogy for this. It's data driven. It's taking a look at your bottom line, how much money you have and then achieving your ends, and I think there's something very similar going on in some of the processes we're talking about here today.

The other thing, just because he's here, Judge Alm from Hawaii is going to be presenting tomorrow. It's tomorrow, right? I encourage you to take a look at his project, Hawaii HOPE, which basically holds probation violators accountable with swift and certain but mild sanctions. They did a randomized controlled trial supported by the National Institute of Justice that found that fewer folks are using drugs, being arrested, being revoked to prison and occupy fewer bed days. So what you have there is less drug use, less crime, lower taxpayer expense and lower need for punishment, which again gets at this issue, can you have the win-win-win. I think you can, both in a program like HOPE but also in a policy process like justice reinvestment.

So, just to close up here, back when states' budgets were flush, it was, I think, an okay thing to say “How can we be tougher on crime? How can we prove that we take this seriously and we're going to really nail these criminal offenders?” Not just because of the recession; I think there's other reasons for this, but the policy question is shifting. How do we get a better taxpayer return on our investment in public safety?

Anne Rice: I'm from New Hampshire, and New Hampshire, beginning last September, became involved with The Council for State Governments and the Pew Center to look at justice reinvestment in New Hampshire, and I want to talk about what has happened since September, but before that I want to give you a little background on New Hampshire and where we stood when we started our initial efforts of justice reinvestment.

New Hampshire is a very rural state. It's about 1.4 million. Our largest city is 120,000. So we're kind of like a drop in the bucket of the states here.

It is considered the safest state in the nation, and it is an extremely frugal state. There is no income tax. There is no sales tax. Any political candidate who doesn't take the no-new-tax pledge stands a good chance of losing when they run for office.

So, as a result of that, we have seen a great deal of decline in community services and mental health programs that are available in our communities. We have also seen a significant decline in the number of programs that are available to inmates at our New Hampshire state prisons. Simply the lack of money, no revenue coming in, so all of these programs are declining.

And we're a bifurcated system. We have a state prison system, and we also have 10 independent county jails that don't have any working relationship with the prison. So, if you are in a jail population, it depends on what county or what part of the state you are as to whether you get any kind of services whatsoever, either in the jail facility or out in the community.

So crime rate in New Hampshire has been stable for the past 10 years, but over the past 10 years, our prison population has increased 31 percent. So there's 31 percent with no new crimes, and the cost over that same period has doubled. And because of all of those things, we had very significant concerns on all branches of government in terms of what are we going to do about correction costs and why are we seeing this increase in the corrections population.

There were a number of groups looking at it. In fact, when The Council of State Governments came in to look at justice reinvestment in New Hampshire, one of the concerns was that there were actually too many people looking at this. There were too many groups that were trying to find solutions, and there was a concern that people were starting to work at cross purposes, so we needed to get a statewide effort focused on corrections spending and corrections policy.

So one of the things: New Hampshire is a truth-in-sentencing state, which means that a person who is sentenced to prison is required to serve their full minimum sentence. They do not earn good time off that minimum sentence, but they are required to serve the full minimum, and then anything beyond that is determined by our parole board.

So one of the things that our legislature was looking at in terms of how to deal with corrections spending was to eliminate truth in sentencing so that the prison population would go down. People wouldn't be spending as much time in prison. But there was a significant outcry with that, saying, “Why are you going to cut people and ship them out to the community when there is absolutely no support in the community for these folks?” New Hampshire spends zero dollars on community support for probationers and parolees. There is nothing in the community other than the probation and parole officers themselves who provide supervision.

So it was decided the truth in sentencing was not something that we could eliminate at this point. It didn't make policy sense because, if we send these people out and there's no support for them, the likelihood is that we're just going to see people recidivating.

So, what really — we were in a chicken-and-the-egg situation. We knew we had to cut corrections cost, but, at the same time, we were fearful of just eliminating or cutting the prison populations and sending these people into the community where there was no support for them, where we knew that people were coming out with significant mental health and substance abuse treatment problems, but there was nothing in the community to support them.

And that was where we were in the beginning of last year, was everyone in the government was recognizing that we had to do something, but there was absolutely no agreement on how to get it done, how to reduce spending and keep our safe state. In March of 2009, a representative from The Council of State Governments came to New Hampshire to talk about the process of justice reinvestment, looking at how to shift your costs and allocate more effectively.

And, at that meeting, it happened that we had the president of the Senate, the speaker of our House, I believe the chief justice of the Supreme Court was there, and legal counsel for the governor, and all of them listened to this presentation and thought this is something we really need to do in New Hampshire. And, within four months' time, we had a bipartisan inter-branch group that got together and worked very hard to have The Council for State Governments and the Pew Center provide a grant to New Hampshire so that we could have a justice reinvestment study in New Hampshire. And that happened, and we had people on the ground in September. So this was a very quick decision on our part, and Pew was wonderful in agreeing to support our efforts.

And it was agreed that our goals would be to reduce corrections spending while increasing public safety but that there be no change to truth in sentencing, and, politically, it was decided that it made more sense to, instead of looking at the front end and trying to divert people away from prison, which obviously is a very good goal in and of itself, but, politically, it was decided that it was going to be more viable if we looked and focused on the back end of corrections system, looking at who is in the corrections system and, when they leave the prisons, how are they being dealt with, how can we keep people out of prison once they've been released.

So what the justice reinvestment process is is that there is a group of people that come in and do very extensive data analysis in the state. They then put that analysis together and come up with policy options for the states to follow and then assist in the implementation of any policy options.

We had a team of people come in, start in November — excuse me, September — did a very intense data mining of our corrections data and also did a series of stakeholder focus groups in New Hampshire. We pulled together groups from a variety of different places, including law enforcement; prosecutors; county attorneys; the probation/parole officers, which I'm going to call PPOs; victim advocates; crime victims; mental health and behavioral health associates and professionals; county jail superintendents. We had people from the courts, and there were focus groups with each one of these groups of stakeholders to identify what they saw as the issues relating to corrections population coming back into the community and also what was driving the cost and what was driving the prison population increases.

And there are a number of findings that were … there were a number of findings that were very significant for our policy direction, and I'll just go over a couple of those with you. It was recognized that admissions for new crimes were stable. So our crime rate had been stable and so had admissions for new crimes. We weren't seeing any growth there.

What we were seeing was a growth in the number of people that were recidivating, people that were coming back on probation and parole violations. It turned out that we had 57 percent of our new admissions were probation and parole violations. So that's a huge number, and that was the big driver for our prison population growth.

Also, we saw that the number of people that were being released and still on probation and parole had almost doubled over the time period of eight years, and when people were leaving prison, there was no way, no established assessment process to determine who was at high risk for recidivating. So people were leaving the prison system and going on to probation and parole, and there was no assessment being done of who was someone who needed a great deal of community supervision and who really didn't need a lot of community supervision. As a result, our PPOs were spending … were really focusing on all of their probation and parolees with the same amount of effort and really not allocating their effort to the people who needed supervision the most.

We also looked at parole setbacks. When people were violated on parole and they were sent back to prison, most of the revocations occurred within eight months. So that seemed to be the critical time period for someone who was not going to succeed in the community was eight months, and once they got back to the prison, they were in prison for an average of 10 to 11 months, and they sat in what we call the “Reception and Diagnostic Unit.” They were not in a unit that provided any programming. So people were pulled back from the community into prison, and no programming was done to help them move back out and be successful in the community.

And it was also discovered that about 75 percent of the people that were coming back recidivating were there because they were either … it was a result of mental health problems or drug, substance abuse problems. Those were the causes of recidivation. So they were coming back into the prison because of substance abuse and mental health problems, and there was nothing for them once they got back in.

And, finally, another thing that became very apparent during the data gathering was that there was nothing in the community to provide intermediate sanctions for people on parole and probation. So if someone violated their conditions of probation or parole, the PPOs had two options. They either kind of ignore it and say “bad boy” or they violate the person and send them back to prison. There was nothing available for these PPOs to encourage people, incentivize good behavior in the community.

So those were some of the things that the findings of justice reinvestment people came up with and then came up with policy options. And one thing I want to talk about here is when justice reinvestment folks, the team, came into New Hampshire, it was required that we have a leadership team in New Hampshire of policymakers who were going to work with the data analysts as they mined their data and talked about these policy options. It was an ongoing dialogue.

So we had a leadership team that consisted of all the major policy decision makers in the state. That included the president of the Senate, the speaker of the House, legal counsel for the governor, the attorney general, the chief justice of our court system, and they sat down with the data analysts on a regular basis to talk about what kinds of information was being discovered. They could talk to the analysts and say, “Well, this doesn't appear to be correct. We think that you need to focus on a certain area.” So there was an ongoing dialogue to make sure that what the data was showing was actually what we thought was happening in New Hampshire.

And one of the good things about that ongoing dialogue was that the policy options that were developed really came as part of that dialogue. The data analysts and the folks who came in were recommending positions or policies that we should take. This didn't come at the end of a long period of research where there was no communication between the analysts and the policymakers, but there was this ongoing dialogue. So everyone knew the directions that we were going.

And some of the policy options that were recommended and have, in fact, been enacted into legislation were: one, that we focus our community supervision on those who are at high risk of recidivating, and so it requires the department of corrections to establish an assessment program, so that any person who is going on to probation or parole is going to be assessed using an objective assessment tool to assess them for their risk of recidivating. And those on high risk of recidivating are going to require the most and intensive supervision. Those who are low risk and medium risk are going to be covered under a much less … a much shorter period of probation or parole active supervision and then put on what we'll call a “monitoring status” so the PPOs can focus their efforts on the population that needs it the most.

Another option that has been implemented, was recommended and implemented, was the development of swift and certain sanctions in the community for violations. So it requires, our legislation now requires, the department of corrections to establish programs in the community, a seven-day residential treatment facility that will allow people who are violating their parole conditions to go into a facility in the community for seven days so that they don't go back and spend 10 or 11 months in the prison. Every effort is made to try and keep the people in the community where they can maintain jobs … and housing.

Another policy option is that nonviolent offenders will now serve no more than 120 percent of their minimum. We found that there were people who were sitting in prison; they'd served their minimum sentence, and they were not released because either they had not participated in the programming that they were supposed to; their disciplinary conduct may have kept them back. But it was a substantial cost. People were there for an average of 500 days more than their minimum sentence. So the policy option was to make sure that people get out, nonviolent offenders get out at 120 percent of their minimum.

And, finally, another policy option which was all agreed on was the reinvestment of money. As we decreased the correction population by these efforts, by making sure that people have more strict supervision if they need it in the community and getting people out more quickly, the money that we are going to save from that is going to be reinvested in community services.

Now, it was decided that, politically, that policy option could not be implemented in legislation. So we have a bill that went through the state and went through the House and has passed that sets out all of these things that the department of corrections needs to do and probation and parole, but the funding piece, the critical reinvestment piece is something that is not in our legislation, and the reason for that was because nothing was passing our legislature this year if it had any kind of additional funding.

So, when we went and testified on this piece of legislation, we had to make very clear that this process will only succeed and reach the goal of reducing spending while maintaining public safety if the legislature was committed to reinvesting that money in the future. And the current legislature assured us that they would go along with that and that that money would be reinvested.

I don't know. We have elections every two years in New Hampshire. Who knows what elections are going to bring this fall and whether we're going to have a similar group of legislators who are similarly committed to this reinvestment process?

But now New Hampshire has these policy options in place, and we are just on the stages of implementing it. So I would like to say that I could come back in two years and tell you that we have drastically reduced our prison population and that we're not seeing the kinds of recidivism that we're now seeing, but we are on the cusp of all of that right now.

So John will speak next, and he … Michigan is much further along in the process, and I think that he can tell you how these policy options have worked out. Thank you.

[Applause.]

John Lazet: Don't you just love headlines like this: “Criminal Arrested for the 99th Time”? And this is the kind of thing that drives everybody crazy. If I may encourage you all, if you're a researcher, please remember that every policymaker, every individual, every institution operates in a context, and because in a former life I was a chemist — I worked in a chemical factory; I was an instructor — I understand the specificity and sometimes narrowness of what we do as researchers. However, when you get to policymakers, they operate in a much broader context. If you're a policymaker, please understand in the 21st century, what we're learning in Michigan is we need to govern by collaborative. You have to govern within networks. You can't just impose from on high. And if you're a practitioner, we're learning that you have to stay focused because things will come along, they will upset the apple cart, and the best laid plans of mice and men are just that.

So what we had in Michigan between 1992 and 2002, in that 10 year frame, we completely restructured felony sentencing in Michigan, everything completely redone. We completely overhauled our approach to controlled substances. We implemented truth in sentencing. We had a significant restructuring of the courts, how they did everything. We had drunk driving reform. We had a governor who eliminated a lot of our public defense work. We had our Detroit crime lab — one-fourth of all the evidence in the state — they had compromised procedures. The entire lab closed and flooded the entire system. School programs are being cut. The safety net was being shredded in places, and crime victim rights were out there with a fairly strong voice in Michigan.

So, in the middle of all this, here's what we're being told by Michigan opinion leaders, the complaint: “Well, guess what? We put too many people in prison, and we need to have more people diverted,” and they compared us to surrounding states. What a surprise. “Too few prisoners are being paroled. They serve longer than they need for public safety.” So the solution, we were told: “Divert more people. Parole more people from prison. We'll save tens of millions of bucks. If we close prison, we'll save tens of millions of dollars. If we reduce the number of employees, we'll save tens of millions of dollars.” That's what we were told.

Here's what drove it. Take a guess. Right about there, we had a sex offender with three prior sex offense convictions and three prior prison sentences who walked, and he committed a heinous crime in a large area, metro Detroit area, and as a result, the public went berserk, policymakers went berserk, prison populations spiked up, maxed out our capacity. There was a move to try and mitigate that. It didn't work. It bounced back up. And during that three-year little hiatus right here, there was an agreement we're going to build more prisons. Those prisons started coming online right here, and they jumped up over 50,000, over 51,000 in 2002.

We have term limits in Michigan, and in 2002, everybody got broomed. All the legislators got broomed. The administration, they all got broomed. Everybody is brand new. So, after these 10 years of all these reforms, all the stakeholders have kind of got their hands on their thighs, pulling their shorts. We're all kind of huffing and puffing, and all these new people come in and say, “OK. Time to do reform.” So we had all these competing pressures out there, and we all understood, people going to prison, we spend too much. This is what really drove people bonkers.

The prison budget had gone up from about three-quarter billion to over two billion during a time frame. Now I'm going ahead into the future to take you where we are.

Curiously, while the prison population was going up, the percentage of felons who were adjudicated and being sent to prison was dropping. It was at a high of 40 percent in late '70s. By 1990, one in three felons went to prison. Today in Michigan, about 19 percent go to prison. What this led to is local stakeholders, key groups saying, “This isn't working.” They would point to our violent crime in Michigan. Violent crime would go up and down maybe 10 percent over the last decade, but the number of arrests has dropped by 40 percent. The closer rate for murders in Detroit was down around 30 percent, and we had the community screaming bloody murder.

Then they looked at the number of felons. Curiously, we were actually adjudicating more felons over that same decade. The number of felons adjudicated went up by over 40 percent, but the number of felons going to prison remained flat, and that's why the percentage dropped down. That was a very conscious effort to divert felons, and the question is did we accomplish public safety, did we take offenders to keep them from reoffending.

We had a very robust program called the Michigan Prisoner ReEntry Initiative. If you have any interest in reentry issues and looking at something that's relatively mature, go to michigan.gov. Go to Michigan Department of Corrections. Pull up MPRI. The program they have down is a fantastic problem. It is a seamless plan of services and supervision. They're theoretically working with individual offenders. They work with the locals, theoretically, and they start in prison and they go on. Their goals: promote public safety, increase success of offenders, do risk management, accountability with offenders and system officials, and get victim participation.

Now, as we went through this and we're implementing MPRI, what we found is that one of the strengths is we standardized risk assessment across the state. If you don't know what works, how do you know what to fund? And if people say something works here and people there say, “We don't believe it,” if you have standardized risk assessment, it works.

We're working with a supreme court in Michigan to get this in all the courts. It makes a big difference. We predicated what we were doing on evidence-based practices. If you don't have employment, you have a problem; if you don't have housing. We worked on mental health issues, substance abuse issues, education issues, transportation. The MPRI, what we're doing in Michigan is putting our funds specifically into these areas because the research says this makes a difference. When we talk to local stakeholders, they say it makes a difference.

The weakness? This thing was budget-driven. All the people who are running this cared about, it seemed like, was are we spending less money all in all on the budget, and then they had implementation issues. What works in Detroit doesn't work in Kalamazoo, doesn't work in Traverse City.

It kind of looks like this. This is my drive home. I took this Saturday, and you can't see it clearly, but in a quarter-mile stretch, the speed limit goes from 70 to 60, back up to 70, down to 45, and back to 70 over a quarter mile. Now, whoever is doing the construction for the department here, it makes sense to them. The problem is in real life, guess what we all do when we drive through that zone? We're all doing 70.

We found that if I'm sick, I go to a doctor, I get a prescription, I go to the pharmacy, I get my meds, I go home. If you're in the state correctional system and you need a pharmaceutical, it's a 54 step process, and every step got added in by somebody doing it. It all made sense, “You know what? We can do better than this.” So the stakeholders came up with a blueprint for a safer Michigan.

If you go to the Prosecuting Attorneys Association of Michigan website, the PAAM website, you pull up their blueprint for safer Michigan. They have a far more detailed presentation on this. What they said is, “You know, we need to get down to business and start influencing policymakers.” So we had the administration aggressively pushing for reforms. We had the stakeholders saying, “These don't work. Let's do something different.” So what this blueprint group said is we need to have adequate police in Detroit, and I can tell you from personal experience, if something happens to you, you have to go down and file a police report. They will not come out and see you. Your home gets broken into, they will not come out and see you, generally.

Now, if there's an armed robbery and somebody's entered, you got somebody bleeding out, you know what, they'll show up, but it's hard to get a response. They want adequate prosecution resource.

They also want adequate representation for criminal defendants, and they want adequate courtroom resources. A judge is not going to have programming as part of the conditions of a sentence if it's not available.

For corrections, we want adequate local resources, we need jail space, probation and parole, which is run by the state, but the state seems to default to oversight as opposed to really offering resources, which are by the locals anyway. And then we need to make sure we don't do too much with our reducing prison population.

So, through 2002, this is a prison population. Between a robust reentry program, between stakeholders all sitting down and agreeing to work with us, we were able to do this. We basically stabilized a population. People are still saying, “You spend too much,” and here's what drove everybody crazy.

Slope compares to slope. Starting at the top left-hand corner, this is the budget compared to the number of employees. The fewer employees, the more the budget went up.

The next one to the right is the number of prisoners versus a population. Now, that last little drop in the prisoner population is a projection. We don't think they're going to be able to actually meet that, but even so, the budget is going up.

Right below that, we have closed … last year we closed 11 prison facilities. Major. We released a lot of parolees, and the costs still go up. And what we found is at the bottom left-hand corner. The red is savings we have gotten from closing prison facilities, and the blue is the economic increases, even though we have fewer employees. And so reentry was what was driving the closing of prisons, but it really was masking other drivers out there. So, if you get into reentry, please understand you have a lot of agendas going on out there.

We learned a lot like New Hampshire. Thank you, BJS. According to them, the national average for supervised offenders going back to prison is 33 to 36 percent. In Michigan, we're around the 60 percent level. Our reentry, while robust, was not particularly effective. It's driving our prison population.

So here's the policy debate in Michigan. We stopped the growth in prison population, too many criminals back on the street. Crime is going down. Crime isn't increasing. We have four of the most violent cities in America, and Detroit is one of the 10 most dangerous in the world. So what do you do? You call in the experts at Justice Center and Council of State Governments, and so we did.

They come in, did an analysis. We loved this. For the first time, they got past all the individual biases, institutional memory. They went across all the departmental lines, all the databases. They look at everything. They gave us analysis. Everybody is cheering and saying, “Oh, yeah. Look. See? See, I told you. Look at that. Look at that.” “Yeah, we'll look at this. We told you.” So everybody was right, but nobody had a global picture. CSG gave that to us.

Then they gave us policy options. When this was presented — and Marshall probably well remembers this. When they came up with the final savings that we could get aggregated, those people who are budget driven said, “It's not enough money,” and they wanted to walk away from the table. And the locals leave the room, and they're all cheering and high-fiving because, guess what, they're recognizing that we can do things in the community. It doesn't have to just be parolees. It can be about probationers, and it can be about crime prevention, also.

So here's what CSG did, pulled together over 80 stakeholder groups. I'm sitting in this room with all these people, and I know almost all of them, never had them all together before, and everybody is kind of like, “Look, it's them.” “Oh, look, it's them.” And the level of interest, caution, mistrust, it was an interesting blend in that room.

CSG analyzed data across all department databases. They gave us an overview of what was happening all across the state. They gave us a three-tier, prevent crime, better interventions earlier on before they arise with the threshold of going to prison, and they gave us a … then for once you're in prison, what can we do to help reduce reoffending and, therefore, cost, theoretically.

So they identified the pressures we had, all those people going through all the reforms. They looked at the programs we had, the MPRI that was in place, other issues. We have therapeutic courts throughout there; we were acting independently. Everybody agreed to be factually driven. I'm trying to give you graphs because we don't go by opinion. We go by the facts.

We agreed to a comprehensive approach. Everybody is part of this, all the key stakeholders, and we agreed to implement a balance of deliverables. Everything gets something. Nobody gets everything, but everybody gets something.

Let me back up. Here's part of our problem. We had the Detroit Free Press, and they are always on our case about we spend too much on corrections and we can do things better. And last Wednesday, we get this wonderful edition, and the Senate majority leader came over and dropped this on my desk. And here's a guy, he was a serial killer, he got out, he killed again. So here's a paper saying “Stop sending people to prison,” and what do we get top of the fold?

And, oh, guess what, here's Kwame Kilpatrick, mayor of Detroit. He didn't pay his restitution. He was hiding assets. So he got sentenced to prison. You shouldn't go to prison for this, but he did. Now they want to send him to boot camp, and the whole city of Detroit is all upset that Kwame Kilpatrick is not going to spend enough time in prison. So how do you determine if your reentry is successful? Because he was out on probation. He was in Texas, wasn't paying his restitution.

We put down an agreement for what success is. We put this into the budget law, first of all, so the budget-driven people had this, and then we're working now on putting it into the policy, so it goes beyond, because at the end of 2010, term limits, we're all getting broomed again, entire legislature gone. Executive branch, everybody is gone. We start over again. All the stakeholders are kind of groaning, “What do we do?”

So we said this: “Report to your field agent.” Absconders were not considered failure, and all the department used to report was failure, so they must be successes. We don't believe that. We looked at evidence-based practices, substance abuse issues, mental health, sex offender. You can see these. We looked at, just like the unemployment system, are you working, are you getting training to get a job, are you genuinely and actively investigating employment opportunities, do you have housing. Sex offenders, we have exclusionary zoned residence for sex offenders, and we do the circle graphs on your city. What you find is you have these weird little areas where sex offenders can live, and you go back to Detroit, you go back to Kalamazoo or Pontiac or Flint, and you look at where they can live. We have a problem. Then we have transportation problems. You may have gang issues that play into it.

So we're finding that we've had to put more money into housing than we anticipated doing, but if you don't have a place to live, I'm telling you, it's just trouble waiting for you.

Getting a state identification card, it is so frustrating that … what's one thing a guy has behind bars? He's got time on his hands. Start getting the documents you need, so that when you walk out the door, you go down to Secretary of State who has homeland security issues. You can say, “Here's my A documents, my B documents, my C documents. I can get a state ID card.” Not going back to prison, not going back to jail, you have to find success.

What have we done? We have a first responder program. We sat down with another collaborative, mental health professionals, providers, law enforcement, and the community of persons with mental illness. We put together a 200 page manual. If you want to see what we've put together, you can e-mail me. I don't know where this is on the Web, but I will get you a website reference to it.

We have trained across the state on this. We do it regularly. I have heard outstanding comments from people on the field that you go to a crime scene, and somebody presents with a mental illness, it doesn't matter why, a significant difference. We have found that the jail commitment, just being arrested, has dropped dramatically in the counties where we're doing this aggressively.

We're expanding therapeutic justice: drug courts, sobriety courts, mental health courts. Funding is tough. If you tell people it makes a difference and you've got local buy-in, this will work.

We've enhanced resources for both pretrial and adjudicated offenders. This gets the locals happy. You can also … by doing pretrial, you can start before you actually get to it. If you know the guy has a substance abuse problem, get on it right now.

State department of corrections. This is what we do for MPRI.

My clock is running out here.

So here's what we had for a prisoner population. It had grown. We had stabilized it. Council of State Governments came into town. They spent a year. They analyzed everything, sat everybody down, went back and forth. They came out with our policy options, January of '09, and through a series of a really kind of rough and ragged … working with everybody else, this is now where we are in our prisoner population. The question is can we sustain this. By closing prisons, we have not much option, but reentry has to work. Reoffending has to stop. Victimization has to go down. If you do this, you'll be successful. If you don't do this, we're back where we were 10 years ago.

Reinvestment. Here's what we were able to do. You'll notice that the prisoner reintegration, right here, we actually spent $24 million in here, and not a single person walked out the door, and we raised a stink, like, “Look it, I know you're setting up an infrastructure. I know you're getting contracts in place, but that's a lot of money in a state that has a budget revenue situation that has no bottom. It just goes down and down and down.” But because we had the reentry and because Council of State Governments has come in, we are now spending just under $58 million, current year, and we're looking at about the same for next year. And this is in the middle of schools being cut, higher education being cut, social programs being cut. We're doing reentry. We're able to get the funding because everyone has worked together. We've all agreed to be collaborative, and it's working.

So what have we learned? We've learned that individuals have a lot of things going on in their lives. They interact with institutions. It might be schools. It might be providers. It might be the jail. Who knows? Then you got policymakers. And, ladies and gentlemen, if you want to do reentry and you want it to work for the long haul in your state, find a sweet spot. Find where you get the most agreements and the most consensus and work on it. You can work in these areas. Keep trying to expand this area, but you have to get everybody on the same page working together.

Some of our principles: read the research, do what works. When they complained about Kwame going to boot camp, here we go, the Detroit News came out, “Boot Camp Could Cut Kilpatrick's Sentence,” and they're complaining. So they called our office. I said, “Well, let's go to the NIJ website,” and we looked at some of the research out there, and [inaudible] done some work. So they came out the next day and said, “Oh. Hey, guess what?” and they looked at the research. Find out what works, what doesn't.

Understand the long-term history and perspective of the stakeholders you're working with. You have to use a collaborative approach. You have to step outside your own organization and your own safety zone. If you're a policymaker, please understand you can't impose. You have to work with everybody.

Involve your funding units. They are not going to fund if they don't think it works.

Involve your victims. We had a hearing several weeks ago on abolishing truth in sentencing. We had one homicide's … the wife of the homicide victim came in and testified, and when she got done, the committee went on another hour and a half, but we were done, and it was the end of the package.

Agree right up front. Use a comprehensive approach. Focus your resources, evidence-based practices. Be data driven. You're going to have ugly incidents, and you're going to have … the Detroit Free Press said, “Oh, look, he was a serial killer. He walked. He killed again.” You have to stay focused, look at the broad picture, have a proactive media plan, work with them, tell them what you're doing. Be flexible, so you can change whatever it is, wherever you are in the state, and don't mitigate problems. Solve them if you can.

Critical goals. Set priorities. Stick to priorities. Settle priorities. Don't get off task. You have to keep on, despite setbacks. You need to have local buy-in and trust. If your local probation officer culture is one that doesn't like the people they work with, you have a problem. You need to work with them, and I'm not saying that is the case.

Have a comprehensive stakeholder approach. You need sufficient funding to not only start but to keep going. Inform the media. Take a balanced approach, funding and policy, local, state, punishment, rehabilitation.

Real fast, dangers: You stop doing what works. You do what you want to do. You start overbalancing. We're spending too much on housing, not enough on employment issues. You implement too quickly. In Michigan, we went from no people being involved. The next year, we had thousands of people involved.

As a consequence, they have changed the standards of who qualifies for this program. They just changed it last week. I talked to one of our statewide providers on Friday, complaining bitterly. They just found out they no longer qualify because the standards just changed. It gets everybody frustrated.

Looking good versus doing good. Folks, you have to have results and make a real-life difference. You can't just have reports that say everything's fine, we're doing good, your money is being spent wisely. You have to have results to back it up.

And you have to be sure that as you work with people, you get buy-in from everybody. No time for Dilbert.

So, my parting advice to you: give people hope, and give them a plan. If you give people hope, they'll go running off in 10 different directions. The vendors will get a hold of them. Stakeholders will get a hold of them, and the next thing you know, everybody is running in a different direction, and it's not evidence-based practices necessarily, and everyone wants funding for what they're doing. If people have a plan and no hope, they'll go through the motions. They'll go around what you want to do, and it won't be particularly effective. Give people hope, and give them a plan.

You all come to Michigan, and when you do, give me an e-mail first. I'll hook you up with the appropriate group, people you can talk to there. We'd love to have you out there. It's a beautiful country. Thank you.

[Applause.]

Marshall Clement: Good morning. I know what it's like for me when I go into meetings with John Lazet and Michigan State Senate in my head. After an hour, I leave; my head's spinning. I've never met someone who talks faster and imparts more content in such a short period of time. It speaks to the quality of the legislative staff at Michigan, which has been just wonderful to work with.

So my name is Marshall Clement. I'm the project director for justice reinvestment with The Council of State Governments. I'm just going to try to connect the dots between some of the presentations thus far and provide a little bit of background on the process that we used in these two states and other states where we've worked.

So I don't need to reiterate the problem, right? Spending on corrections is up in states around the country. This is New Hampshire, which Anne mentioned, recidivism rates. Within just two years, their recidivism rate climbed from 40 percent to 51 percent. We talked about reentry programs dropping recidivism by 10 percent or 5 percent. They've got a problem situation where recidivism is increasing by over 20 percent, right, 11 percentage points within just a two-year period.

And, obviously, state spending at current levels is just simply unsustainable, let alone the projected increase at a lot of states, like Indiana, North Carolina, Oregon, other states are seeing for their prison populations and associated costs that would come along with that.

So, in this climate, policymakers need good data. They need a strategy, obviously. They recognize the need to have a strategy to reduce spending on corrections, to increase public safety, but they have also good reason to be skeptical, right, of just picking something off the shelf or picking up the latest trend, investing in programs, et cetera, and the evidence for this isn't better than coming from Ohio where Dr. Ed Latessa — many of you are probably familiar with his research looking at residential treatment programs and residential community corrections programs and the outcomes that they're getting. We're working in Ohio now. They spend $100 million a year on residential community corrections program. This is a huge, huge investment, especially for the current budget times that we're in.

John talked about the $57 million they're investing in reentry programs just for parolees. This is $100 million for probations and parolees.

The good news is there's a program that Ohio is funding that reduces recidivism for that group compared to a match sample by 50 percent. That's tremendous. But the problem is there's also a program in another part of the state that increases the recidivism rate by 50 percent, and on and on we go until basically you look at this graph and you realize that pretty much the hundred million is a wash, right? All their impact on recidivism has been outstripped by those programs that are increasing recidivism by not following evidence-based practices, by focusing on low-risk offenders, et cetera. And so, in fact, Ohio for spending a hundred million is basically where New Hampshire is, having not spent a dime and not having any impact on recidivism.

So what can policymakers do to get good data to make sure that the strategies they take are going to have the appropriate impacts on both crime, public safety and corrections spending?

For the last six, seven years, we've developed this justice reinvestment approach with support from BJA, Pew Center on the States and other private foundations and working in states across the country. It starts with an interest and commitment by state leaders from all three branches, and this is what makes it unique from most any other research or work in the states that's trying to influence policy. We start with a commitment from the governor. The senate president, the speaker, the chief justice have to agree that they want to actually work together and pursue this data-driven approach. They have to set up a work group or a task force. They'll have their direct representatives.

In New Hampshire's case, this completely blew us away. Every time we would go to New Hampshire, Anne's boss, the attorney general, would chair a meeting that included, literally every meeting for three hours, the chief justice, the two presiding justices, the speaker of the House, the Senate president, the legal counsel, and the commissioner of corrections and other cabinet members. It was a truly remarkable group. For hours on end, they would dive into the research. In no other state have we seen that, but it's a kind of a testament to the commitment that state policymakers have to address these issues.

We have kind of a three-phase process, pretty straightforward: first, to just analyze the data and develop the policy options in that iterative process that Anne mentioned; second, to work on the implementation issues that John highlighted are so critical to actually making sure you're getting the impact you want; and then, third, making sure the state has the capacity and a dashboard of indicators that are going to actually help them track whether or not these policies are having the intended impact.

So, as you heard from some of the presentations, we look at a comprehensive view of the criminal justice system. We're looking at crime, arrests, conviction, jail, prison, probation/parole data over a 5 to 10 year period, case-level data to the extent we can get it to really analyze this issue from all angles.

We engaged hundreds of stakeholders in each state from all parts of the system. There's only so much the data can tell you. Often what the data tells you is actually wrong because bad data systems, as you all are familiar with, analyze parts of the system.

We map prison admissions and social services to look at mismatching and spatial opportunities, and then, finally, we developed a set of policy options that, as John and Anne both mentioned, first are … recognize the political context, right? We start with what the policymakers agree to. So, in both New Hampshire and Michigan, the first conversation we had with them was what's off the table. Well, addressing mandatory gun laws, mandatory drug laws and truth in sentencing, they're off the table. Find out what else we can do. And so we start with a context in which we understand what policymakers can actually be enacted and will be pragmatic.

We worked to identify the assistance from BJA and other entities that are going to be needed to actually translate these policies into practice. We'd help to coordinate an implementation plan that keeps policymakers engaged, and then we'd deploy the reinvestment strategies.

And then, finally, this tracking impact. Inevitably, some policies aren't going to be implemented correctly. Some policies aren't going to work or have unintended consequences that no one can see, when you need to make mid course corrections.

So you've heard about New Hampshire and Michigan, two states that are still early on in the process. I wanted to share just some of the state experiences from states we've worked in in years past and what they have seen in terms of some of the outcomes.

We've worked or are currently working in 13 states around the country that are indicated here, starting back as early as 2004 in Connecticut and then moving through some of these other states in the intervening years.

In Kansas, where we worked in 2007 with Republican legislators and Governor Sebelius, a Democrat, at the time, they were facing projected growth in their prison population of 27 percent. They were going to have to spend about a half a billion dollars over the next 10 years in order to absorb that growth in construction and operating costs. And so the question that Republican majority leader was asking was, “If we're going to spend half a billion dollars to impact crime, is building the next wave of prisons going to be the best investment we can make? Let's first examine why our prison population is increasing to understand if that's the case.”

So we looked at their prison admissions, similar to some of the other states you've heard from, probation and parole revocations driving admissions, right, not only driving admissions but a huge percentage of the population, even though they spend less time, still almost 30 percent of the prison population. And the point of the state was, “You're already spending $53 million on reentry. It's just, you're spending it when people fail in the community and come back.”

So, in response to this analysis and obviously much more analysis that we developed for policymakers in Kansas, did a number of things as part of Senate Bill 14, which they passed overwhelmingly in 2007. The key thing they did was take some lessons they'd learned from reducing parole revocations through risk assessment, through evidence-based practices, cognitive behavior, motivation, and other training they'd done with staff, and had to figure out how to apply that to their probation system, which was scattered, locally run community corrections programs.

And so the legislature established a grant program, put $4 million into it, and said, “It's competitive, but the goal is to reduce probation revocations in each of these jurisdictions by 20 percent. Come up with a credible plan that uses evidence-based practices that focus on the high-risk offenders to do that, and we'll award these grant funds accordingly.”

They did that, and since 2006, they've seen probation revocations drop 26 percent statewide. Some jurisdictions have dropped even by 40 or 50 percent within just two years.

Parole revocation rates, which had already been on the decline, continued to decline with additional funding, reinvestment from the state by 48 percent. Parole absconders are down 70 percent, and the parole reconviction rate, a key indicator to know if we're just ignoring behavior or if we're actually changing behavior among those on supervision, down 35 percent from the two-year averages that we're looking at there.

As a result, they've averted all the growth, and it projected in their prison population. This is the 2007 projection and the 2008 projection, which assumed the bill that they passed, Senate Bill 14, and then you can see the actual population in the blue line down below. They are actually getting better results than we anticipated. In part, that's because they've done better. In part, that's also because we're incredibly conservative about the impact that these policies can have when we estimate the cost savings.

As a result, they not only avoided half a billion dollars of prison construction they were facing back in 2007, but they've also actually been able to close a couple of small prisons over the last couple of years as a result of these declines.

In Arizona, their prison population was growing tremendously but primarily due to resident population growth, not because of other trends, and so policymakers there, Republicans in the Senate, who were kind of taking the lead on this effort with us, really wanted to identify a crime-fighting strategy. Arizona has one of the highest crime rates in the country, and instead of looking at reentry, which is a fairly small population because of truth in sentencing, very few people on post, really, supervision in Arizona, they wanted to look at probation, which is a large population and a large opportunity to reduce crime.

And so they set up a performance funding incentive mechanism to drive performance from the state level, at the local level in terms of county probation. So every year now, the legislative budget staff calculates the probation failures, people in each county and by county, and they asked for each county, “Has crime gone up? Has the rate of new convictions, new felony convictions for people on parole/probation gone up or down?” If it's gone up, there's no funding, end of story, but if it's gone down and the revocation rate went down to prison, then the legislature has said in law that they will provide the county with 40 percent of the averted costs at the state level, and that's calculated on a fully loaded $20,000 a year, $20,000 per revocation, because there's a number of contract beds that Arizona could close to generate those savings. And the local probation agency can reinvest that in drug and mental health services as well as victim services. Those are the two designated areas for which they can use those funds.

In just one year, from 2008 to 2009, we've seen probation revocations to prison fall 13 percent in the state. They didn't just shift these to jails, which are county funded in Arizona. Those also dropped 14 percent, and there had behold or slightly reduced, depending on the county, the rate at which people on probation were committing new felony convictions. As a result, the state estimated savings at around 17.8 million in just one year after implementation.

Number of issues to discuss we can go into in terms of Arizona and actually carrying out the reinvestment. Guess what? Budget crisis. “Sorry, we can't do it this year.” Passed a law saying, “We'll come back next year and pick this up again when we have new resources,” so huge implementation challenges in terms of actually coming through with a reinvestment that they promised back in 2008.

In Texas, we worked in 2007. They were facing this 2007 baseline projection. Their prison population was going to grow from 155,000, about the size of my hometown in Oregon, in 2006 to over 163,000 in 2009, just five years. They are going to have to build 17,000 new prison beds. And policymakers there, after building 100,000 prison beds over the last decade and a half and spending a billion dollars to do so, started asking themselves, “So when does this, you know, prison-building thing stop? Right? We just keep doing it and doing it, and it's got to stop at some point, right? What's the answer?”

And so a researcher, director of research, in Texas came in, did an exhaustive analysis and found a number of areas, number of reasons. One reason was that due to budget cuts, the legislature had to cut in-prison drug treatment. Guess what? The parole board didn't stop demanding that people participate in drug treatment, even if they were low-risk drug offenders, and so you had 2,000 people taking up, you know, several prisons in Texas because they were waiting for that program that had been defunded. That, probation revocations, parole revocations, a number of things that were causing the prison population to continue to increase.

They put forward a proposal, a set of massive investments, $241 million in a two-year biennium, invested in intermediate sanctions for probation and parole revocations, expanded outpatient drug treatment, mental health treatment, the nurse-family practitioner partnership program — a range of things across the state of Texas that enabled them to actually pursue over the last three years this, basically, aversion of all that growth in the system. Just in two years alone, they saved $443 million, and same time as that, parole revocation is down 25 percent, probation revocations statewide down 4 percent.

So I want to … I know we're running out of time. We want to leave time for question and answers. We've been throwing a lot of stuff at you, but just to highlight some key themes and challenges that maybe we can pick up in some of the question and answers.

The first is that in justice reinvestment, we're analyzing trends across the entire criminal justice system. So there's some confusion — not confusion but disagreement within our staff about whether this is actually able to be read by policymakers or if it's just a confusing mess of a presentation, but, like it or not, this is the mess that is the criminal justice system, and it's confusing. But we need to understand all the boxes and what's happening in each of these trends in order to actually see what's happening, right? Green boxes, you're getting better outcomes. Red boxes, you're getting poorer outcomes. And you need to use this to diagnose kind of where the problem is in the system and what you need to be focusing on.

So, for New Hampshire, which is this, this chart is for, helped us realize that crime is low. There's some front end pressure in the courts, but, guess what, it's not coming to prison. It's not what's driving a prison population. Probation revocation is stable. It's all the back end. It's all parolees that are driving the increase in the prison population there.

This is true in most of the states we're working in, not all, where we're seeing parole and probation revocation as being the main driver.

The second major area — and this is not a surprise to researchers; the research has been clear for a very long time about targeting resources on those that have the — most likely to reoffend, right? We come up with fancy names for this, “risk principle” and all these other things, but the end of the day, it's just targeting resources to those who are most likely to cause the most harm. It's a pretty simple concept.

And we know we have risk assessments that can, you know, group people into low, medium and high risk, and the re-arrest rate will vary with those groups. Yet, unfortunately, in a lot of the states we're working, the supervision levels based on those risk levels looks pretty much like that, right? There's a huge mismatch between what we know about risk and where the states are.

The state we're working in currently, which I won't mention the name of, 100,000 people on probation and parole. They are supervising … they do a risk assessment, have been doing it for five years. It's very clear the differences we see between the high , medium and low-risk groups. Everyone is supervised with the exact same contact standards, regardless of their risk level. And guess what, they're running out of probation resources, so they're hiring a bunch more probation officers because they need to reduce caseloads.

Another point is we know we need to reallocate program researchers to increase the impact on recidivism, right? Thanks to a lot of the meta-analyses that have been done and supported by NIJ and Steve Voss' work and others, we know what works and what works more than other programs, but when we look at most … when we go into a state and we look at where dollars are allocated currently, they're not aligned with the research. Obviously, there are different programs, and people have different preferences.

The biggest is that, for instance, in New Hampshire, there's a fair amount of money being spent on programs to reduce recidivism, but it's all being spent behind bars, right? It's not being spent on the community.

For instance, we know that drug treatment provided in the community is going to be roughly twice as effective as it is behind bars in prison, and so a huge reallocation is not only between programs that don't work, like boot camps, to those that do, but, also, from inside the facilities, out into the community, we're looking to have a bigger impact.

The other issue is not just funding those programs but making sure your funding programs are high quality. In every state we've worked in applied justice reinvestment, maybe with the exception of Michigan, there has never been a full-on evaluation of any of the programs they're currently funding, and it's just not feasible to do that, right, even though that's what we'd prefer.

And so using, whether it's a CPAI or CPC tools, tools that assess the quality of the program, very cheap, very quick, and easy to administer, we can get a sense of how high quality these programs are and be able to correlate that and some of the research with expected impact on recidivism.

So here's some of Latessa and his colleagues' research again, looking at Ohio programs. Fourteen percent of the programs they're funding are rated poor, and when they looked at the recidivism rates for those programs, they were increasing recidivism by 15 percent.

The good news is some programs were actually decreased in recidivism, but only about 5 percent of the programs were in a major way, and 26 percent actually saw a 12 percent decrease. That's not too bad. But over half the programs weren't having basically any impact at all. Researchers need this kind of real-time evaluation research in order to know what programs are working and which ones aren't. They can't wait around for long-term research findings when they're funding, especially in the case of Ohio and Michigan, lots of programs at lots of different levels.

The final point — and this is more of something that we're eager to maybe talk about with many of you who are studying this kind of work — this is research from Michigan where we looked at everyone who had been arrested and everyone who's on probation and parole and looked at the overlap because we wanted to know … we know we've got a crime problem in Michigan. We want to know to what degree is probation and parole an effective tool at reducing crime, serious crime, and so we looked at the arrests for index crimes, which we know are low. There's low clearance rates in Michigan. What we found is only 10 percent of those arrested for index crimes in 2007 were on probation or parole at the time of their arrest. Now, obviously a lot more were on probation or parole in their past or at some point in their history, but this has huge implications for how policymakers think about what probation and parole can do to reduce serious crime. There's no doubt they can do a lot, but even if we reduced recidivism to zero for everyone on probation and everyone on parole, we'd still have 90 percent of our index crimes, this research suggests, occurring. And so this has huge implications for how we think about and how we communicate with policymakers and how we need to reinvest not only in probation and parole systems but also in law enforcement, crime prevention and other parts on the front end of the system.

So, with that, I'll stop, and I think we want to take questions from those of you who have them.

[Applause.]