Skip to main content

Ending Mass Incarceration: Charting a New Justice Reinvestment

April 24, 2013
This report outlines how the original mission of Justice Reinvestment can be achieved by focusing on reducing incarceration and targeting investments in high incarceration communities.

In the decade since Justice Reinvestment was launched, much has been accomplished.  With assistance from the Justice Reinvestment Initiative, around 27 states have participated one way or another in this data-driven reform, which has revealed previously unacknowledged realities about U.S. correctional systems.  Throughout this paper, “Justice Reinvestment” connotes the concept and originating idea and the “Justice Reinvestment Initiative” or “JRI” refers to the formal implementation strategy spearheaded by the Council of State Governments (CSG) and its now principal funders, Pew Charitable Trusts (Pew) and Bureau of Justice Assistance (BJA).  Of the 27 states that have participated in the JRI, approximately 18 have enacted JRI legislation for the purpose of stabilizing corrections populations and budgets.1)For purposes of this report, we consider the following 27 states to have participated in JRI: Alabama, Arizona, Arkansas, Connecticut, Delaware, Georgia, Hawaii, Indiana, Kansas, Kentucky, Louisiana, Michigan, Missouri, Nevada, New Hampshire, North Carolina, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, South Carolina, South Dakota, Texas, Vermont, West Virginia, and Wisconsin.  We consider the following 18 to have enacted JRI legislation: Arizona, Arkansas, Connecticut, Georgia, Kansas, Kentucky, Louisiana, Missouri, Nevada, New Hampshire, North Carolina, Ohio, Oklahoma, Pennsylvania, Rhode Island, South Carolina, Texas, and Vermont.  We do not include Michigan in the list of states that have enacted JRI legislation because we understand Michigan to have dropped the bill proposed in connection with the JRI process there.  JRI has played a major role in educating state legislators and public officials about the bloated and expensive correctional system, persuading them to undertake reforms not previously on the table. Considering the country’s four-decade addiction to mass incarceration and harsh punishment, the general refusal to acknowledge its failures and the monumental resistance to change, JRI’s most enduring contribution to date may be its having created a space and a mindset among state officials to seriously entertain the possibility of lowering prison populations.

Because Pew and CSG maintain that the JRI, as they have administered it, has helped states reduce incarceration, it makes sense to look at the impact of JRI thus far.  Our analysis, described in the pages that follow, leads us to the conclusion that while JRI has played a significant role in softening the ground and moving the dial on mass incarceration reform, it is not an unmitigated success story; the picture is complex and nuanced.  The Justice Reinvestment Initiative, as it has come to operate, runs the danger of institutionalizing mass incarceration at current levels.

The destructive effects of mass incarceration and harsh punishment are visited disproportionately upon individuals and communities of color.  Justice Reinvestment was conceived as part of the solution to this problem.  The intent was to reduce corrections populations and budgets, thereby generating savings for the purpose of reinvesting in high incarceration communities to make them safer, stronger, more prosperous and equitable. Increasingly, the formalized JRI has moved away from these progressive goals, seeking to reduce the rate of prison growth rather than reduce the number of prisoners. Instead of producing savings that enable reinvestments in high incarceration communities, in too many places we are likely to institutionalize our massive incarceration rates for many decades to come, largely ignoring its impact on communities of color in particular. This is not the future we want.

Some  have  argued  that  compromise  necessitated  by  politics  and  the  need  to reach  stakeholder consensus has meant that avoiding projected prison growth has become more feasible than achieving substantial reductions in actual prison populations.  Possible savings in the form of “averted costs” for JRI work have been either returned to the general coffers or used to augment community corrections and law-enforcement government budgets. By focusing on state-level political and administrative policymakers, the JRI process has too often marginalized well-established local advocates and justice reformers who bring knowledge of local conditions and politics to the table, and who have a vested interest in providing long-term implementation oversight and ensuring sustainability of reforms.

We believe this is an important moment to take stock of the JRI, especially within the context of an unusually favorable climate to challenge mass incarceration, and assess how to get a more ambitious Justice Reinvestment movement back on track. We believe that a revived, reoriented Justice Reinvestment  effort  could significantly reduce  U.S.  corrections  populations  and  costs; and through smart, targeted and locally determined reinvestments, could aid substantially in repairing the destructive impact of high levels of concentrated incarceration on poor communities of color.

Achieving this goal will require meaningful, multi-year support for a range of coalitions and networks to pursue strategic goals consistent with the principles of Justice Reinvestment aimed at reducing admissions to corrections systems and lengths of stay, changing incentives for systems players, and reinvesting in public safety by strengthening community institutions.  We are not naïve to the challenges associated with long-term corrections reform, but we believe it would be irresponsible not to seize today’s opportunity for change with a much more substantial impact in mind.


After a half century of relative stability, the American penal system for the last forty years has been dominated by relentless growth.  The culture of punishment, in part driven by political expediency with “tough-on-crime” policies marketed as the solution to “fear of crime,” has been aggressively implemented at every stage of the criminal justice process:  arresting, charging, sentencing, confining, releasing, and supervising.

Today, there is general agreement that this vast expansion of correctional control occurred not by accident, but as the result of deliberate policy choices that impose intentionally punitive sentences that have increased both the numbers of people entering the system and how long they remain under correctional control, as illustrated in Table 1.

Table 1: Changes in Correctional Populations 1980-20112)Unless otherwise indicated, all data presented in Tables and Figures throughout this paper is from the Bureau of Justice Statistics, Correctional Populations in the United States data series.
Population 1980 2011 % Increase
Prisons 319,598 1,504,150 371%
Probation 1,118,097 3,971,319 255%
Parole 220,438 853,852 287%
Jails 182,288 735,601 304%
Total 1,840,421 6,977,700 279%
U.S. Population (millions) 227 311 37%

Of the seven million people currently under correctional control in the U.S., a disproportionate number come from a small subset of neighborhoods in the major cities of each state.  Overwhelmingly black, Latino, and poor, the residents of these neighborhoods are those most likely to suffer from high rates of unemployment and poverty; homelessness; and sub-standard schools, healthcare, and other basic services.   These are the same neighborhoods to which the vast majority of people return (“reenter”) after being released from prison and where more than four million people are under the surveillance and supervision of the state.  This excessively punitive and racially charged system exacerbates injustice, breeds resentment, and undermines the legitimacy of the justice system itself.  The consequences are well documented: a five-fold increase in the size of the penal system at tremendous social and financial cost to the country, producing a level of mass incarceration on a scale never before experienced.

For decades, the culture of incarceration and extreme punishment has thwarted or watered down most reform efforts. Today, however, these dynamics seem to be shifting, and the JRI, as mentioned above, has been a factor in this shift.  In nearly every state of the country, a political premium has developed in favor of containing correctional costs, scrutinizing proposals for further growth, and considering strategies to downsize correctional populations and budgets that were out of the question just a few years ago.

Fiscal pressures are partially responsible for this shift, but other forces are in play as well. Falling crime rates have reduced the salience of fear-of-crime as a political imperative, weakening the unspoken calculus that being tough-on-crime may be bad policy but good politics.  Other concerns, such as the war on terror, jobs, healthcare, education, immigration and climate change have replaced public safety as top national priorities.  The failure of the War on Drugs, along with its destructive socio-economic consequences, has increased public support for effective alternatives that are less costly and punitive.

The prospects for state finances look bleak for the near future, though they will likely ease as the country comes out of the recession.  While today’s new reality for penal reform may not last forever, other factors that play a prominent role in altering the dynamics of penal policy could outlast the current fiscal crisis and sustain momentum for change even as state fiscal problems ease.

We are encouraged by these trends—and are active participants in many of the efforts now underway to take advantage of them—but we are also concerned that the moment calls for more. Current efforts, particularly under the rubric of the JRI, aim too low, achieve too little and lack the local organized support—eclipsing  even—much  of  the  reform  energy  and  support  needed  to realize sustained reductions in the U.S. correctional population. This has resulted in legislative achievements which do not go deep enough.  We think it is time to bring to the table a more comprehensive and effective justice reinvestment strategy.

Justice Reinvestment was developed as a public safety mechanism to downsize prison populations and budgets and re-allocate savings to leverage other public and private resources for reinvestment in minority communities disproportionately harmed by the system and culture of harsh punishment. Initially,  it  sought  to capitalize  on the  nascent  shift  away  from  “tough-on-crime”  sentiment  by highlighting the trade-offs between primarily punitive (and expensive) prison spending and prospective public safety investments in local community-building institutions and services.

In spite of its successes, JRI has been unable to reduce correctional populations and budgets below the historically high levels which persist today.   Nor has it steered reinvestment toward the communities most weakened by aggressive criminal justice policies.  The JRI’s limitations are certainly not due to lack of competence or rigor.  On the contrary, the JRI’s methods are sophisticated and effective and have succeeded in building coalitions of state lawmakers and public officials to support passage of legislation in state after state.  But an important source of the JRI’s strengths—special access to state officials to provide detailed information, analysis and recommended policies—is also the source of one of its limitations: primary accountability to those state officials.

There is little question that the JRI has brought conservative politicians (sometimes aligned with more liberal politicians) to the justice reform table in a number of states.  Indeed, among many lawmakers, the “Justice Reinvestment” label has come to stand for any correctional reform effort that is expected to save states money and improve public safety, but without the concomitant reinvestment in community and, it turns out, without significantly reducing correctional populations.   Increasingly, stripped of its core purpose—to downsize corrections and increase community reinvestment—Justice Reinvestment risks becoming all things to all people.

The major reasons that the current JRI approach will not achieve the dual objectives of sustained reductions in state correctional populations and stronger, safer communities can be summarized as follows:

  1. Current JRI efforts that focus on crafting legislation often incorporate statutory reforms that will not significantly reduce admissions and lengths of stay – especially for people convicted of serious and violent crime  When promising strategies for affecting these two drivers of the prison population have been proposed, they are too often the first to be compromised as a part of the legislative process. This is in part because the JRI puts a premium on the passage of legislation, even when the legislation will not lead to meaningful reductions.
  2. JRI  activities  are  typically  focused  on  state  government  policy  makers  and  state-level reforms,  eschewing  and  sometimes excluding other important  state  and  local constituencies.  Efforts may have been made  to reach  out  to  other  stakeholders,  for example in the form of town hall meetings, but this is not the same as having local stakeholders directly involved early on to help develop a decarceration strategy.
  3. The JRI focus on intensive, short-term analysis and technical assistance, aiming for near- term legislative packages in a large number of states, does not build state and local capacity to assume responsibility for monitoring and evaluating implementation and outcomes for genuine justice reform over the long-term.
  4. Increasingly, JRI has abandoned reinvestment in high incarceration communities as a key element and goal of the initiative which, absent demand-side advocacy from coalitions of local stakeholders and leaders for local reinvestment, leaves resources vulnerable to the claims of other criminal  justice  agencies,  including increased investment  in law enforcement.  A key but missing element of Justice Reinvestment has been an organized coalition of demand for prison reductions and reinvestment in community institutions from local coalitions of city and county officials, grassroots leaders, residents, and service providers.
  5. There has been insufficient  attention  to the problem of structural disincentives that discourage and inhibit officials at all levels of government from pursuing local, innovative, non-incarcerative public safety strategies.  The state subsidy for imprisonment and harsh punishment relieves public officials across the criminal justice spectrum of the consequences—both social and financial—of the punitive policies, practices, and discretionary decisions which have led to mass imprisonment.

Despite these limitations, there are opportunities to pursue meaningful, long-lasting reform that can build upon the JRI work conducted to date.  We must aim higher and go further than the JRI has been willing or able to go in recent years.

In this paper, we set forth a vision for a revamped, reenergized Justice Reinvestment program that we think will achieve these goals.  This vision calls for the creation of multi-sector campaigns coordinated by coalitions of locally based grassroots organizations, grass-tops leaders and in-state advocacy groups, national advocacy organizations, state and local lawmakers, researchers and policy analysts, and communications professionals.  Together, these coalitions could identify the drivers of state and local corrections populations, the policy mechanisms needed to make major reductions in these correctional populations, and the pertinent political pressure points.  They could mount sophisticated, multi-faceted public education campaigns.  The overarching goals would be to create sustained demand for long-term corrections reform, major cuts in overall correctional populations, and establish investment in high incarceration communities.

The political challenges are many.  But such a revised strategy that directly focuses on high incarceration communities would have distinct and measurable advantages:  immediate and tangible impact on the ground; sustainable reforms rooted in local interests, practices and needs; viable state and local-level policy prescriptions; and ongoing implementation oversight.  The JRI has played a critically important role in turning lawmakers’ attention to the importance of correctional reform.  Its first decade of high- paced activity has laid the groundwork.  We believe the ideas offered in this paper will inform and reinvigorate the next phase of Justice Reinvestment work.

A Decade of Justice Reinvestment

Justice Reinvestment originated as an ambitious strategy to reduce reliance on incarceration and repair the attendant harm to individuals and communities through reinvestment in neighborhoods with high concentrations of residents in criminal justice systems.  But it became weaker over time, for the reasons discussed in this section.

Origin and Evolution: 2002-2012

The initial purpose of Justice Reinvestment was to make state government accountable to impoverished communities—mostly (though not exclusively) black and Latino—where the burden of punishment and incarceration has been heaviest.  These already disadvantaged neighborhoods were being driven deeper into perpetual economic divestment, social isolation, political disenfranchisement, and physical distress by the coercive, downward mobility caused by locally concentrated mass incarceration and the forced migration of residents to and from prison.  The most locally concentrated pockets of incarceration were dubbed “Million Dollar Blocks,” because of the millions being spent each year on prison cells for high proportions  of  working-age  male  residents  for  an  average  of  three  years.    Million  dollar  blocks dramatized the trade-offs for specific neighborhoods between locally concentrated incarceration spending policies, and alternative, locally focused investment policies that could yield greater returns in public safety, strengthened community institutions, and expanded neighborhood networks.

The JRI was launched during the recession of 2001–2003, when state budgets were under considerable stress and no new dollars were available to fund reform efforts. With support from the Open Society Institute (now Open Society Foundations), criminologists and other experts under the leadership of the JFA Institute and the Council of State Governments (CSG) undertook to transform the broad Justice Reinvestment concept into a specific set of initiatives, and JRI was launched as a three-part strategy:

  • Work with state legislatures to analyze criminal justice populations and  budgets and recommend ways to reduce them to generate savings for reinvestment in local high incarceration communities;
  • Engage development experts to identify and steer investment opportunities; and
  • Organize demand by affected communities, advocates and institutions for neighborhood reinvestment.

Between 2002 and 2008, JRI legislation adopted by Connecticut (2004), Kansas (2007), Texas (2007), Rhode Island (2008), and Arizona (2008) achieved initial successes and galvanized latent interest for reform in state government, particularly around stemming the high cost of imprisonment.   In these states, the JRI sought to target administrative policies and practices of state correctional systems, intentionally aiming for low-hanging fruit, such as: reducing revocations for technical violations of parole and probation, holding parole hearings at the point of parole eligibility, or re-establishing earned “good time” credits.  Early Justice Reinvestment proponents believed that these measures could win bipartisan support, achieve some level of systemic reduction in correctional populations, and serve as a wedge for more ambitious sentencing reforms in the future.

JRI legislation in these early states has shown mixed results:

  • In Connecticut, the state prison population was already on the decline when JRI legislation was enacted in 2004. From 20,720 in 2002, it fell to 19,442 in 2005, and jumped six percent in 2006 and another two percent to an all-time high of 20,924 in 2007.  Finally, the numbers started to decline and by 2011 the population was at 18,324.
  • Kansas’s population stood at 8,539 in 2008, and grew to 9,327 by 2011.
  • In Texas, the prison population went from 171,790 in 2007 up to 173,648 in 2010, then down slightly to 172,224 in 2011.
  • Rhode Island’s prison population jumped from 3,654 in 2006 to 4,045 in 2008, and then fell to 3,337 in 2011.
  • Arizona’s prison population started at 39,589 in 2008 and grew to an all-time high of 40,627 in 2009, before declining modestly to 40,020 in 2011.

In most of these early states, JRI measures to reduce prison populations were explicitly tied to commitments by the state to invest some portion of the savings in targeted “Million Dollar Blocks.” To date, however, there has been virtually no reinvestment in education, employment, community revitalization or affordable housing development in those communities.  Instead, JRI-guided legislation has increasingly channeled modest reinvestment into community corrections and, more recently, law enforcement agencies.

In these early years, the JRI’s approach as implemented by CSG and JFA stuck relatively close to its enunciated four-step strategy:

  1. Analyze the prison population and spending in the communities to which people in prison often return.
  2. Provide policymakers with options to generate savings and increase public safety.
  3. Quantify savings and reinvest in select high-stakes communities.
  4. Measure the impact and enhance accountability.

But between 2008 and 2012, the language used to articulate the JRI’s revised three-step approach became increasingly generalized and open-ended, reflecting a change in orientation and purpose that produced a different implementation strategy:

  1. Analyze data and develop policy options.
  2. Adopt new policies and put reinvestment strategies into place.
  3. Measure performance.3)Council of State Governments, “The Strategy: How Justice Reinvestment Works.”

The increasingly generic language signaled less of a focus on reducing prison populations. Hence, the terms of JRI reform today are expressed in the language of crime control and law enforcement, rehabilitation and treatment, recidivism reduction, and expanded, often tougher, parole and probation supervision.

This shift in orientation is viewed, by some, as necessitated by political realities—that the focus on law enforcement has been necessary to win support and consensus among legislators and other public officials for the JRI.  Indeed, the popularity of the JRI among state officials has, to a large extent, made “justice reinvestment” synonymous with any and all criminal justice reform, in some cases crowding out and marginalizing other reform efforts with more aggressive agendas.  No longer is the JRI about reducing correctional populations or reinvesting in high incarceration  (which morphed into “high stakes”) communities. For example, in Wisconsin the JRI used incarceration mapping for the last time in 2009 to link state prison population reduction to reinvestment in high incarceration neighborhoods. Since then, the focus has been on crime mapping, which moves the essence of the reform agenda definitively away from the community-level impact of mass incarceration to crime control.

The net result is that the primary purpose of the JRI is no longer systematic reduction of prison and other correctional populations and racial disproportionality accompanied by reinvestment in high- incarceration communities.

Limitations of the Current Strategy

We have no doubt that some important milestones have been achieved that would not have occurred had JRI not existed.  And we acknowledge areas where JRI states have been leaders in important reforms.  But our conviction is that unless a more aggressive and focused JRI emerges soon, high incarceration rates that persist today will become the status quo. There are five major limitations to the current approach.

Insufficient Targeting of the Key Corrections Reduction Policy Mechanisms

Correctional populations are a function of two factors: admissions (including revocations to prison from probation and parole) and length of stay.  If policy makers want to reduce the costs of corrections, they have to reduce the number of people who enter the system, their length of stay, or both.  This point is important, because, many reform efforts shy away from policy mechanisms that can reduce lengths of stay or overall admissions in favor of programmatic strategies that seek to reduce recidivism rates.  For example, reducing recidivism and re-incarceration is a laudable goal in and of itself, and recidivism- reduction measures might have an impact on correctional populations.

But recidivism reduction measures are insufficient to check overall corrections growth, which has more significant drivers such as new commitment admissions and increases in length of stay –especially for people convicted of violent crimes.  Too many of the current reform efforts try to achieve population reductions by programmatic initiatives, such as increasing the availability of drug treatment slots, strengthening reentry-related services and supervision, and funding police reform (see below).  These tactics have limited capacity to reduce admissions or lengths of stay.  For example, in January 2013, a JRI legislative package put forth a set of proposals to West Virginia lawmakers that focused almost exclusively on reducing recidivism.  Its proponents projected that adopting the package would result in a 2018 prison population largely unchanged from its 2012 size. The package contained no recommendations to reduce new commitments or length of stay. Reform efforts that are founded on programmatic changes cannot affect prison population in substantial ways.

Limited Involvement of Well Established Local and State Advocacy Groups

Every state has an active core of criminal justice reformers. These are groups that have been pressing for progressive reform before the advent of Justice Reinvestment, and they will be working in the state long after outside reform efforts have left. There is a critical need to bring these local advocates into the Justice Reinvestment process, in order to promote a more substantial change agenda and to build a sustainable foundation for continuing reform. County and municipal officials need to be at the Justice Reinvestment table to help strike a Justice Reinvestment deal that furthers local interests and considerations and is responsive to the needs of their constituents.  This is particularly so in urban locales, where the history of reform is more robust. Some DOJ initiatives have successfully engaged policy-makers at the state and now local levels of government in questioning the value of incarceration as the primary crime control strategy. For example, BJA is implementing its Byrne Criminal Justice Innovation Program (BCJI) which is designed to reduce crime in persistently distressed areas which impede community revitalization efforts.  To its credit DOJ has recognized the need for a local JRI effort which it has funded on a limited basis.  But it too suffers from some of the same issues raised at the state level JRI (e.g., short-term Technical Assistance, lack of involvement by local advocacy groups, etc.).

Organizing and maintaining demand for an ambitious vision of criminal justice reform requires the inclusion of reform coalitions rooted in the longer-term interests and sustainable commitment of state and local constituencies, especially minority leaders and elected representatives of high incarceration communities, who are often markedly missing. But this cannot be accomplished through meetings with these constituencies in which the participants are not partners in decision-making.  Local leaders and advocates genuinely brought into the process could work to reduce incarceration rates, for example, by advocating for a change in policing strategies at the local level while also making the case for state law reform. They have a vested interest in ensuring reinvestment into high incarceration communities while sharing political risk with state lawmakers.4)JRI actors often tout the experience in Ohio as an example of effective interaction with state and local advocacy organizations and stakeholders. We agree that Ohio is, in a sense, an exception to the JRI’s general strategy of claiming that participation in town meetings constitutes collaborative policy analysis and development. But it is important to clarify that the dynamic was different. In Ohio, a modest bill was introduced in the legislature before the JRI came to Ohio and was later stalled once JRI got to Ohio. Ohio has an established network of active criminal justice reform advocates with a close working relationship with state correctional officials, who insisted that the advocates have a seat at the table. Advocates continued to fight for reform and the modest bill, including some JRI recommendations, was reintroduced in the next session.

For example, the JRI entered Oklahoma in 2011 to work with a bipartisan working group of state officials. JRI recommendations focused on providing increased funding for police departments via an arrest-rate-driven grant program and allowing prosecutors to have veto power over an individual’s request  for  sentence modification.  The  primary  reform  focus  was  no  longer  prison  population reduction; it had shifted to increasing funding for law enforcement and probation/parole staff, with a secondary emphasis on cost containment.  These concessions to law enforcement were made early on, before any local advocacy organizations or community groups were brought into the discussion. This resulted in a weak bill in 2012; notably, the state’s 2014 budget contains no mention of the 2012 law and the law’s original supporters are now urging repeal.

Limitations of Short-Term Technical Assistance

Under the current model, a state applies for JRI assistance to analyze and assess the drivers of its correctional  population.  Based on that  analysis,  the  state receives  a  set  of recommendations  that includes estimated impacts of proposed reforms on populations and budgets.5)Notably, JRI does not analyze the impact of its proposals on racial disproportionality or on specific communities.  The state then drafts legislation to adopt the recommended reforms. The entire process typically takes approximately six months to complete, and allows limited time or resources, if any, for implementation oversight, tracking the impact of reforms, or assuring that projected savings are reinvested.  One side effect of this strategy is the absence of input from well-established justice reform advocates;  another is that no one is onboard to ensure that reforms are implemented as intended.  Absent local analytical capacity or participatory authority to track and oversee JRI implementation and reform outcomes, short-lived technical assistance eventually gives way to increasingly watered-down, risk-averse policy mechanisms and inadequate quality assurance of implementation. While it is true that the BJA has recently funded Phase 2 follow-up technical assistance and monitoring grants, these monies were awarded to the Vera Institute as opposed to local organizations.

Insufficient and/or Misdirected Reinvestment

The original idea of Justice Reinvestment proposed that savings from reduced correctional populations be used to build stronger infrastructure in exceedingly high incarceration communities.  The reasoning for this was simple: these communities are the feeder system for prisons, jail, probation and parole, and strategic investment in them would be essential for long-run reduction of demand for “correctional services.” This was also an ethical argument.  Individually and collectively, residents of these communities—already suffering from social exclusion due to race, poverty, disenfranchisement, etc.—have been disproportionately subjected to the further destabilizing and downwardly mobile consequences of high incarceration rates; therefore, it is incumbent upon policy leadership to make investments that promote greater economic and social equality and stability.

The current JRI has not maintained the logical, pragmatic, and ethical links between prison reductions and reversing the systemic social and economic obstacles facing communities with high concentrations of criminalized residents.  In fact, we worry that some of the more recent reinvestment packages are likely to strengthen the very corrections policies that have played such a prominent role in the system’s growth in the first place. Increased funding for “intensive community supervision” (i.e., closer control and scrutiny) can result in higher rates of return to prison by widening the net of social control.  Even investment in rehabilitation services, such as drug treatment, can backfire if services are inappropriate for the individual or sub-par since relapse (which is common among recovering addicts) can result in revocation to prison.

The lack of targeted reinvestment in high incarceration communities is probably the most glaring weakness of JRI.  For example, JRI in Pennsylvania in 2012 made no attempt to advance a goal of community reinvestment, and instead promoted  redirecting  the  highest  proportion of anticipated savings to local law enforcement authorities, even though the state’s overall crime rate had dropped 14 percent.  A bill enacted in October 2012 required that a portion of savings go to a special fund to be distributed to state agencies and local governments:   $1 million to the Pennsylvania Commission on Crime and Delinquency for a statewide victim notification system and other victim services; another $400,000 to the Pennsylvania Sentencing Commission to establish risk assessment models for judges to use at sentencing; and the remaining funds to local police and county probation departments, the Department of Corrections and the state Board of Probation and Parole. Anticipated savings are not being directly re-invested in high incarceration communities.

Failure to Reduce Structural Disincentives to Local Innovation That Reduce Corrections Populations

Some of the chief impediments to reducing the overuse of state prison are the structural disincentives (financial and risk) inherent in current law enforcement and corrections systems.   If localities seek to develop alternative approaches to state prison, it is the localities which must bear the financial burden of establishing and maintaining those alternatives, which discourages local innovation and reinforces the easier and cheaper (to localities) option of prison, where the state picks up the costs.   Similar disincentives are present in probation and parole systems, where the risk of taking a less punitive approach falls squarely on individual officers, who have no incentive to take the risk of a client’s being rearrested — unless, of course, an administration deliberately and intentionally reverses the structural incentives to discourage revocations and encourage early discharge from probation, as has happened recently at the NYC Department of Probation.

Moreover, certain federal grant requirements create incentives for local police to increase their arrest numbers, often through “easy” arrests for low-level, nonviolent drug offenses and petty misdemeanors in minority neighborhoods.  Similarly, the massive 1994 federal crime bill provided $8 billion for state prison construction, in part conditioned on states’ adoption of “truth in sentencing” policies that increased length of stay in prison.  Some rural jurisdictions, persuaded by the promise of jobs to allow prisons to be built in their communities, are now fearful that reducing prison populations will devastate their fragile economies.   Whereas JRI addressed these kinds of disincentives early on in Connecticut and Kansas, attention to removing them has waned in later years.

Performance measures for prosecutors reward convictions, which incentivize prosecutors to “over- charge” so that defendants will plead guilty to avoid the risk of being convicted of a more serious charge at trial.   The same rational encourages prosecutors to oppose bail reform, since avoiding pre-trial detention is a strong predictor of a defendant’s avoiding conviction and incarceration.

Outcomes of Today’s Justice Reinvestment Strategy

One  of  the  major  trends  in  incarceration  has  been  the  stabilization  and  slight decline  in  prison admissions (new court commitments and parole violations) that began in 2006 and, in 2009, outpaced a similar decline in releases. Some stabilization and reduction in admissions can be attributed to reforms in technical violation policies, local police practices, and sentencing reforms (in both JRI and non-JRI states).  But other factors have also played a role, including declining crime rates and declining arrest rates.

Many of the JRI actors claim that their efforts have averted (and occasionally reduced) incarceration, but we believe these conclusions are often based on a misunderstanding of the available data on prison admissions, populations, and projections.   As prison admissions slowed and even declined in recent years, prison population projections that assumed no such stabilization began producing significant errors  in their  long-term  projections.  Over the  last decade,  several  states, often  in anticipation of working with the JRI, have made projections based on the assumption that prison admissions would not stabilize within three to five years of the projection being issued. It turns out that this was not the case. So prior to JRI being implemented in many states, the admission assumptions and projections were appropriately lowered. This means that many pre-JRI projections were no longer valid.

No criminologist or social scientist can legitimately claim that the difference between the pre- and post- reform numbers is being caused by the reform unless all of the other possible and external explanations and variables have been ruled out or controlled for in the analysis. Specifically, comparing a pre-reform estimate that had incorrect assumptions with a post-reform time period that had corrected the pre- reform assumptions has nothing to do with whether or not a reform’s claimed effect is or is not valid.

If the goal is to reduce mass incarceration, there is scant evidence of success. More alarming, there is little indication that historic rates of incarceration will be reduced in the future. A fair question to ask is what has been the impact of JRI thus far in reducing incarceration?

In assessing the outcomes of today’s JRI strategy, a good starting point is to compare the past trends of state that have been identified as JRI states with those that have not.   Figure 1 compares prison populations for the eight states that enacted JRI legislation prior to 2009 with those that have not.6)These 8 states are CT, RI, TX, KS, NV, AZ, PA, and NH.  Other various combinations of JRI and non-JRI states were tested with basically the same results. These eight states, as noted earlier, passed JRI legislation several years ago so we should be seeing some effects by now.  Other JRI states that have since passed legislation are included in the “Other” state category. Perhaps the primary finding is that for both groups and almost all of the states, there have been negligible, if any, reductions in prison populations. The non-JRI states began to show a slight decline beginning in 2009 while the early JRI states have not seen such a decline.

What Figure 1 masks, however, is the fact that only four major states have significantly reduced their prison populations—New York, New Jersey, Michigan, and California (Figure 2). The reductions were achieved by varying methods.   In New York, reductions in the number of felony arrests coupled with increases in non-prison sentences (which only occurred in New York City) have reduced the state’s prison population by nearly 25 percent.

In New Jersey, higher parole-grant rates brought on by litigation, sentencing reform of drug crimes, and reductions in parole revocations have reduced the prison population also by nearly 25 percent.  In Michigan, sentencing reforms as well as reductions in parole revocations and a higher parole grant rate have lowered the prison population by 17 percent.  And in California, beginning in 2007 with legislative efforts to increase the use of probation through financial incentives followed a federal court order to depopulate, the state prison population has declined by nearly 25,000 or 13 percent.

Only one of the early JRI states depicted in Figure 1 (Rhode Island) has shown a significant prison population reduction from a high of about 4,000 in 2006 to 3,200 in 2011. This reduction is largely due to legislation passed in 2008 as part of the JRI which increased the amount of good-time credits all individuals  could  receive  by  participating  in meaningful  programs.  There were  also adjustments to parole board decision-making and a general decline in prison admissions (non-JRI factors).



But  aside  from  these  five states, no others, whether JRI or not, have achieved such results.

Figure  3  shows  that  while prison admissions have dropped nationwide (since 2008) prison populations are down only slightly.  There are two reasons for this.  First, as depicted in Figure 4, while national state prison admissions are down, admissions have dropped dramatically in the four states described above.  Again, these same four states are doing almost all the work to drive admissions down.


Second, for all states there has been very little progress in reducing the average length of stay, especially for persons convicted of violent crimes, who occupy a majority of prison beds.  Unless all states begin to reduce the length of time people are required to serve, which can be accomplished as well for persons  serving  time  for violent offenses without compromising public safety, it will not be possible to significantly lower prison populations.

It can be argued that such a comparison is unfair to JRI states as they were targeting so called “high incarceration” states whose internal projections were suggesting rapid growth unless new policies were adopted.  In others words what would have been the impact had JRI legislation not passed. This is a fair question that is beyond the  scope  of  this  paper  and  should  be  answered  by  an  external  evaluation  like  the  NIJ  national evaluation now being conducted by the Urban Institute.  But there have been claims that the averted savings have been dramatic and due only to JRI and not to other external factors that were occurring both prior to and after JRI was implemented.7)One of the well-documented realities of prison forecasting is that forecasts are heavily driven by dynamic policy and legislative factors that are constantly changing and cannot be anticipated. This is why most states update their forecasts each year to adjust for the new policies and laws that have emerged. So using a pre-JRI baseline forecast to assess the impact of JRI is problematic.

For example, CSG has indicated that JRI has averted over 45,000 beds in prison population growth in 11 states through 2012.8)The 11 states are CT, KS, TX, AZ, RI, VT, MI, NH, NC, NV and OH.  Of the 11, four have shown actual reductions greater than 5 percent (RI, MI, NC, and NH) with RI having the largest drop, about 17 percent (or 800 inmates).

The largest portions of the 45,000 figure are caused by “averted growth” in four states that have not reduced their prison populations (Nevada, Kansas, Texas and Arizona). Averted prison population is based on projections that were done at the time legislation was passed which showed continued growth in the absence of JRI. The problems with such analyses are that they assume 1) the projections were valid and 2) there were no other factors post JRI causing the stabilizing trajectory.  Any credible claim that a policy or new law is causing a prison population to change must try to rule out competing explanations.

For example, as has been documented by Texas Legislative Budget Board (LBB), the stabilization in the Texas prison population was the result of two faulty assumptions by the LBB on prison admissions (too high) and parole grant rates (too low). The admissions assumptions were based on an historic linear growth rate that failed to account for more recent stabilization in admissions due to declining crime and arrest trends.  The LBB parole grant rate failed to account for the recent adoption of new parole guidelines that increased the parole board’s grant rate. Both of these developments were unrelated to the Texas JRI legislation. Once those two faulty assumptions were corrected by the LBB, its pre-JRI projection was no growth, suggesting that the JRI is far less responsible for averted growth in Texas than it has claimed.

In terms of measuring the impact of the JRI reforms, Texas is a particularly complicated example because of discrepancies between Texas data as reported by the Bureau of Justice Statistics (BJS) and that reported by the Texas Department of Criminal Justice (TDCJ). According to TDCJ’s fiscal year statistical reports, the incarcerated population grew by 2.4 percent between 2006 (the year before the JRI legislation was enacted) and 2011, while BJS reports just 0.1 percent growth during the same period. This is not a population reduction success story by either measure.

Similarly, the averted Arizona prison population was the result of using the high projection of new court commitments as opposed to the flat admission assumption. The pre-JRI projection completed by the JFA Institute noted that the prison admission stream in 2008 was stabilizing. If one used JFA’s flat prison admissions, the long-term forecast was significantly reduced. And that is exactly what has happened.  JRI has had a positive impact on probation violations and the amount of time probationer’s serve under direct supervision. But one must recognize that a large portion of the averted prison population in Arizona was related to factors that predated JRI that were not directly addressed by the JRI legislation.

CSG and Pew have argued that the JRI’s current work is more comprehensive and effective, and cite North Carolina as an example, claiming that North Carolina’s prison population is projected to decline by 3,600, or about 10 percent, over the next five years solely due to JRI.  This claim is based on a 2010 projection by the North Carolina Sentencing and Policy Advisory Commission (NCSPAC). But there are two reasons to doubt that Pew and CSG’s projection will be realized.

  • First, the 2010 projection assumed a linear increase in prison admissions based on the past few years’ activity, but in actuality, prison admissions peaked and stabilized in 2009, as did the prison population, which has been averaging about 39,000 to 40,000 since 2008.
  • Second, a portion of the projected reduction in the state prison population is due to relocating some 1,200 people convicted of misdemeanors to local jails, which will not have an impact on the state’s overall incarceration rate.

A later projection also done by the NCSPAC shows that the pre-JRI prison population estimates and the post-JRI estimates are virtually the same, which suggests the JRI has had little impact.  Even more troubling is that the JRI legislation is also expected to add nearly 2,000 people to the projected incarcerated population by increasing  the  lengths  of  sentences  for  breaking  and  entering  and lengthening the terms of parole supervision. Even a casual review of the official data would suggest that the impact of JRI is much less than Pew and CSG have advertised. The current prison population has stabilized at the 37,550 level and is projected to remain constant for the next ten years – thus institutionalizing the size of North Carolina’s incarcerated population.

One JRI state where there was averted prison growth was Nevada, which modified good time policies for prisoners and probationers and made the reforms retroactive for the current correctional populations.  But even here the averted growth was due to many factors, including reductions in crime, arrests and prison admissions—changes that were not a result of the JRI legislation.

The data shows that the JRI has not made significant progress in reducing prison admissions or length of stay. If its efforts had succeeded, there would be a significant reduction in prison populations, which has not occurred (nor is it projected to occur) in JRI states. One reason for this result is that most states have not succeeded in reducing sentences and lengths of stay for people convicted of violent or sex crimes, drug sales, and second/third felonies.9)In at least one state, significant changes have been made in this regard. New York prison admissions for drug sale offenses have declined markedly since 1999; the length of stay for this offense has also declined. And in 2009 the Rockefeller Drug Law reform eliminated mandatory prison sentences for second and third felony drug sales, as well as for non-violent felonies if the defendant is drug dependent and willing to seek treatment. It is insufficient to say that elected officials will not consider these changes without helping them understand that unless length of stay is addressed, prison populations will remain much as they are today.

Perhaps more troubling is the fact that prison populations where JRI policy is incorporated in the projections are expected to increase slightly over the next four years in 14 of the 18 JRI states, as illustrated in Figure 5.10)The 14 states where official agency forecasts were secured are NV, AZ, KS, OK, TX, OH, RI, NC, CT, KY, AR, LA, PA and GA.  While it may be argued that these projections would be even higher without the  intervention  of  the  JRI, it is  clear  that  current  efforts  will  not  lower  the  historically  high imprisonment rates in these jurisdictions. It also negates the claim that recent reforms that have been adopted in JRI will produce substantial declines in the near future.


The same dismal future trend can be expected for the non-JRI states, including states that have made the most progress to date. For example, California and Michigan are now projecting increases in their prison populations.  New Jersey and New York are seeing no further declines in their prison populations. This is not a healthy prognosis if the goal is to reduce the level of mass incarceration.

A More Effective, Inclusive and Sustainable Justice Reinvestment Strategy

A smart, principled Justice Reinvestment campaign to dramatically reduce overall   corrections populations is both timely and necessary.  The first step is to identify a set of clear policy goals that all states should consider adopting.  Then, incentives have to be structured in a manner that rewards justice and fair play.  Finally, effective strategies must be designed and tactics deployed to reverse the prime drivers of our bloated, expensive and excessively punishing corrections systems, including community reinvestment.

Clarifying the Goals of Justice Reinvestment

The following policy objectives have been put forth by reformers for decades but, with the exception of reducing admissions from probation and parole revocations, have generally not been adequately pursued. Of course, states and local governments may wish to pursue different strategies depending on their particular circumstances. But to reduce state and local correctional populations on a meaningful scale, we need to adopt some combination of the policy proposals set forth in this section.

Reduce admissions to prison for new convictions and re-entry into the system

States and local governments must consider additional strategies for reducing the number of people who enter the system. To reduce front-end and back-end entry into the correctional system, state and local governments should adopt policies designed to:

  • Reduce unnecessary arrests, especially for drug crimes. Research shows that police departments can safely reduce their felony arrest rates, which can have a dramatic impact on state correctional populations.
  • Eliminate unnecessary pretrial detention.  Nationally, approximately 60 percent of the jail population is awaiting trial; many charged with nonviolent offenses languish in jail simply because they cannot afford bail.
  • Reclassify some drug and other crimes, thus reducing the overall number of individuals housed in local jails and sentenced to prison.
  • Eliminate mandatory minimum sentences, which have a racially discriminatory impact and unnecessarily send thousands to prison every year who have been convicted of nonviolent and/or drug offenses and imposed sentences that are not proportional to the crimes committed.
  • Eliminate revocations to prison for violations of probation and parole conditions.
  • Require Racial Impact Statements, which can help reveal the hidden racial impacts of proposed changes to the criminal justice system.

Reduce length of stay in the system

Reducing length of stay in jail and prison and on probation and parole even by a small amount such as a few months would have a dramatic impact on the size of a state’s overall correctional population.  For example, the current average length of stay reported by the Bureau of Justice Statistics is 29 months.  If we returned to the earlier LOS of 21 months that existed for many years, the nation’s state prison population would decline by over 400,000.11)James Austin, “Reducing America’s Correctional Populations: A Strategic Plan” (2010). Justice Research and Policy, Vol. 12, No.1, pp. 1-32. PEW recently released a study showing that this increase in prison length of stay was costing the states about $10 billion per year.12)Time Served: High Cost Low Return of Longer Prison Terms, June 6, 2012. Washington, DC: Pew Center on the States. But to reduce length of stay, state and local governments should adopt policies that:

  • Acknowledge the  unnecessarily  harsh  nature  of  our  penal  systems  by  undertaking  major sentencing reforms, such as:
    • Reducing sentence lengths overall to be more fair and proportionate to the offense (including, for example, reclassifying some drug and other crimes and eliminating mandatory minimum sentences, as discussed above);
    •  Eliminating or reforming habitual offender laws;
    • Prohibiting life sentences for all but the most serious crimes; and,
    • In jurisdictions with indeterminate sentencing systems, presumptive parole for low risk individuals who have served their minimum prison terms.
  • Increase eligibility  for  earned  discharge  from  probation  and  parole,  as  well  as  work  and education release, so that individuals who succeed under supervision can be released from their correctional obligations and move forward with their lives.
  • Develop mechanisms for back-end release from jail and prison, including earned eligibility for good time, merit time programs, and increased eligibility for parole in states that retain it.

Many states (both JRI and non-JRI) states have implemented some of these reforms.  The problem is that, with a few exceptions, they are not sufficiently aggressive to produce the significant declines in correctional populations.

Incentivizing Decarceration

As discussed above, one of the chief drivers of excessively punitive policies is the system of incentives that encourage officials at all levels of government to pursue punitive policies and practices.  Pernicious incentives and financial interests permeate the criminal justice system.  Funding allocations make state imprisonment cost-free to local jurisdictions but require those jurisdictions to pay for placement into treatment or probation; these kinds of anomalous government  funding streams that  incentivize localities to push people upstream to state prisons instead of deploying more effective local strategies must be changed.  Similarly, effective campaigns must be mounted to challenge federal grant programs that reward unnecessarily aggressive policing, and to counter the bail bond industry’s opposition to pretrial bail reform and the private correctional corporations that lobby for excessively punitive penal and immigration policies, as well as pressures on courts to speed through dockets to the detriment of defendants’ due process rights.

These problems have to be addressed if we are ever going to move away from a system of disproportionately severe punishments.  Performance metrics aimed at ending mass incarceration must be adopted system-wide.  Specifically, federal grants (as well as state and local) should encourage practices that improve public safety and discourage high rates of disproportionately minority arrests for low-level, nonviolent crimes.  Prosecutors should be rewarded for filing charges that reflect the merits of the case, not charges intended to extract guilty pleas.  Probation and parole officers should be incentivized to help clients successfully complete their terms of supervision rather than making seemingly risk-averse recommendations for revocation.   And counties should be compensated for creating programs and policies that retain individuals locally and charged for committing them to state prison.

What It Will Take: A New Strategy for a Revamped Justice Reinvestment

If we are to take advantage of the current climate for progressive change, we must move quickly, with determination and creativity.  To achieve the goal of dramatically reducing the U.S. correctional population requires a substantial commitment to develop and sustain the capacity of key players to do the work.  Furthering this necessary and ambitious agenda will require development of different sectors—both in-state and nationally—that can create sustained demand at the local and state levels for far-reaching sentencing and corrections reform and smart investments in local communities.

A key aspect of our proposed strategy is to focus on involvement of local municipal and county officials, as well as local system reform advocates, to press in coalition for reforms that will actually reduce the “local concentrations” of mass incarceration that prevail in the communities to which they are politically accountable.  By virtue of their accountability to local constituents, local justice reinvestment coalitions would not just add to the demand for state legislative reforms, but they would be in a position to share the political risks entailed in legislating sentencing reforms and other difficult-to-attain legislation.  And local officials have the authority to implement local policy changes that could result in major reductions in statewide correctional populations.

Some may suggest that the authors of this paper simply ignore the political realities of state sentencing and corrections reform.  However, no honest accounting of community-level public safety can fail to register the debilitating impact of high-scale prison migration.  And we know from experience that success is possible.

In West Virginia in 2012, for example, a strong coalition of advocacy organizations developed a comprehensive plan for overhauling the state’s bloated corrections system.  S.B. 342, the Public Safety Offender Accountability Act, would have made significant front-end changes to West Virginia’s corrections systems, such as drug law reform (including the reclassification of certain manufacturing and distribution crimes to misdemeanors), mandatory citation in lieu of arrest for most misdemeanors, and presumptive pretrial release for presumptive probation offenses.   The state legislature was under pressure to pass this set of reforms because it had been defying three court orders mandating that it address the severe overcrowding of its jails, which also hold state-sentenced individuals.   S.B. 342 passed in the Senate unanimously and had a majority of support in the House Judiciary and Finance Committees. It lost a few votes by the time it got to the House floor last spring and was eventually pulled.  Still, the bill was thoroughly debated and widely supported, and advocates and litigators were keeping the pressure on by threatening to file contempt motions in order to ensure that the legislature passed the 2012 bill in 2013.  Then JRI entered West Virginia late in 2012 and initiated its own process, which had the effect of causing the state to jettison the earlier 2012 bill.  The JRI’s 2013 recommendations omit all of the 2012 proposed front-end reforms, and focus instead on post-release supervision and substance abuse treatment.  As explained above, these goals may be laudable ones but they cannot, on their own, achieve meaningful reductions in the state’s correctional population. In fact, if the 2013 recommendations are passed in toto, the state’s prison population will be about the same in 2018 (7,418) as it was in 2012 (7,531).

Major reform has taken place in New York, where decades of advocacy and activism in New York City aimed to reform the state’s drug laws, reduce drug enforcement efforts by police and prosecutors, and shift city drug policies away from law enforcement and toward a public health response.  Advocacy for change was intensified through broad-based drug reform campaigns coordinated by two New York City (NYC)-based advocacy campaigns: “Drop the Rock” and “Real Reform.”  In 1998, felony drug arrests in NYC had hit a high of 45,978.  The campaigns against the Rockefeller Drug Laws helped to foster a major shift in public opinion about drug enforcement.   In 1999, a widely-publicized poll of New York State (NYS) voters conducted by pollsters at Zogby International revealed that twice as many said they would be more inclined to vote for state legislators who would reduce sentences and give judges greater discretion in drug cases than the number who said they’d be less inclined to support them.  That same year, New York Police Department enforcement priorities shifted, and felony drug arrests began a sharp decline.  By 2011, the number had plummeted to 21,149, and felony drug filings had also declined. Accordingly, prison sentences for felony drug offenses meted out by NYC judges fell from 8,614 in 1998 to 2,224 in 2011.  And between 1999 and 2012, the NYS prison population fell by nearly 25 percent.  The steep decline in felony drug arrests in NYC contributed greatly to a decade-long decline in the NYS prison population, and finally—in 2009—state lawmakers voted to reform the harsh Rockefeller Drug Laws.

Yet another example of effective state-level reforms resulting from local and state-based coalition building comes from Colorado, where the adult prison population has been declining (down seven percent since 2010) after decades of unabated growth.  The Colorado Criminal Justice Reform Coalition (CCJRC), founded in 1999 to reverse the trend of mass incarceration in Colorado, has had substantial influence in spurring this decline.  To accomplish its goals, CCJRC conducts public education, research, policy analysis, community organizing, coalition building, litigation, and legislative campaigns.  Since its founding, CCJRC has become one of the most effective state-based advocacy organizations in the U.S. The small staff has built a vibrant statewide, movement for criminal justice reform.  Today, CCJRC includes a network of over 112 organizations and faith communities and over 6,800 individual members throughout the state; it has intentionally built a diverse, bipartisan coalition that includes criminal justice reform organizations, educators, students, prisoners, former prisoners, prisoners’ family members, attorneys, mental health advocates, substance abuse treatment providers, racial justice advocates, victims’ advocates, child welfare professionals, various faith communities, fiscal conservatives, civil libertarians, business owners, and women’s organizations.  Through involvement with the state’s Commission on Criminal and Juvenile Justice, many of CCJRC’s reform goals have gained the traction needed to enact important legislative victories, including many of the reforms proposed in this paper, such as establishing probation eligibility for people with two or more felony convictions, enacting drug sentencing reform, excluding felony drug possession as a trigger for habitual offender sentencing, reducing new admissions from parole revocations, and increasing earned good time.

Early on, Kansas succeeded in passing JRI legislation that specifically targeted northeast central Wichita for reinvestment because it had the highest concentration of incarcerated and returning residents in the state. Compared to other JRI states, Kansas perhaps made the most headway toward implementing two key Justice Reinvestment features which have been absent elsewhere: organizing local demand and engaging development experts.   Meeting regularly for about a year, the New Communities Initiative (NCI), formed with the help of CSG and Urban Strategies (a St. Louis-based non-profit), was composed of city officials and representatives from a local hospital, Wichita State University, banks and local foundations (including for a while the Koch Foundation); admittedly absent were local residents and advocates. Having identified five areas in desperate need of reinvestment and revitalization,  NCI decided to tackle the demand for affordable housing because of the large number of tax delinquent and abandoned properties and otherwise poor and inadequate housing stock.   Vacant lots stood where there had once been grocery stores, hairdressers, cafes and hardware stores—victims of the economy, lack of opportunity, and resistance to desegregation, and mass incarceration which had turned a once vibrant working/middle class community into something of a wasteland.  The City issued an RFP for development of a housing master plan; a grant was ultimately awarded to McCormack, Baron and Salazar, whose principals had helped design the federal HOPE 6 grant program.  Their plan for new and in-fill housing development was considered likely to receive millions in federal support from HUD. Unfortunately, the Wichita reinvestment project has stalled for a number of unforeseeable reasons; nonetheless, the Kansas example (which followed the original Justice Reinvestment concept), demonstrates that it is possible to undertake reforms pursuant to the strategy we propose here. (See Appendix A for more details about new JRI efforts in Kansas, which appear not to include reinvestment in Wichita.)

We do not claim that these states have been able to resolve all of their criminal justice problems; we simply provide these examples to demonstrate that many lawmakers have the political will to make major reforms.  But it takes the work of local constituents to realize these gains. For example, Louisiana repealed  mandatory minimum  sentences  for  simple  drug  possession  and  many  other  non-violent offenses in 2001 (and cut minimum sentences for drug trafficking in half) long before the JRI entered the state.  Michigan has steadily been cutting mandatory sentencing laws from its books for the last decade, without the help of the JRI.

We propose a series of focused campaigns in selected jurisdictions consisting of cross-disciplinary, multi- sector coalitions of players who can seize the moment to challenge the status quo and replace it with a system that is fair, just, and equitable.  These campaigns should be able to identify the drivers of their correctional populations, the scope and extent of their local concentration, the policy mechanisms needed to significantly reduce correctional populations, and the points where the necessary political pressure can be applied.   Reinvestment in high incarceration communities should be an essential feature.13)Where cost savings from corrections reforms are scarce, funds from other DOJ and ONDCP grant programs (e.g. Byrne Criminal Justice Innovation Initiative, Community-based Violence Prevention Initiative, Byrne Justice Assistance Grants, High Intensity Drug Trafficking Areas [HIDTA] Program) and the DOJ Asset Forfeiture Program could be redirected to or reinvested in high incarceration communities.  Campaigns will also have to include sophisticated, well-planned, and effectively executed public education campaigns. We acknowledge that there are many ways to structure such campaigns and we do not claim that a one-size-fits-all model will work; nonetheless, we offer the following model as an example of what such campaigns might look like.

Imagine, for example, the power of coalitions composed of the following:

  • Locally based grassroots organizations to marshal demand for Justice Reinvestment and keep the ethical issues of justice and equality at the forefront of the struggle;
  • Grass-tops leaders and in-state advocacy organizations with the political sophistication in state policy matters to engage state policymakers, litigate strategically, and mobilize community groups for change;
  • State lawmakers who are willing and able to share the political risks associated with being part of such a coalition;
  • Local lawmakers who have the political guts to join the effort, share the risks, and use their legislative savvy and clout to move this agenda and ensure reinvestment into their high- incarceration communities, as well as authority to implement local policy changes that can impact statewide correctional populations; and
  • National organizations  that  can  support  the efforts of in-state campaigns and local communities by providing technical assistance and strategic advice based on experiences in other states, urging and crafting deeper reforms using examples from other states, recommending third party “validators” from other states that enacted reforms and kept crime rates down, and ensuring that state-based efforts are part of a national movement to end mass incarceration.

Outside of working toward state legislative reform, a key strategy to tackle, for example, mandatory minimum and truth-in-sentencing legislation, we could also create a small number of Community-Based JRI (CB-JRI) demonstration projects to work intensely over a five-year period to help design and implement a different JRI strategy aimed at local reinvestment and policy change. Such a demonstration initiative would focus on high incarceration communities within a jurisdiction rather than state level agencies and prison rates, though the initiative would hopefully help create demand for state legislative reform and potentially impact state incarceration rates.

These national/state/regional/local coalitions should work with major researchers and policy analysts to identify the drivers of their corrections populations, the scope and extent of their local concentration, and the health, education, and employment indicators that prevail in those communities so as to make the data-driven case for   important but challenging criminal justice reforms that incorporate reinvestment logic from the beginning.  These policy narratives should be coordinated by communications professionals, who can design and build effective public education strategies. Additionally, a concerted strategy should be developed to partner with multi-sector economic and community development entities that have the credibility and expertise to help identify and forge reinvestment opportunities.

These kinds of reform campaigns are not won overnight.  They involve a combination of long-term cultural transformation and imminent policy opportunities.   And funding will likely be a challenge for such coalitions. To meet this need, interested foundations could form a funding collaborative to develop grant programs, issue guidelines, and invite proposals for significant multi-year funding.  Funders might want to develop criteria for and identify promising jurisdictions for reform, perhaps with the help of leading criminal justice reform advocates, researchers, and analysts, as well as experts from the community investment, organizing and communications field, who might also participate in designing the grant program.  Proposal guidelines should be based on a well-articulated theory of change, and balance prescriptive strategies and goals with room for creativity and regional differences.  Interested coalitions could be invited to submit initial inquiries, requests for planning grants, and/or full proposals. Prospective applicants would be challenged to form effective coalitions of appropriate but diverse stakeholders.


For the first time in several decades, public dialogue seems to be reflecting genuinely progressive principles and values with regard to reducing mass incarceration.  This might be an once-in-a-lifetime opportunity to make sure that people mired in the injustices of the U.S. punishment system are not left out, forced to the back of the line again, and told to wait until all the other big problems facing the world have been dealt with.  In fact, this may be the perfect “teachable moment” to raise our collective consciousness (and conscience) and make clear how the progressive agenda cannot be achieved without dismantling the U.S. punishment system. The reasoning is clear: the combination of excessive incarceration and harsh punishment is a blunt instrument for social control that perpetuates the country’s painful, historical legacy of injustice and inequality, and deprives masses of black and brown people unfairly of freedom and opportunity.  It is the site of today’s civil rights struggle.  So while Justice Reinvestment may not be the answer, it could perhaps be a good instrument for righting some wrongs and setting some people free.  But to do so, it needs a new orientation, and a new future.

Appendix A: Case Studies in the Early History of the Justice reinvestment Initiative

Texas: In May 2007, the Texas legislature enacted a package of criminal justice legislation that many policy makers consider to be the most expansive redirection in state correctional policy since the early 1990s.  The new policies focused on an expansion of treatment and diversion programs and enhancement of parole and probation policies.14)The law also authorized bond funding for the construction of three new prisons, but only if the state failed to implement the new policies and programs effectively and the construction was deemed “necessary.”  Proponents argued that these reforms alone would eliminate prison growth and did  not push for the types of changes needed to make deeper cuts. Choosing a strategy that focused on treatment, parole, and probation ultimately proved to be ineffective.  Though the reform legislation mandated a lower parole caseload, it contained no provision for increasing parole rates.  The projected savings from JRI’s proposed expanded treatment program relied on previous prison population projections by the Legislative Budget Board.   However, those previous projections turned out to be false.  Once the Board adjusted its projections, it became clear that the JRI treatment proposals were unlikely to generate any savings.

In fact, prison populations did stabilize after the 2007 law was passed.  JRI principals argued that this stabilization was due to the recently enacted treatment program.  However, those arguments ignored the fact that the Texas Board of Pardons and Paroles began to grant parole at a higher rate, and admissions to prison were decreasing.   Neither of those developments was attributable to the JRI treatment program.

In the last several years, Texas’s prison population has risen from 171,790 at the end of 2007 to 172,224 at the end of 2011, and is projected to increase further.  The JRI trumpets Texas’s “success,” and the Texas reforms were a success in one sense: Texas is one of our toughest-on-crime states, so any progress on criminal justice reform is an accomplishment.  However, if the metric is reduced to corrections populations and costs, the Texas JRI program must be viewed as a failure.

Connecticut:  In Connecticut, the JRI was intended to address prison overcrowding through a collaborative effort between key state agencies—the Departments of Correction; Labor; and Mental Health and Addiction Services; the Board of Pardons and Paroles; and the Judicial Branch—to develop and implement a comprehensive reentry strategy for people released from prison, as well as new efforts to cut down on the number of people who are sent to prison for technical violations of both probation and parole.  To that end, CSG submitted “Building Bridges:  A Proposal to Reinvest Corrections Savings in an Employment Initiative,” to legislators in 2003.  It contained a series of maps created by the Justice Mapping Center as a vivid display of the Million Dollar Blocks, high-incarceration neighborhoods where targeting of community-based programs offering transitional employment, drug treatment and mental health services could allow non-violent individuals to be paroled more quickly without jeopardizing public safety.

“Building Bridges” presented three options for leveraging correctional savings to expand employment opportunities in the communities that had been identified:

  • Invest in community development financial institutions that would reinvest the dollars in small businesses, job creation and general community development targeted to low-income neighborhoods and/or criminal justice populations;
  • Match savings with funds available through federal “pass through” grant programs such as the Temporary Assistance for Needy Families (TANF) and the Workforce Investment Act (WIA); or
  • Capitalize on tax incentives established to assist employers in hiring people without jobs or receiving public assistance.

In 2004, the Connecticut legislature appropriated $13.4 million to provide expanded supervision and program services, including contracts for new residential and drug treatment beds and expanded “alternatives to incarceration” services.  New staff positions were created for probation and parole officers.  In keeping with the original JRI concept, $1 million was earmarked for creation of pilot projects in New Haven and Hartford to aid re-entry for parolees.  But important considerations about which kinds of investments might best improve the circumstances of people returning to the neighborhoods so vividly mapped in “Building Bridges” received scant attention.  Rather than concentrating the modest resources provided for employment initiatives within a single neighborhood in each city to which people might return from any of the 20 state prisons, the pilot projects were designed to provide transitional housing for a few hundred people released to any part of the target cities from a prison nearby.

Kansas:  The JRI entered Kansas in 2006 to deal with the state’s growing prison population, high failure rates for people on supervision, and high rates of release of people who had not completed risk- reduction programming.15)See Tough and Smart: Opportunities for Kansas Policymakers to Reduce Crime and Spending, Council of State Governments, Justice Center, (last visited November 13, 2012).  Kansas policymakers moved more aggressively to embrace the original JRI philosophy when they  adopted the most  ambitious  Justice Reinvestment projects ever undertaken under the JRI.  The 2007 legislation provided $4 million dollars to strengthen and expand capacity at local community corrections programs, with the proviso that programs receiving the additional funding had to reduce their revocations by 20 percent within two years.  The community corrections programs actually achieved the required reduction goal within one year and exceeded it in the second year.

At the same time, policymakers understood that a lasting reduction in crime and recidivism rates would depend on efforts to revitalize the neighborhoods where people had been living when they committed the crimes that sent them to prison, as well as on the provision of substance abuse, mental health, employment, and housing in the communities to which they would return.  Aided by mapping analysis, they were able to pinpoint the multiple problems that gave northeast central Wichita (Council District 1) the highest incarceration rate in the state, accounting for $11.4 million in spending for its prison commitments over the course of a single year.  People from District 1 were using more than twice the number of prison beds as any other Wichita council district.

State agency officials began working with staff at local city agencies on a comprehensive plan for neighborhood revitalization.  Members of the state legislature collaborated with city council members, and a community advisory committee, which included representatives of the local housing and police departments, along with people from the faith community. The heart of the revitalization plan—“The New Communities  Initiative”—was  a  neighborhood-based  housing development  project that  would have targeted the district’s hundreds of abandoned and tax-delinquent houses and blighted properties and garnered significant support from HUD.  Several banks, hospitals, private foundations, schools, universities and municipal officials joined The New Communities Initiative, which was short, however, on local residents.

The initial results were positive: Kansas’s prison population declined three percent from 2007 to 2009. But after the 2008 fiscal crisis hit the state, the statewide community corrections system was hit with deep budget cuts, and plans for revitalization in Wichita were shelved.  Since 2009, the prison population has grown a troubling eight percent.  Males are once again above the state’s prison capacity and the projected growth over the next few years is significant.  In order to addresses these alarming developments,  the JRI returned to Kansas and has identified  a  number of trends: new court commitments have been rising (up 37 percent from 2006-2011); arrest rates have been steadily rising for violent, property, and drug crimes; prison sentences are up nine percent since 2009; probation revocations decreased from 2006-2009 but have been increasing since 2009; and fewer people are receiving programming, both in prison and in the community.

Arizona:  In Arizona, where the high rate of failure among people on community supervision had been a primary driving factor behind prison growth, the legislature enacted the “Safe Communities Act” in 2008 to create incentives for success for those sentenced to county-based probation supervision. Probationers were made eligible to have their supervision term reduced by 20 days for each month of compliance with probation conditions, performance of community service, and payment of restitution to victims. Any county probation agency that saw a reduction in recidivism and revocations was to receive 40 percent of the prison bed savings to provide greater access to drug treatment and training programs, and to expand services to victims of crime.

The “Safe Communities Act” appeared to help speed progress toward adoption of “evidence-based” practices—validated risk-assessment tools and less punitive supervision strategies—for probation supervision, an effort already undertaken by the state’s Administrative Office of the Courts. Compared to 2008, by 2011 revocations to prison fell by 39 percent.   During the same period, the rate of new felony convictions among people on probation fell by 42 percent.  Facing a huge budget crisis, however, Arizona legislators did not make any appropriations of the savings available for the purposes set forth in the “Safe Communities Act.”  The Act itself was repealed in 2011, but the provision that allows a reduction in probation supervision terms according to performance remains.

Rhode Island: Rhode Island is another example where significant reductions in the prison population have been achieved without jeopardizing recidivism rates and overall crime rates.  This was largely achieved by the work of CSG and JFA Institute that identified an antiquated good time policy that provided no incentives for   individuals serving longer sentences to participate in meaningful rehabilitative programs. Legislation adopted this system that served to reduce length of stay.

Appendix B: Reducing the Juvenile Justice Population

We need not look to other countries or other times in our nation’s history to find a workable model for this level of de-incarceration – one is occurring right under our noses, in our country, at the present time in America’s juvenile justice system.

As recently as the 1990s, reducing juvenile incarceration rates was unthinkable.  For example, in Body Count (1996), co-authored with leading conservative thinkers William J. Bennett and John P. Walters, John J. DiIulio, Jr. wrote “America is now home to thickening ranks of juvenile ‘super predators’ — radically impulsive, brutally remorseless youngsters, including ever more preteen age boys, who murder, assault, rape, rob, burglarize, deal deadly drugs, join gun-toting gangs and create serious communal disorders.”

With the advent of the super predator era, the juvenile incarceration race was on.  All but one state passed laws during the 1990s making it easier to try juveniles as adults, reducing confidentiality protections or both.  Taking a leaf out of Dr. DiIulio’s book, Congress proposed the “Violent Youth Predator Act of 1996” which would have allowed 13-year-olds to be housed in adult jails.   America exited the 1990s as one of only a handful of nations that allowed the death penalty for juveniles.16)The Supreme Court abolished the juvenile death penalty in 2005 in Roper v. Simmons.

But recently, there has been a dramatic change in the handling of juveniles who run afoul of the law, one that had begun brewing even during the most punitive days of the 1990s.  The decline in the number of juveniles behind bars has been deep and broad.  It has affected red states and blue states in rural and urban areas, at a city, county and state level.  And it has generally occurred without an increase in youth crime.

Between 2001 and 2010, there was a 33 percent decline in the number of youth “placed” post- adjudication (the juvenile justice system’s euphemism for sentenced to prison) in America’s juvenile facilities, with declines in 43 states (see map showing reductions in juvenile placements, attached). A major reason for the decline is that juveniles cannot receive long and excessive prison sentences unless they are sentenced as adults. Consequently, the length of stay cannot be extended like it can for the adult system.

Indeed, some of the nation’s largest states, counties and cities have become very creative at not only reducing youth incarceration, but also assuring that the funds freed up from the incarceration decline partially follow the young people from their cells back to their communities:

California reduced the number of young people locked up in its state system from approximately 9,572 in 1996 to 1,082 at the end of 2011.  Ninety-nine percent of adjudicated youth in California are now housed or supervised by counties, with county costs defrayed by $93.4 million in state funds last year alone.17)California Legislative Analyst’s Office: The 2012-13 Budget: Completing Juvenile Justice Realignment.  Changes in California’s youth crime rates support the contention that the massive downsizing of state juvenile corrections there has not come at the expense of public safety.  According to the California Department of Justice, while the incarcerated state youth population in California declined by 84 percent from 1996 to 2008, the juvenile arrest rate declined by 32 percent.  Meanwhile, as the adult prison population was increasing by 21 percent during that same time, the adult arrest rate declined by a more modest 15 percent.  Buoyed by these kinds of data, California Governor Jerry Brown has suggested completely shuttering the state’s juvenile justice agency and transferring its authority to the counties.  To assure the new system would be adequately funded, he has proposed changing the state’s constitution.

Between 2002 and 2011, New York City reduced institutional placements of delinquent youth by 62 percent, while experiencing a 31 percent decline in major felony arrests of juveniles.  In 2012, Governor Andrew M. Cuomo signed the “Close to Home” bill into law that, by the end of 2013, will transfer jurisdiction of almost all youth sentenced in Family Court from the state to the city.   Close to Home passed with bipartisan support.  The city has already issued contracts to non-profit organizations to operate 35 small facilities (of 24 or fewer beds each) for the youth being transferred to the city, and the city is expanding the number of alternative to placement programs to further reduce the number of incarcerated youth.

In Texas, the number of youth incarcerated in the Texas Juvenile Justice Department (formerly known as the Texas Youth Commission) declined by over 63 percent between 2006 and 2011.18)National Juvenile Justice Network: Advances in Juvenile Justice Reform 2009-2011 (July 2012). After a highly publicized sex scandal involving TYC staff and youth created a climate for reform, the legislature passed bills precluding misdemeanants from being sentenced to TYC and reducing lengths of stay for youth convicted of felonies, reallocating $58 million to counties to work with these youth, about half of what it would have cost to incarcerate them.  When lawmakers were able to close two remote facilities, saving the state $115 million, probation departments in Texas were provided $45.7 million to operate proven diversion programs in exchange for agreeing to reduce their targets for commitments to TYC.19)Right on Crime.

In Michigan, Wayne County (Detroit) had 731 youth confined in state facilities in 1998; last year it had four, with about 185 in secure housing in facilities within the county.  State and county officials agreed to realign care and funds from the state to the county.  County officials then contracted with five Care Management Organizations (CMOs), which are similar to HMOs.  The CMOs receive a block grant for the delinquent youth in their catchment area, which incentivizes them to place youth in successful and cost- effective programs located in their home communities.  Juveniles living in their home communities have less than a one percent felony conviction rate during active enrollment with a CMO.20)Douglas N. Evans, John Jay College of Criminal Justice: “Pioneers of Youth Justice Reform: Achieving System Change Using Resolution, Reinvestment and Realignment Strategies.”

In 1991, Ohio was home to four of the 20 most overcrowded juvenile facilities in the nation.  In 1994, to address overcrowding and the violence that went along with it, the state launched RECLAIM Ohio. RECLAIM carefully uses fiscal policy to incentivize county juvenile justice innovations.  If counties reduce state juvenile placements in a given year, they earn more money the following year.  RECLAIM was piloted in nine counties in 1994.  Pilot counties quickly reduced state commitments by 42 percent, while commitments from the non-pilot counties actually increased.  Most significantly, low-level felony commitments  from  pilot  counties  dropped  precipitously.  In 1995, RECLAIM Ohio was launched statewide to include the remaining 79 counties.  By December 2011, the number of RECLAIM-funded programs initiated statewide topped 600, while the number of youth sent to state facilities has dropped to approximately 650 (from a high of more than 2,600 in 1992).21)Ohio Department of Youth Services: “What is Reclaim Ohio?”

In 1994, the Annie E. Casey Foundation launched the ambitious Juvenile Detention Alternatives Initiative (JDAI) in five pilot sites nationwide to safely reduce the number of youth detained “pre-adjudication” (i.e. pre-trial).  JDAI is now operating in over 180 jurisdictions in 39 states and the District of Columbia housing 40 percent of the youth detained nationwide.  As JDAI sites have made more parsimonious use of pretrial detention, their commitments to state facilities have dropped commensurately.

Cook County (Chicago), Illinois, one of JDAI’s model sites, reduced state training school commitments by 45 percent between 1997 and 2010.  Model site Santa Cruz, California, which historically committed few youth to state custody sent only two youth in 2010, while far fewer youth (52) went to any out-of- home placement (down from 139 in 1996).  JDAI sites in Virginia reduced commitments by 59 percent, while Alabama’s four JDAI sites reduced commitments from 1,127 to 495 or 56 percent since 2007. Overall, across all JDAI sites, 4,237 fewer youth were committed to state placements, representing a 38 percent drop when compared to pre-JDAI implementation.22)Annie E. Casey Foundation, JDAI 2010 Annual Results Report Summary.

These significant declines in youth incarceration, affecting the vast majority of states in the U.S., are as deep as they are broad.  While they have not necessarily occurred perfectly, and while issues affecting juveniles are somewhat different than those that affect adults, they offer both a political and policy lesson as policy makers struggle to substantially reduce adult prison populations and realign dollars that now fund prisons to communities that are disproportionately affected by the cycle of crime and imprisonment.

43 of 50 States Reduced Placement Rates Between 2001-2010



41 States Experienced a Decrease of More than 10% in Placements Between 2001 and 2010


State With +-3% none are in the “no” change group
Percent change 2001-2010 Map codes
Rates Rates Rates
Total Public Private Total +- 3 Total+- 10%
United States -33% -34% -30%
Alabama -33% -34% -30%  1  1
Alaska -12% -2% -39%  1  1
Arizona -50% -48% -61%  1  1
Arkansas 11% 26% -61%  3  3
California -38% -39% -36%  1  1
Colorado -18% -6% -31%  1  1
Connecticut -57% -47% -65%  1  1
Delaware -21% -14% -41%  1  1
District of Columbia 22% 50% -5%  3  3
Florida -34% -44% -26%  1  1
Georgia -35% -31% -46%  1  1
Hawaii 17% -34% -50%  3  3
Idaho -17% -11% -47%  1  1
Illinois -36% -33% -57%  1  1
Indiana -39% -41% -35%  1  1
Iowa -30% -34% -28%  1  1
Kansas -22% -18% -34%  1  1
Kentucky -15% -2% -63%  1  1
Louisiana -53% -57% -35%  1  1
Maine -8% 13% -84%  1  2
Maryland -25% 16% -66%  1  1
Massachusetts -47% -45% -50%  1  1
Michigan -37% -25% -48%  1  1
Minnesota -51% -36% -66%  1  1
Mississippi -47% -51% 200%  1  1
Missouri -10% -11% 17%  1  2
Montana -20% -1% -47%  1  1
Nebraska 10% -16% 70%  3  2
Nevada -38% -36% -62%  1  1
New Hampshire -37% -50% -15%  1  1
New Jersey -44% -45% -33%  1  1
New Mexico -30% -36% 28%  1  1
New York 37% -48% -23%  1  1
North Carolina -42% -44% -39%  1  1
North Dakota 9% -2% 21%  3  2
Ohio -34% -31% -58%  1  1
Oklahoma -26% -10% -51%  1  1
Oregon -17% -21% 9%  1  1
Pennsylvania 7% -14% 17%  3  2
Rhode Island -14% -42% 25%  1  1
South Carolina -31% -37% -13%  1  1
South Dakota 12% -19% 70%  3  13
Tennessee -55% -49% -68%  1  1
Texas -44% -37% -76%  1  1
Utah -40% -31% -48%  1  1
Vermont -39% -41% -37%  1  1
Virginia -37% -32% -96%  1  1
Washington -38% -34% -81%  1  1
West Virginia 24% 66% -12%  3  3
Wisconsin -39% -43% 31%  1  1
Wyoming -16% -34% 5%  1  1


Related Posts
July 20, 2022

Comment on Home Confinement Under the Coronavirus Aid, Relief, and Economic Security (CARES) Act

The Sentencing Project's Executive Director Amy Fettig submitted comments to the Office of the Attorney General on behalf of The Sentencing project regarding the United States Department of Justice’s proposed rule on CARES Act Home Confinement.
July 27, 2022

Comments to the Food and Drug Administration on the Proposed Ban of Menthol Cigarettes

The Sentencing Project submitted comments to the U.S. Food and Drug Administration on the proposed Tobacco Product Standard for Menthol in Cigarettes.