You are here

Changing Priorities: State Criminal Justice Reforms and Investments in Education

October 28, 2014

Most states’ prison populations are at historic highs after decades of extraordinary growth; in 36 states, the prison population has more than tripled as a share of the state population since 1978.  This rapid growth, which continued even after crime rates fell substantially in the 1990s, has been costly.  Corrections spending is now the third-largest category of spending in most states, behind education and health care.  If states were still spending on corrections what they spent in the mid-1980s, adjusted for inflation, they would have about $28 billion more each year that they could choose to spend on more productive investments or a mix of investments and tax reductions.

Even as states spend more on corrections, they are underinvesting in educating children and young adults, especially those in high-poverty neighborhoods.  At least 30 states are providing less general funding per student this year for K-12 schools than before the recession, after adjusting for inflation; in 14 states the reduction exceeds 10 percent.  Higher education cuts have been even deeper:  the average state has cut higher education funding per student by 23 percent since the recession hit, after adjusting for inflation.  Eleven states spent more of their general funds on corrections than on higher education in 2013.  And some of the states with the biggest education cuts in recent years also have among the nation’s highest incarceration rates.

This is not sound policy.  State economies would be much stronger over time if states invested more in education and other areas that can boost long-term economic growth and less in maintaining extremely high prison populations.  The economic health of many low-income neighborhoods, which face disproportionately high incarceration rates, could particularly improve if states reordered their spending in such a way.  States could use the freed-up funds in a number of ways, such as expanding access to high-quality preschool, reducing class sizes in high-poverty schools, and revising state funding formulas to invest more in high-poverty neighborhoods.

State incarceration rates have risen primarily because states are sending a much larger share of offenders to prison and keeping them there longer.  States can reduce their incarceration rates – without harming public safety – by reclassifying low-level felonies to misdemeanors where appropriate, expanding the use of alternatives to prison (such as fines and victim restitution), shortening jail and prison terms, and eliminating prison sentences for technical violations of parole/probation where no new crime has been committed. 

A number of states have enacted criminal justice reforms in recent years.  Some have reduced prison populations sharply; reforms in New Jersey, New York, and California for example, helped drive down prison populations in each of those states by roughly 25 percent – while crime rates have continued to fall.[1]  In most states, though, reforms have not had a large impact on the size of prison populations, which remain extremely high nationally.  Moreover, states rarely have directed the savings from reform explicitly to human capital investments (such as education) or low-income neighborhoods.

States wishing to use the savings from criminal justice reforms for productive purposes would do well to adopt planning and budgeting mechanisms that can help them shift priorities. These include:

  • A high-quality, long-term forecast of the savings from specific reforms, made available to lawmakers when they are considering reform bills;
  • An accepted process to estimate the annual savings from the reforms once enacted;
  • An established mechanism to shift those savings to productive uses, especially human capital investments; and
  • An independent commission to monitor implementation and enforce compliance with the reforms.

This is not to say that states can use criminal justice reforms to fully finance the increased education investments they need.  First, there is a timing issue; major savings from reducing incarceration likely will accrue over a number of years, as reforms lead to prison closures and a reduction in the prison population, but states need to invest more in education more rapidly than that.  In addition, states will likely spend much of the savings from criminal justice reforms elsewhere, in investments such as effective rehabilitative programs that allow formerly incarcerated people to address mental illness and addiction and lead productive lives, or in a mix of investments and tax reductions. 

State Prison Populations Have Grown Rapidly in Past 35 Years

State corrections systems incarcerate the vast majority of prisoners in the United States.  State prisons account for 87 percent of the total prisoner population, with the remaining 13 percent under federal jurisdiction.[2] When one considers the broader population of incarcerated people — that is, including inmates in local jails either awaiting sentence or serving a term of less than one year — state prisons account for just under 60 percent of all people behind bars at any given point in time.[3]

The overall state prison population has grown sizably since the late 1970s, from roughly 270,000 inmates in 1978 to more than 1.3 million in 2013.[4]  That growth far outpaced U.S. population growth.  In the late seventies, states imprisoned around 120 individuals for every 100,000 U.S. residents; in 2007 the state incarceration rate peaked at 450 individuals per 100,000 residents and has fallen only slightly since.[5]


Incarceration rates have more than tripled in 36 states since 1978 and have increased six-fold in four states (Mississippi, New Hampshire, North Dakota, and Pennsylvania).  Rates remain at near-peak levels:

  • Ten states had incarceration rates above 500 prisoners per 100,000 residents in 2013:  Alabama, Arizona, Arkansas, Florida, Georgia, Louisiana, Mississippi, Missouri, Oklahoma, and Texas.
  • Five of those states — Alabama, Louisiana, Mississippi, Oklahoma, and Texas — had incarceration rates above 600 prisoners per 100,000 residents.
  • In one of those states, Louisiana, the incarceration rate stood at nearly 850 prisoners per 100,000 residents.


State Policies, Not Crime Levels, Are Biggest Drivers of Rising Incarceration Rates

Incarceration rates are a function of four variables:  the crime rate, the ratio of arrests to crimes, the share of offenders sent to prison, and the length of prison sentences.  Research shows that the last two of these are the biggest drivers of the increase in state incarceration rates over the past several decades.  Both the share of people sent to prison and the length of their prison stays are under policymakers’ direct control.  Reforms aimed at reducing prison populations will need to target these two areas.

More specifically, research on the causes of rising incarceration rates has found:

  • Crime rates have risen and fallen independently of incarceration rates.  Crime rates began rising in the early 1960s, roughly a decade before state incarceration rates began rising.  In the 1980s, violent and property crime rates fluctuated (falling in the first half of the decade, then rising in the second), while incarceration rates continued rising, undergoing their greatest decade of growth.  By the end of the 1990s, crime rates had fallen to levels not seen since the 1970s, and they have continued to fall throughout the 2000s; yet incarceration rates continued to grow well into the 2000s, peaking in 2007.[6] 


  • Arrests per crime have been relatively stable.  Incarceration rates may rise even when crime rates remain stable if police become more effective at apprehending offenders (in other words, if the arrest rate per crime increases).  However, the likelihood of arresting someone who has committed a crime remained relatively stable between 1980 and 2010.  “[B]y the measure of the ratio of arrests to crimes, no increase in policing effectiveness occurred from 1980 to 2010 that might explain higher rates of incarceration,” a recent National Research Council report concluded.[7]
  • The share of offenders sent to prison has climbed dramatically.  For all major crime types, the likelihood that an offender will go to prison has risen sharply over the past 30 years.  This is especially true for drug offenses; the likelihood of being sent to prison for a drug-related crime rose by 350 percent between 1980 and 2010.  The National Research Council study estimated that the increase in the share of offenders sent to prison accounts for 44 to 49 percent of the long-term growth in state incarceration rates.[8]  
  • Length of stay in prison has grown for all types of crimes.  Between 1990 and 2009, the average time served rose by nearly 25 percent for property crimes and by roughly 37 percent for violent and drug crimes, the Pew Center on the States estimates.  Overall, Pew estimated that individuals released from prison in 2009 spent nine months longer behind bars than offenders released 20 years ago.[9]  The increase in average sentences has contributed as much to the growth in incarceration rates as the rise in the share of offenders sent to prison, and possibly slightly more.

High Incarceration Rates Impose Significant Human Costs, Especially in Certain Neighborhoods

While incarceration rates have risen in every state in recent decades, the impacts have been most acute for a small but geographically concentrated number of neighborhoods.  A 2010 paper by two Harvard criminologists found that incarceration rates in the early and mid-2000s were below 500 per 100,000 adult residents for the majority of Chicago neighborhoods but were more than eight times greater— over 4,000 per 100,000 adult residents — for a small subset of clustered Chicago neighborhoods.[10]  Another study found that North Carolina’s incarceration rate in 2000 was 335 per 100,000 residents statewide, well below the national average at the time, but was 8,000 per 100,000 adult residents in one neighborhood.[11]

Communities most afflicted by high incarceration rates have high levels of poverty, unemployment, and racial segregation.  “To state the claim bluntly, imprisonment and its effects are concentrated in neighborhoods where black people and poor people live,” political scientist Traci Burch has written.[12]

High levels of incarceration impose significant human and economic costs.  People with criminal convictions face serious challenges in finding stable and adequate employment.[13]  Time behind bars is generally time lost developing the skills and education increasingly necessary in today’s labor market, a particular problem given that formerly incarcerated people typically have lower levels of education.[14]  (Nearly 40 percent of state prisoners had not finished high school and only 11 percent had attended college, a Justice Department study conducted in 2003 found.)[15]  For many, incarceration also carries a strong stigma, which can discourage potential employers from hiring.  In addition, in some states, legal barriers prevent people with criminal convictions from holding certain jobs.[16]  

Even those who do find employment typically earn less than otherwise-similar people who have not been incarcerated.[17]  A Pew study found that men with a previous criminal conviction worked roughly nine fewer weeks, and earned 40 percent less, each year than otherwise similar non-offenders.  These effects accumulate over time.  The study also found that men’s total earnings by age 48 are less than half among men who have been incarcerated than among comparable men who have not been incarcerated.  In addition, the study found that overall, incarceration reduces the total earnings of all black men — not just ex-offenders — by 9 percent.[18] 

10-28-14sfp-f4.pngIncarceration also increases poverty, for those who have been to prison as well as other household members, including children.  Many inmates are also parents and/or partners, and their incarceration leaves households with one less potential wage earner.  One study examining poverty and state-level incarceration rates between 1980 and 2004 determined that if incarceration rates had not increased, the official poverty rate would have fallen by roughly 20 percent over that period instead of remaining relatively stable.[19]  A 2008 study estimated that more than 2.7 million children had a parent behind bars, and that this significantly increased the children’s likelihood of being poor.[20]

Because high levels of incarceration are heavily localized, the individual and family effects of imprisonment accumulate to limit entire communities’ economic and social opportunities.  Removing large numbers of working-age men and women from the community depletes the human capital needed to build stable neighborhoods.[21]  That depletion, in turn, tends to reduce economic and social opportunities even for community members with no interaction with the criminal justice system.  A 2003 study found that as incarceration rates rise in a given county, unemployment rates subsequently rise for the county’s non-incarcerated African Americans.[22]

High Incarceration Rates Present Mounting Fiscal Challenge

As the number of individuals connected to the criminal justice system has ballooned, so has state corrections spending, which more than doubled between 1986 and 2013 (after adjusting for inflation), from $20 billion to over $47 billion.[23]  Spending rose in every state except Virginia, by more than four times in nine states (Arkansas, Colorado, Idaho, North Dakota, Oregon, Pennsylvania, West Virginia, Wisconsin, and Wyoming) and by five times in three of those states (Colorado, Idaho, and Pennsylvania).

The share of state general-fund dollars going to corrections rose as well between 1986 and 2013, from 4.7 percent to nearly 7 percent nationally.[24]  For most states, corrections spending is now the third-largest category of spending, behind only education and health care.[25]  In four states (Arizona, Michigan, Oregon, and Vermont), corrections accounts for more than 11 percent of state general fund spending.

Growth in corrections spending has outpaced growth in expenditures in other critical areas of state budgets, such as K-12 and higher education.  State spending on higher education — that is, money spent through the state budget, not by students and families through tuition — rose by less than 6 percent between 1986 and 2013, after adjusting for inflation.  State support for K-12 education grew by 69 percent over this period.[26]  But corrections spending jumped by 141 percent.  Eleven states spent more general funds on corrections than on higher education in 2013; Oregon spent more than twice as much.  In 12 other states, corrections spending was at least 70 percent of state support for higher education.[27] 

If states were still spending the same amount on corrections as they did in the mid-1980s, adjusted for inflation, they would have about $28 billion more available each year for education and other productive investments.

Table 1
State Corrections Spending Has Risen Significantly Since 1986
(in millions of dollars)
  Fiscal Year 1986 (in 2013 $’s) Fiscal Year 2013* Difference
Alabama $270 $460 $190
Alaska $166 $349 $183
Arizona $347 $965 $618
Arkansas $81 $385 $304
California $3,329 $8,618 $5,289
Colorado $130 $670 $540
Connecticut $225 $642 $417
Delaware $100 $265 $165
Florida $772 $2,316 $1,544
Georgia $449 $1,467 $1,018
Hawaii $79 $218 $139
Idaho $32 $213 $181
Illinois $782 $1,295 $513
Indiana $296 $687 $391
Iowa $155 $362 $207
Kansas $151 $354 $203
Kentucky $196 $558 $362
Louisiana $389 $713 $324
Maine $74 $133 $59
Maryland $780 $1,313 $533
Massachusetts $353 $1,261 $908
Michigan $836 $2,064 $1,228
Minnesota $185 $469 $284
Mississippi $117 $312 $195
Missouri $204 $589 $385
Montana $45 $178 $133
Nebraska $74 $198 $124
Nevada $96 $247 $151
New Hampshire $32 $95 $63
New Jersey $670 $1,511 $841
New Mexico $140 $266 $126
New York $2,762 $2,918 $156
North Carolina $538 $1,717 $1,179
North Dakota $17 $81 $64
Ohio $668 $1,798 $1,130
Oklahoma $208** $459 $251
Oregon $196 $802 $606
Pennsylvania $393 $2,111 $1,718
Rhode Island $74 $185 $111
South Carolina $257 $492 $235
South Dakota $28 $85 $57
Tennessee $472 $880 $408
Texas $872 $3,191 $2,319
Utah $115 $239 $124
Vermont $34 $131 $97
Virginia $940** $1,174 $234
Washington $344 $913** $569
West Virginia $45 $219 $174
Wisconsin $279 $1,138 $859
Wyoming $30 $123 $93
    Total Spending Increase: $28,006
*FY 2013 spending levels are estimates collected by the National Association of State Budget Officers. Actual state spending levels may differ.
**Oklahoma data reflect corrections spending in FY 1987; Virginia data reflect FY 1990;  Washington State data reflect FY 2012.
Source: CBPP analysis Data from National Association of State Budget Officers.

States Are Underinvesting in Educating Children in Low-Income Neighborhoods

State economies — and, in particular, the economies of many low-income neighborhoods — would be stronger over time if states spent less in maintaining extremely high prison populations and more to educate children and young adults.  

In recent years, though, states have cut education funding, in some cases by large amounts.  At least 30 states are providing less general funding per student this year for K-12 schools than in state fiscal year 2008, before the Great Recession hit, after adjusting for inflation.[28]  In 14 states, the reduction exceeds 10 percent.  The three states with the deepest funding cuts since the recession hit - Alabama, Arizona, and Oklahoma - are among the ten states with the highest incarceration rates.

Cuts in state funding for colleges and universities have been even deeper.  The average state has cut higher education funding per student by 23 percent since the recession hit, after adjusting for inflation.  The two states with the deepest cuts - ver 40 percent - are Arizona and Louisiana, both in the top ten for incarceration rates.[29]

Many states have also cut funding substantially for preschool programs.  Of the 40 states that help fund preschools, 28 now have lower per-child funding than before the recession hit.[30]  Many of the deepest cuts occurred in the highest-incarcerating states.  Six of the ten states with the highest incarceration rates - Arizona, Florida, Georgia, Missouri, Oklahoma, and Texas - cut preschool funding per child by more than 15 percent between 2008 and 2013.[31]

By reordering their priorities to invest more in education, states could start repairing the damage done by these recession-era cuts and otherwise improve their education systems, especially in high-poverty neighborhoods most directly affected by high incarceration rates.  They could, for example:

  • Expand access to high-quality preschool.  A substantial body of research indicates that children from low-income families who attend a high-quality preschool program improve their cognitive skills and tend to earn more as adults.[32]  Yet only a little over one-third of 3- and 4-year-olds from low-income families nationally are enrolled in preschool, including both public and private programs.[33]  Only New Jersey and the District of Columbia have enrollment rates exceeding 50 percent.

    10-28-14sfp-f5.pngWhile some high-incarcerating states perform comparatively well on this measure, others do not.  Mississippi enrolls nearly half of its low-income 3- and 4-year-olds in preschool, for example, but Arizona enrolls only about 27 percent.

  • Reduce class sizes in high-poverty schools.  Evidence suggests that reducing class sizes can boost achievement, especially in the early grades and for low-income students.[34]  After the recession hit, though, class sizes rose nationally and in a number of states.[35]  Kansas schools, for example, had 19,000 more students last school year than they did in 2009, but 665 fewer teachers.[36] 

    Further, in a handful of states, teachers in high-poverty schools have more students, on average, than teachers in low-poverty schools - the opposite of what is generally required to produce an equitable education system.[37]  In Alabama, for example, the average teacher in a high-poverty school district had 19 students in 2011, while the average teacher in a low-poverty district had 13.  Four of the states with this backward arrangement - Alabama, Arizona, Florida, and Texas - are among the ten highest-incarceration states.

  • Revise state funding formulas to invest more in high-poverty neighborhoods.  Schools receive most of their funding from a combination of state and local sources.  The primary local source is the property tax, which tends to generate more revenue for schools in wealthier areas than in poorer ones because it is based on property values.  State funding can help counteract that inequity.[38]  Unfortunately, a number of states provide less funding for high-poverty schools than for low-poverty schools, while some others provide about the same funding to high- and low-poverty districts.  As of 2011, only 14 states provided at least 5 percent more funding per student for high-poverty districts than low-poverty districts.

    Further, many states provide inadequate funding for schools overall.  While some interstate differences in funding levels are to be expected, given variations in wage rates, poverty levels, population density, and other factors, a number of states have low per-student funding levels even after controlling for these factors.  Based on one study that accounted for these factors, none of the states with the ten highest incarceration rates ranked in the top half of states for school funding per student in 2011.

  • Increase college enrollment and graduation rates for students from low-income families.  Students from low-income families are much less likely to enroll in college than students from wealthier households.  Only about half of recent high school graduates from households in the lowest income quintile (earning less than $18,300) enrolled in postsecondary education in 2012, compared to more than 80 percent of students from households in the top quintile (earning over $90,500 a year).

    In addition, low-income students who enroll in college are much less likely to graduate than their higher-income counterparts.  One study found that only a quarter of low-income students who began college in the 2003-04 school year had attained a bachelor’s degree six years later, compared to nearly 60 percent of high-income students.[41]  

    Adding to the difficulties facing low-income students, sharp reductions in state funding for higher education have led to significant tuition increases.  Tuition at the average public college or university nationally is up $1,936 (28 percent) since the 2007-08 school year, after adjusting for inflation.[42]  While more federal financial aid is now available to offset rising tuition for low-income students, increases in costs of room and board along with a higher “sticker price” at many colleges and universities likely cause some students to choose not to enroll.[43]

Reforms Can Reduce Incarceration Rates and Produce Savings 

States can reduce their incarceration rates and realize significant long-term budget savings without harming public safety.[44]  To do this, state policymakers need to enact reforms that target the main drivers of high incarceration rates:  the number of people admitted (or re-admitted) into correctional facilities and the length of their prison stays.  States should consider four basic kinds of reforms:

  • Decriminalize certain activities and reclassify certain low-level felonies.  The increased use of prison — and longer prison sentences — to punish crimes such as the possession of certain drugs, like marijuana, has contributed heavily to the growth in mass incarceration.  Lawmakers should look to reduce or eliminate criminal penalties for such crimes when doing so would not affect public safety. 
  • Expand the use of alternatives to prison for non-violent crimes and divert people with mental health or substance abuse issues away from the criminal justice system altogether.  Policymakers should assess the range of sentencing alternatives available in their state, such as drug and mental health courts and related treatment, community correction centers, community service, sex offender treatment, and fines and victim restitution.  Whenever possible, people whose crimes stem from addiction or mental illness should be diverted into treatment programs rather than sent to prison.  These treatment programs should be high-quality and adequately funded.[45]
  • Reduce the length of prison terms and parole/probation periods.  Policymakers should reform unnecessarily harsh sentencing policies, including “truth-in-sentencing” requirements and mandatory minimum sentences, and allow inmates to reduce their sentences through good time or earned time policies.  States also should expand programs that enable inmates meeting certain requirements to receive favorable decisions in parole hearings, especially in states where parole grant rates remain low.  Funding for programs to help inmates meet these requirements, in areas such as substance abuse, anger management, literacy, or higher education, has not kept pace with the growth in state prison populations.[46]
  • Restrict the use of prison for technical violations of parole/probation.  The share of individuals entering prison due to a parole violation grew rapidly between the late 1970s and the late 2000s.  While it has fallen more recently, parole revocations accounted for more than a quarter of admissions to state prisons in 2013.[47]  Some of these violations are technical, such as missing a meeting with a probation officer or failing a drug test.  States should heavily restrict the use of prison for technical parole violators and implement graduated sanctions for more serious parole violations.

    States can also adopt more effective probation policies.  For example, Hawaii has sharply reduced probation revocations with a program that punishes infractions more quickly and with more certainty, but with much shorter periods of incarceration.[48]

These reforms are complementary; adopting just one or two won’t shrink a state’s prison population as much as a more comprehensive set of reforms that improves “front-end” sentencing and admission policies as well as “back-end” release and re-entry policies.

What Policy Mechanisms Do States Need to Support Those Reforms?

States wishing to use savings from criminal justice reforms for more productive purposes would do well to adopt planning and budgeting mechanisms that can help them shift priorities, including the following.

High-Quality, Long-Term Forecast of Potential Savings

Lawmakers often don’t have the information needed to make educated decisions on proposed criminal justice reform legislation because they lack an official estimate of the fiscal impact of the reforms, also known as a “fiscal note.”  Roughly 40 percent of the major criminal justice bills enacted in states in 2009-2011 had no fiscal note.  In about half of the states where a fiscal note was produced, the notes projected fiscal impacts no more than two years into the future.  Moreover, some states had little or no process to ensure that the fiscal notes were credible, such as a review by independent analysts. (See Box 1.)[49]

Box 1: Fiscal Note Best Practices

When drafting fiscal notes, states should strive to make them:a

Consistent.  Fiscal notes should be produced in a consistent format by trusted, non-partisan staff.  All major bills that have reached a certain stage in the legislative process should be analyzed for their fiscal impact.

Properly researched.  Fiscal notes should estimate savings and potential costs and include a detailed explanation in instances where an estimate cannot be calculated.  At a minimum, they should seek to forecast five years into the future.

Detailed.  In complex reform bills, the analysis should extend to the bills’ individual major provisions.  Fiscal notes should also attempt to estimate impacts on local as well as state finances and on the size of prison and jail populations.

Accessible.  Fiscal notes should be clearly written and available online and should include contact information for the analyst or staff responsible.

a See Michael Leachman et al., “Improving Budget Analysis of State Criminal Justice Reforms: A Strategy for Better Outcomes and Saving Money,” Center on Budget and Policy Priorities and American Civil Liberties Union, January 11, 2012,

Accepted Process to Estimate Annual Savings Once Reforms Are Enacted

To capture the savings from criminal justice reforms for more productive human capital investments (including education), states need to estimate how much specific reforms save following enactment.  These estimates need to be produced each time a state writes its budget - annually in most states - so that lawmakers can incorporate the savings into the budget.  Estimates need to be produced in a manner that is accepted as credible by legislators, other policymakers, and the public, so that the process does not bog down in arguments over the numbers.

A California ballot initiative (Proposition 47) could serve as a model.  It would reform sentencing policies and place all resulting savings in a special fund to be used for certain specified purposes, primarily addiction and mental health treatment and school programs for high-risk youth.  (See Box 2.)  To determine how much to deposit into the fund, Proposition 47 would require the state finance department to estimate the savings attributable to the measure annually. 

Alternatively, states can choose to allocate savings based on estimates produced during the initial fiscal note process.  For this to occur, it is important that fiscal notes be properly researched, consistent, and provide detailed estimates of savings far enough into the future.  Such a process would be similar to what occurs at the federal level when the Congressional Budget Office estimates the potential costs or savings of proposed legislation and these estimates are then used to determine the savings that can be spent in subsequent years.[50]

Established Mechanism to Shift Savings to Investments in Human Capital

States typically have three alternative ways to shift savings from criminal justice reforms into human capital investments.  They can:

  • Establish a mechanism to automatically estimate the savings and divert them into a special fund.  One approach is for policymakers enacting criminal justice reforms to simultaneously create a “reinvestment fund” that captures savings from the reforms and allocates those savings largely to human capital investments, including educational programming targeted to high-poverty communities and improved mental health and substance abuse treatment programs.  As the state reports each year the estimated savings from the reforms, an equal amount is automatically deposited into the fund.  The legislation establishing the fund would specify how the money is to be spent.  California’s Proposition 47 follows this approach.
  • Appropriate savings through state budget processes.  Through the annual budget process, policymakers can estimate the savings from criminal justice reforms and determine how much of those savings to reinvest and in what programs.  This option gives policymakers greater flexibility and control.  On the other hand, it can place reinvestments in future years in jeopardy as political priorities change or fiscal and economic conditions shift.
  • Reallocate spending at the department or agency level.  Within departments or agencies, states can shift dollars away from incarceration and into human capital investments.  For instance, after North Carolina enacted criminal justice reforms in 2011, policymakers shifted $16 million into community-based treatments by drawing on resources that were already in the Department of Public Safety’s budget but would no longer be needed for corrections costs.[51]

Box 2:  California Ballot Proposal Offers Model

Proposition 47 (“The Safe Neighborhoods and Schools Act”), which will appear on California’s statewide ballot on November 4, contains criminal justice reforms designed to reduce the state prison population, coupled with measures to reinvest the savings.

California has the nation’s second-largest prison system, with over 130,000 individuals under state jurisdiction.a  With the system operating at roughly 140 percent of capacity, California has located more than 15,000 prisoners in private prison facilities and out-of-state prisons.b  In addition, a federal court ruling that prison overcrowding is preventing the state from giving prisoners adequate medical and mental health care means that California must reduce its prison population.

To accomplish this, Proposition 47 would:

  • Make targeted sentencing reductions by reclassifying certain offenses from felonies to misdemeanors, for both current and future offenders.  Proposition 47 would reclassify seven types of non-violent drug and property crimes (such as shoplifting, drug possession, and petty theft) from felonies to misdemeanors,c thereby shortening the maximum penalty from a multi-year prison sentence to one year in jail.  Since the change would be retroactive, qualifying prisoners could apply for resentencing and see their sentences reduced. 
  • Require the state to calculate the savings from the reforms each year and deposit them in a dedicated fund.  The initiative would require California’s Department of Finance to estimate the state savings attributable to the measure each year and deposit them in a special fund.
  • Earmark the savings for specific investments.  Savings deposited in the fund could only be used for three explicit purposes and in specified proportions:  65 percent would go to mental health services, drug treatment, and “diversion programs” designed to enable offenders to avoid criminal charges and a criminal record; 25 percent would go to supporting at-risk youth in schools; and 10 percent would go to victim services.

California’s Legislative Analyst’s Office has estimated that Proposition 47 would likely cut the state’s prison population by “several thousand” inmates while generating corrections savings in the “low hundreds of millions” of dollars annually.d

a CBPP calculations, Bureau of Justice Statistics, “National Prisoner Statistics,”
b “Proposition 47: Should California Reduce Penalties for Drug and Property Crimes and Invest in Treatment?” California Budget Project, September 2014,
c These seven categories are check fraud, drug possession, forgery, petty theft, shoplifting, receiving stolen property, and petty theft with a prior theft-related conviction.  This change would not apply to individuals with prior convictions for violent offenses or registerable sex offences.
d “Proposition 47: Criminal Sentences. Misdemeanor Penalties Initiative Statute,” California Legislative Analyst’s Office,

Independent Commission to Monitor Implementation and Enforce Compliance

States adopting significant criminal justice reforms can create an oversight commission to craft and recommend further reforms, propose legislation, assist in implementation, and evaluate the results.  The commission should include experts and individuals rooted in communities most affected by high incarceration rates.

South Carolina, for instance, created a bipartisan sentencing reform commission in 2008 to recommend changes to state law.  The commission proposed a set of reforms in 2009 that, among other things, required fiscal impact statements for future criminal justice legislation, eliminated mandatory minimum sentences for drug possession, and expanded parole eligibility for certain offenses.  Projections indicated that these reforms, enacted in 2010, would slow the growth of the prison population over the next five years by 7.3 percent and save roughly $241 million over that period.  At the commission’s urging, policymakers also created a standing oversight committee with the authority to spend part of the resulting savings.[52]


Appendix Table 1
Sentenced Prisoners Under the Jurisdiction of State Correctional Authorities, 1978-2013
  1978 2013 Population Increase, 1978-2013
Alabama 5,529 31,354 25,825
Alaska 490 2,682 2,192
Arizona 3,450 39,062 35,612
Arkansas 2,578 17,159 14,581
California 19,550 135,981 116,431
Colorado 2,474 20,371 17,897
Connecticut 1,863 12,162 10,299
Delaware 1,005 4,112 3,107
Florida 21,243 103,028 81,785
Georgia 10,833 53,478 42,645
Hawaii 486 3,618 3,132
Idaho 802 7,549 6,747
Illinois 10,765 48,653 37,888
Indiana 4,396 29,905 25,509
Iowa 1,961 8,654 6,693
Kansas 2,339 9,506 7,167
Kentucky 3,390 20,330 16,940
Louisiana 7,291 39,298 32,007
Maine 577 1,972 1,395
Maryland 7,966 20,988 13,022
Massachusetts 2,785 9,643 6,858
Michigan 14,944 43,704 28,760
Minnesota 1,954 10,289 8,335
Mississippi 2,633 20,742 18,109
Missouri 5,637 31,537 25,900
Montana 680 3,642 2,962
Nebraska 1,242 4,929 3,687
Nevada 1,350 12,915 11,565
New Hampshire 283 2,848 2,565
New Jersey 5,422 22,452 17,030
New Mexico 1,505 6,687 5,182
New York 20,459 53,428 32,969
North Carolina 12,268 35,181 22,913
North Dakota 138 1,513 1,375
Ohio 13,107 51,729 38,622
Oklahoma 4,186 25,496 21,310
Oregon 2,873 15,180 12,307
Pennsylvania 7,722 50,083 42,361
Rhode Island 524 2,039 1,515
South Carolina 7,130 21,443 14,313
South Dakota 505 3,641 3,136
Tennessee 5,835 28,521 22,686
Texas 24,575 160,295 135,720
Utah 908 7,071 6,163
Vermont 351 1,575 1,224
Virginia 7,882 36,982 29,100
Washington 4,487 17,947 13,460
West Virginia 1,237 6,812 5,575
Wisconsin 3,432 21,285 17,853
Wyoming 433 2,310 1,877
Total State Prison Population* 268,005 1,321,781 1,053,776
See data download for full time series and state-specific footnotes.
*1978 total includes prisoners held under the jurisdiction of the District of Columbia.  As of 2001, sentenced felons from the District of Columbia are the responsibility of the Federal Bureau of Prisons.
Source: Bureau of Justice Statistics, National Prisoner Statistics Program, 1978-2013.
Appendix Table 2
Imprisonment Rate of Sentenced Prisoners Under the Jurisdiction of State Correctional Authorities per 100,000 U.S. Residents, 1978-2013
  1978 2013 Percent Increase, 1978-2013
Alabama 144 647 349%
Alaska 122 364 198%
Arizona 137 586 328%
Arkansas 115 578 403%
California 86 353 310%
Colorado 89 384 331%
Connecticut 60 338 463%
Delaware 168 442 163%
Florida 233 524 125%
Georgia 205 533 160%
Hawaii 52 257 394%
Idaho 88 466 430%
Illinois 94 377 301%
Indiana 81 454 460%
Iowa 67 279 316%
Kansas 100 328 228%
Kentucky 94 462 391%
Louisiana 179 847 373%
Maine 52 148 185%
Maryland 189 353 87%
Massachusetts 48 192 300%
Michigan 162 441 172%
Minnesota 49 189 286%
Mississippi 106 692 553%
Missouri 116 521 349%
Montana 87 357 310%
Nebraska 80 263 229%
Nevada 188 460 145%
New Hampshire 32 215 572%
New Jersey 74 252 241%
New Mexico 120 321 168%
New York 115 271 136%
North Carolina 214 356 66%
North Dakota 21 211 905%
Ohio 121 446 269%
Oklahoma 144 659 358%
Oregon 114 385 238%
Pennsylvania 65 391 502%
Rhode Island 55 194 253%
South Carolina 234 447 91%
South Dakota 73 428 486%
Tennessee 131 438 234%
Texas 182 602 231%
Utah 67 242 261%
Vermont 70 251 259%
Virginia 149 446 199%
Washington 115 256 123%
West Virginia 64 367 473%
Wisconsin 74 370 400%
Wyoming 100 395 295%
Overall imprisonment rate* 119 417 250%
See data download for full time series and state-specific footnotes.
*1978 total includes prisoners held under the jurisdiction of the District of Columbia.  As of 2001, sentenced felons from the District of Columbia are the responsibility of the Federal Bureau of Prisons.
Source: Bureau of Justice Statistics, National Prisoner Statistics Program, 1978-2013.
Appendix Table 3
Corrections General Fund Spending as a Share of Total General Fund Expenditures, 1986-2013
  Fiscal Year 1986 Fiscal Year 2013*
Alabama 4.3% 6.7%
Alaska 6.6% 4.8%
Arizona 7.1% 11.3%
Arkansas 2.6% 8.1%
California 5.0% 9.0%
Colorado 3.1% 8.4%
Connecticut 2.9% 3.4%
Delaware 5.1% 7.2%
Florida 5.3% 9.4%
Georgia 4.7% 8.0%
Hawaii 2.3% 3.8%
Idaho 2.6% 7.9%
Illinois 4.4% 4.4%
Indiana 4.8% 4.8%
Iowa 3.4% 5.8%
Kansas 3.7% 5.7%
Kentucky 3.6% 5.9%
Louisiana 4.4% 8.7%
Maine 3.7% 4.4%
Maryland 8.8% 8.7%
Massachusetts 2.6% 4.9%
Michigan 6.8% 22.5%
Minnesota 1.7% 2.3%
Mississippi 3.7% 6.6%
Missouri 3.2% 7.3%
Montana 5.7% 9.1%
Nebraska 4.2% 5.5%
Nevada 9.7% 7.8%
New Hampshire 3.9% 7.5%
New Jersey 4.1% 4.8%
New Mexico 4.7% 4.7%
New York 6.0% 4.9%
North Carolina 4.2% 8.3%
North Dakota 1.5% 3.6%
Ohio 3.9% 5.7%
Oklahoma 4.8%** 6.7%
Oregon 5.7% 13.5%
Pennsylvania 2.0% 7.6%
Rhode Island 3.3% 5.7%
South Carolina 4.8% 7.7%
South Dakota 3.4% 6.5%
Tennessee 6.9% 7.0%
Texas 4.4% 7.3%
Utah 4.4% 4.8%
Vermont 4.2% 13.4%
Virginia 8.5%** 6.6%
Washington 3.6% 5.9%**
West Virginia 1.3% 5.3%
Wisconsin 2.7% 8.1%
Wyoming 3.3% 3.3%
Total State Prison Population* 4.7% 6.8%
See data download for full time series and state-specific footnotes.
* FY 2013 levels are estimates collected by the National Association of State Budget Officers.  Actual state spending levels may differ.
**Oklahoma data reflect corrections spending in FY 1987; Virginia data reflect FY 1990; Washington State data reflect FY 2012.
Source: National Association of State Budget Officers.

End notes:

[1] New York and New Jersey reduced their prison populations by 26 percent between 1999 and 2012. California reduced its prison population by 23 percent between 2006 and 2012. See Marc Mauer and Nazgol Ghandnoosh, “Fewer Prisoners, Less Crime: A Tale of Three States,”The Sentencing Project, July 2014,

[2] CBPP calculations, Bureau of Justice Statistics, “Sentenced prisoners under the jurisdiction of state or federal correctional authorities, December 31, 1978-2013,” Note that while the average person often uses the terms “prison” and “jail” interchangeably, they refer to different criminal justice facilities.  As defined by the Bureau of Justice Statistics; “[j]ails are locally-operated, short term facilities that hold inmates awaiting trial or sentencing or both, and inmates sentenced to a term of less than 1 year, typically misdemeanants.  Prisons are long-term facilities run by the state or the federal government and typically hold felons and inmates with sentences of more than 1 year.” See

[3] Jeremy Travis, Bruce Western, and Steve Redburn (editors), The Growth of Incarceration in the United States: Exploring Causes and Consequences, National Academies Press, 2014.

[4] CBPP calculations, Bureau of Justice Statistics, “National Prisoner Statistics,”

[5] CBPP calculations, Bureau of Justice Statistics, “Imprisonment rate of sentenced prisoners under the jurisdiction of state or federal correctional authorities per 100,000 U.S. residents, December 31, 1978-2013,”

[6] CBPP analysis of incarceration data from BJS and FBI, Uniform Crime Reports, prepared by the National Archive of Criminal Justice Data.

[7] Travis et al., p. 49.

[8] Ibid.

[9] Pew Center on the States, “Time Served: The High Cost, Low Return of Longer Prison Terms,” 2012.  Other studies have produced similar findings. The National Research Council found that between 1980 and 2010, the time served for aggravated assault, burglary, and robbery increased by 83, 41, and 79 percent, respectively. 

[10] Robert J. Sampson and Charles Loeffler, “Punishment’s Place: The Local Concentration of Mass Incarceration,” Daedalus, Summer 2010, pp. 20-31.  (See Figure 3, “Spatial Concentration of Incarceration in Chicago, 2000-2005,” on p. 24.)  See also Visher and Farrell (2005), which found that over half of former male prisoners reentering Chicago from prison returned to just seven of the city’s 77 distinct community areas. 

[11] The study looked at census “block groups.” Traci Burch, “The Old Jim Crow: Racial Residential Segregation and Neighborhood Imprisonment,” American Bar Foundation, 2008,

[12] Traci Burch, “The Spatial Concentration of Imprisonment and Racial Political Inequality,”

[13] See, for example, Holzer et al. (2003), Raphael (2007), and Schmitt (2010) on the impacts of incarceration on employment and wage prospects.

[14] Steven Raphael, “Improving Employment Prospects for Former Prison Inmates: Challenges and Policy,” National Bureau of Economic Research, April 2010,

[15] Caroline W. Harlow, “Education and Correctional Populations,” U.S. Department of Justice, Bureau of Justice Statistics, March 2003.

[16] Jon Schmitt and Kris Warner, “Ex-offenders and the Labor Market,” Center for Economic and Policy Research, November 2010.

[17] Bruce Western, “The Impact of Incarceration on Wage Mobility and Inequality,” American Sociological Review, August 2002,

[18] Pew Charitable Trusts, “Collateral Costs: Incarceration’s Effect on Economic Mobility,” 2010.

[19] Robert H. DeFina and Lance Hannon, “The Impact of Mass Incarceration on Poverty,” Crime and Delinquency, February 23, 2009.  Available at SSRN:

[20] “Collateral Costs.”

[21] Todd R. Clear, Imprisoning Communities: How Mass Incarceration Makes Disadvantaged Neighborhoods Worse, Oxford University Press, 2007.

[22] Clear, citing William J. Sabol and James P. Lynch, “Assessing the Longer-run Effects of Incarceration: Impact on Families and Employment,” 2003.

[23] CBPP calculations, National Association of State Budget Officers.  The year 1986 represents the first year of complete budget data on corrections spending that states reported to NASBO.  State spending levels from years prior to 2013 are expressed in 2013 dollars.

[24] The share declined in only six states: Alaska, Maryland, Nevada, New York, Virginia, and Washington.

[25] CBPP calculations, data from NASBO.  Education spending includes spending for higher education.  Overall in FY 2012, roughly 45 percent of state general fund dollars went to education (K-12 and higher education combined), with another 20 percent going to health care.

[26] The number of K-12 students also rose over this period, by about 10 million students or about 25 percent.

[27] CBPP calculations, data from NASBO.

[28] Michael Leachman and Chris Mai, “Most States Still Funding Schools Less Than Before the Recession,” Center on Budget and Policy Priorities, October 16, 2014,

[29] Michael Mitchell, Vincent Palacios, and Michael Leachman, “States Still Funding Higher Education Below Pre-Recession Levels,” Center on Budget and Policy Priorities, May 1, 2014,

[30] From 2007-08 to 2012-13, the latest data available.  Authors’ analysis of data from National Institute for Early Education Research, The State of Preschool 2013,

[31] Ibid.  Louisiana, also among the ten highest-incarcerating states, is excluded from this list because most or all of its very deep (76 percent) cut in state pre-K funding was offset by an increase in the state’s use of federal TANF funds for the same purpose.

[32] See Timothy J. Bartik, “From Preschool to Prosperity,” W.E. Upjohn Institute for Employment Research, 2014, See also Julia Isaacs, “Research Brief #1: State Pre-Kindergarten,” Brookings Institution, September 2008,

[33] Bruce Baker, David Sciarra, and Danielle Farrie, “Is School Funding Fair: A National Report Card,” Education Law Center, Third Edition, February 2014, p. 32, .

[34] See Diane Whitmore Schanzenbach, “Does Class Size Matter?” National Education Policy Center, February 2014,  See also Matthew M. Chingos and Grover J. “Russ” Whitehurst, “Class Size:  What Research Says and What it Means for State Policy,” Brookings Institution, May 11, 2011,

[35] See National Center for Educational Statistics, Schools and Staffing Survey, 2007-08, Table 8 and 2011-12, Table 7.

[36] Kansas Center for Economic Growth, “Quality at Risk: Impact of Education Cuts,” August 2014,

[37] Baker, Sciarra, and Farrie, Table 8.  A “high-poverty district” in this study has a poverty rate of 30 percent, while a “low-poverty district” has a poverty rate of zero.

[38] See, for example, Bruce D. Baker, “Evaluating the Recession’s Impact on State School Finance Systems,” Education Policy Analysis Archives, Volume 22, Number 91, September 15, 2014.

[39] Baker, Sciarra, and Farrie, pp. 17-18.

[40] Baker, Sciarra, and Farrie, p. 13.

[41] Sandy Baum et al., “Education Pays 2013: The Benefits of Higher Education for Individuals and Society,” College Board, October 2013,

[42] Mitchell, Palacios, and Leachman.

[43] Overall, the cost of attending college has risen for low-income students because the cost of room and board has increased, too.  As a result, the net cost of attendance at four-year public institutions for these students increased 12 percent from 2008 to 2012, after adjusting for inflation.

[44] Marc Mauer and Jenni Gainsborough, “Diminishing Returns: Crime and Incarceration in the 1990s,”The Sentencing Project, September 2000,

[45] See ACLU of Southern California, “A Way Forward: Diverting People with Mental Illness from Inhumane and Expensive Jails into Community-Based Treatment that Works,” July 2014,  See also Families Against Mandatory Minimums, “Alternatives to Incarceration in a Nutshell,” July 8, 2011,

[46] Sarah Lawrence et al., “The Practice and Promise of Prison Programming,” Urban Institute, May 2002,  See also Rand Corporation, “Evaluating the Effectiveness of Correctional Education: A Meta-Analysis of Programs that Provide Education to Incarcerated Adults” 2013, p. 60.

[47] Bureau of Justice Statistics, Number of parole violation admissions of sentenced prisoners to state or federal prisons, 1978-2013.

[48] Significant research has been done on Hawaii’s HOPE program.  See Hawken, Angela & Mark Klaiman “Managing Drug Involved Probationers with Swift and Certain Sanctions: Evaluating Hawaii’s HOPE,” December 2009, U.S. Department of Justice.  For a summary of the program and its results see, Pew Center on the States, “The Impact of Hawaii’s HOPE Program on Drug Use, Crime and Recidivism,” January 2010, and VERA Institute of Justice, “More than the Sum of Its Parts: Why Hawaii’s Opportunity Probation with Enforcement (HOPE) Program Works,” 2012.

[49] Michael Leachman et al., “Improving Budget Analysis of State Criminal Justice Reforms: A Strategy for Better Outcomes and Saving Money,” Center on Budget and Policy Priorities and American Civil Liberties Union, January 11, 2012,  

[50] Regardless of which method is used, states would do well to adopt the practice of publishing a “current services baseline” — a projection of the state’s costs to maintain the same level of services, absent any relevant policy changes —in order to more easily estimate the savings attributable to criminal justice reforms.  Current services baseline projections take into account inflation and other changes in the cost of providing services, changes in the size of the population being served, and past rule changes that are still being phased in.  Such projections are fairly uncommon at the state level; a 2011 CBPP report found that fewer than half of states prepare current services baselines.  See Elizabeth McNichol and Dylan Grundman, “The Current Services Baseline: A Tool for Understanding Budget Choices,” October 21, 2011,

[51] Nancy Lavigne et al., “Justice Reinvestment Initiative State Assessment Report,” Urban Institute, Bureau of Justice Assistance, 2014,

[52]  The committee was required to spend the savings only on stronger parole and probation programs.  American Civil Liberties Union, “Smart Reform is Possible: States Reducing Incarceration Rates and Costs While Protecting Communities,” 2011,