the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

Fewer Prisoners, Less Crime:
A Tale of Three States
Although the pace of criminal justice reform has accelerated at both the federal
and state levels in the past decade, current initiatives have had only a modest
effect on the size of the prison population. But over this period, three states
– New York, New Jersey, and California – have achieved prison population
reductions in the range of 25%. They have also seen their crime rates generally
decline at a faster pace than the national average.
Key findings:
•	 New York and New Jersey led the nation by reducing
their prison populations by 26% between 1999 and
2012, while the nationwide state prison population
increased by 10%.
•	 California downsized its prison population by 23%
between 2006 and 2012. During this period, the
nationwide state prison population decreased by
just 1%.
•	 During their periods of decarceration, violent
crime rates fell at a greater rate in these three states
than they did nationwide. Between 1999-2012, New
York and New Jersey’s violent crime rate fell by
31% and 30%, respectively, while the national rate
decreased by 26%. Between 2006-2012, California’s
violent crime rate drop of 21% exceeded the
national decline of 19%.
•	 Property crime rates also decreased in New York
and New Jersey more than they did nationwide,
while California’s reduction was slightly lower than
the national average. Between 1999-2012, New
York’s property crime rate fell by 29% and New
Jersey’s by 31%, compared to the national decline
of 24%. Between 2006-2012, California’s property
crime drop of 13% was slightly lower than the
national reduction of 15%.

These prison population reductions have come about
through a mix of changes in policy and practice designed
to reduce admissions to prison and lengths of stay. The
experiences of these states reinforce that criminal justice
policies, and not crime rates, are the prime drivers of
changes in prison populations. They also demonstrate
that it is possible to substantially reduce prison
populations without harming public safety.

A Decade of Evolving
Criminal Justice Reform
For more than a decade the political environment
shaping criminal justice policy has been evolving in a
direction emphasizing “smart on crime” and evidencebased approaches to public safety. This has involved
growing bipartisan campaigns at both the federal and
state levels to promote more strategic sentencing and
reentry policies, and to address the unprecedented
growth and cost of the corrections system created over
the past several decades.
The changing climate can be seen in a variety of
legislative, judicial, and policy changes during this period
of time. At the federal level, this has included the Fair
Sentencing Act of 2010 which reduced the disparity in
sentencing between crack and powder cocaine offenses;
the adoption of the Second Chance Act in 2008 which
currently funds about $67 million in reentry services

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

1

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

annually; and the U.S. Supreme Court’s 2005 Booker and in 2012 (the most recent year for which data are
decision making the Federal Sentencing Guidelines available) 27 states experienced a reduction in their
advisory and thereby restoring a greater degree of population.
sentencing discretion to federal judges.
While these trends are encouraging it is also important
At the state level, 29 states have adopted reforms to note that the overall scale of change has been quite
designed to scale back the scope and severity of their modest. The national prison population has only
mandatory sentencing policies over the past decade.1 declined by less than 2% annually in recent years, and
Voters in California approved a ballot initiative in 2012 a disproportionate amount of that decline is due to
that curbed the scope of the state’s notoriously broad California’s “Realignment” policy. In 2012, the prison
“three strikes and you’re out” law and policymakers population reduction of 15,000 in California accounted
around the country have become increasingly for half of the national decline for all states that year.
supportive of Justice Reinvestment initiatives, reducing
parole revocations, establishing treatment courts, and
developing alternatives to incarceration.

Limited Impact on
Incarceration to Date
The impact of these various initiatives on incarceration
has been mixed. At the federal level the prison population
has continued its more than three-decade historic rise,
driven in large part by the ongoing effect of mandatory
penalties for many drug and gun crimes, and increasing
incarceration for immigration offenses.2
At the state level there has been more of a shift in prison
population trends. The number of people incarcerated
in state prisons has declined for three years since 2010,

Substantial Prison
Population Declines in Three
States

The exceptions to the modest scale of decarceration
can be seen in three states – New York, New Jersey,
and California – each of which has reduced its prison
population in the range of 25% over the past decade.
While New York and New Jersey reduced their prisoner
counts by 26% between 1999 and 2012, the nationwide
state prison population increased by 10%. While
California downsized its prison population by 23%
between 2006 and 2012, the nationwide level decreased
by just 1%.3 Six other states achieved double-digit
reductions during varying periods within these years,

80,000

1,400,000

70,000

1,200,000

60,000

1,000,000

50,000

800,000

40,000
600,000

30,000

400,000

20,000

200,000

10,000
0

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012

People in All State Prisons

People in New York & New Jersey Prisons

Figure 1. People in State Prisons in New York, New Jersey, and All States, 1999-2012

All States
New York
New Jersey

0

1	 Ram Subramanian & Ruth Delaney, Playbook for Change? States Reconsider Mandatory Sentences (Vera Institute of Justice 2014),
available at http://www.vera.org/sites/default/files/resources/downloads/mandatory-sentences-policy-report-v3.pdf.
2	 All prison population data taken from various corrections reports of the Bureau of Justice Statistics.
3	 California’s Realignment policy has produced a concomitant increase in the jail population, although of a much smaller magnitude than the prison
decline, as described later in the report.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

2

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

Figure 2. People in California Prisons and All States,
2006-2012

levels of incarceration place the nation well past the
point of diminishing returns in crime control. 

California

180,000

1,400,000

160,000

1,200,000

140,000

1,000,000

120,000
100,000

800,000

80,000

600,000

60,000

400,000

40,000

200,000

20,000
0

2006 2007 2008 2009 2010 2011 2012

People in All State Prisons

People in California Prisons

All States

0

IMPACT OF PRISON POPULATION
REDUCTIONS ON CRIME
The periods in which New York, New Jersey, and
California significantly decreased their prison populations
were ones in which crime rates were declining around the
country.4 Yet in these states, crime rates generally fell at
a faster pace than in the country as a whole. In all three
states, violent crime rates decreased more than they did
nationwide. Property crime rates decreased in New York
and New Jersey more than they did nationwide, while
California’s property crime reduction was slightly lower
than the national average.

though of a lesser magnitude: Colorado, Connecticut,
All three states experienced violent
Hawaii, Michigan, Rhode Island, and Vermont. Coming
after a nearly four-decade historic rise in imprisonment, crime drops that exceeded the
the substantial and sustained reductions in New York, national average
New Jersey, and California should be particularly The violent crime rate measures the incidence of four
instructive for all concerned with excessively high prison crime categories (murder, forcible rape, robbery, and
populations.
aggravated assault) per 100,000 residents. Between
Table 1. State Prisoners in the United States, New York, 1999-2012, the nationwide violent crime rate decreased
by 26%. New York and New Jersey outpaced this
New Jersey, and California: 1999, 2006, 2012
decline, with reductions of 31% and 30%, respectively.
California’s violent crime drop of 21% between 20062012 also exceeded the national decline of 19% during
Prisoners
1999
2012
Change
this period.
Nationwide
1,191,118
1,314,906
10%
NY

72,896

54,073

-26%

NJ

31,493

23,225

-26%

Prisoners
Nationwide
CA

2006

2012

Change

1,331,127

1,314,906

-1%

173,942

134,211

-23%

Table 2. Violent Crime Rates in the United States, New
York, New Jersey, and California: 1999, 2006, 2012
Violent Crime Rate

1999

2012

Change

Nationwide

523.0

386.9

-26%

NY

588.8

406.8

-31%

411.9
290.2
-30%
Because incarceration is ostensibly designed to support NJ
public safety, in this analysis we review how prisoner
2006
2012
Change
reductions in these three states impacted crime control. Violent Crime Rate
Nationwide
479.3
386.9
-19%
While some political leaders warn of a “crime wave”
532.5
423.1
-21%
when prison population reductions are considered, CA
such talk ignores the complexity of how public safety is
produced. Incarceration is a limited factor among many New York and California’s violent crime reductions
that shape public safety. Further, in the era of mass have exceeded nationwide trends despite recent upticks.
incarceration, there is a growing consensus that current Between 2010 and 2012, while the nationwide violent
4	 All crime data are offenses known to law enforcement, taken from the Federal Bureau of Investigation’s Crime in the United States series.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

3

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

crime rate slowed its decline, New York’s violent crime
rate increased each year –  by 3.7% between 2010 and
2012. Because this uptick has only brought the state back
to its 2007 level, New York maintains its historically low
violent crime rate.

unlikely to have been related to the prison downsizing
achieved through Realignment, a policy that went into
effect in October 2011.5 Lofstrom and Raphael compared
monthly changes in crime rates with changes in jail and
prison incarceration rates in each California county in
the twelve months before and after Realignment. “There
is no evidence that realignment resulted in an increase
in murder or rape, with the estimates near zero and
statistically insignificant,” they concluded. When they
examined California crime data without controlling for
broader regional trends, they found that Realignment
had a small and marginally significant effect on robbery
and aggravated assault. But “all evidence of an effect
of realignment on violent crime vanishes,” they noted,
when broader regional trends are incorporated into the
analysis.

Figure 3. Violent Crime Rates in the United States, New
York, and New Jersey, 1999-2012
600
500

New York

400

United States

300

New Jersey

2011

2012

2010

2009

2007

2008

2006

2005

2004

2003

2001

2002

1999

2000

200

Figure 4. Violent Crime Rates in the United States and
California, 2006-2012

All three states experienced
substantial declines in property
crime rates, and two exceeded the
national average

300

The property crime rate measures the incidence of
four crime categories (burglary, larceny-theft, motor
vehicle theft, and arson) per 100,000 residents. While
the national property crime rate decreased by 24%
between 1999 and 2012, New York’s rate dropped by
29% and New Jersey’s by 31%. Between 2006 and 2012,
California’s property crime rate decreased significantly,
but at a slightly lower rate than the national average.
While California’s property crime rate fell by 13% during
this period, the nationwide property crime rate fell by
15%.

200

Table 3. Property Crime Rates in the United States, New
York, New Jersey, and California: 1999, 2006, 2012

600

500

California
400

United States

2006

2007

2008

2009

2010

2011

2012

Property Crime Rate 1999

California’s violent crime drop between 2006 and 2012
also exceeded the national average. Its rate in 2012 was
21% lower than in 2006 despite an increase of 2.9% in
2012.
Magnus Lofstrom and Steven Raphael’s analysis of
county-level variation in crime and incarceration rates has
shown that California’s violent crime uptick in 2012 was

2012

Nationwide

Change

3743.6

2,859.2

-24%

NY

2690.5

1,922.0

-29%

NJ

2988.2

2,047.3

-31%

Property Crime Rate

2006

2012

Change

Nationwide

3,346.6

2,859.2

-15%

CA

3,170.9

2,758.7

-13%

5	 Magnus Lofstrom & Steven Raphael, Public Safety Realignment and Crime Rates in California (Public Policy Institute of California 2013),
available at http://www.ppic.org/content/pubs/report/R_1213MLR.pdf.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

4

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

New York and New Jersey have outpaced the national
decline in property rates since 1999 even while
experiencing modest upticks in some years. Both
nationwide and in New York and New Jersey, property
crime rates have been falling at a slower rate since 2009.

thefts, was related to Realignment.6 “The start of the
increase in motor vehicle theft coincides exactly with the
implementation of realignment in October 2011,” they
write, and there was a statistically significant relationship
between decarceration and motor vehicle theft at the
county level even after incorporating broader regional
trends. But, as the authors note, the post-Realignment
uptick in car thefts only brought the state’s auto theft
rate back to 2009 levels. Given Realignment’s modest
impact on property crimes, Lofstrom and Raphael’s
cost-benefit analysis leads them to conclude that
Realignment’s “benefits in terms of prison expenditure
savings outweigh the costs in terms of somewhat higher
property crimes.”

Figure 5. Property Crime Rates in the United States,
New York, and New Jersey, 1999-2012
4,000
3,500
3,000

United States

2,500

New Jersey

2,000

New York
2011

2012

2010

2009

2007

2008

2005

2006

2003

2004

2001

2002

2000

1999

1,500

Figure 6. Property Crime Rates in the United States
and California, 2006-2012
4,000
3,500
3,000

United States
California

2,500
2,000
1,500

2006 2007 2008 2009 2010

2011

2012

While California’s property crime rate decreased by
13% between 2006 and 2012, the state experienced a
6.8% uptick in 2012. That increase, the first since 2003,
brought the state’s property crime rate back to its 2009
level.
Lofstrom and Raphael’s analysis of California’s monthly
crime data at the county level presents evidence that
the small uptick in property crimes, particularly auto

POLICIES AND PRACTICES THAT
REDUCED PRISON POPULATIONS
IN THREE STATES
The declining prison populations of New York, New
Jersey, and California were not simply the result of falling
crime rates; rather, prisons were downsized through a
mix of policy and practice changes designed to reduce
admissions to prison and lengths of stay. While crime
rates were declining nationally during this period, other
states either experienced continued increases in their
prison populations, or only modest declines. Following
is a brief overview of the key reforms that produced
these outcomes.

New York
New York’s prison population peaked in 1999, with
72,896 prisoners. Mandatory penalties created by
the passage of Rockefeller Drug Laws and related
legislation, along with the intensification of street drug
enforcement in the 1980s and 1990s caused the state’s
prison population to balloon in size with lower-level
drug offenders.7 Other “get-tough” measures, such as
limitations on parole, also added to the state’s prison
population. Through a combination of changes in
policy and practice that largely affected enforcement
and sentencing for drug offenses in New York City, the
state’s 2012 prison population was 26% smaller than its
1999 peak.

6	 Id.
7	 Judith Greene & Marc Mauer, Downscaling Prisons: Lessons from Four States, The Sentencing Project (2010), http://www.sentencingproject.
org/doc/publications/publications/inc_DownscalingPrisons2010.pdf.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

5

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

Felony drug arrests began their sharp decline in New York
City beginning in 1999, following a widely-publicized
poll showing that the public had grown critical of
mandatory drug sentencing. 8 The decline in arrests was
driven largely by a shift in enforcement priorities in the
New York City Police Department. During the 1990s,
there were generally over 40,000 felony drug arrests per
year in New York City.9 By 2003, there were only 23,711
felony drug arrests, and that figure had fallen to 19,680
by 2012.10

since been replicated in a number of other jurisdictions.
Statewide, the proportion of people with felony drug
arrests who were sentenced to prison declined from
23.3% during the 1990s to 13.2% in 2012.13 Recently,
the city and state have also curbed prison admissions
through probation revocations by shortening probation
terms, thereby reducing unnecessary supervision of
low-risk individuals.14

Prison disposition rates also fell, with a growing number
of people with felony drug arrests being diverted to
alternative sentences, enabled by the growth in treatment
programs and their demonstrated efficacy. Initiatives such
as the Drug Treatment Alternative to Prison program,
pioneered by the Brooklyn District Attorney’s Office,
provided high quality substance abuse treatment services
to an otherwise prison-bound population, and have

New Jersey

The state also implemented a “Merit Time Program,”
signed into law by Governor George Pataki in 1997.
At the same time, misdemeanor drug arrests had This program enabled people serving prison sentences
increased dramatically in New York City – doubling for a nonviolent, non-sex crime to earn reductions in
between 1986 and 2008  – in part because of the their minimum term and become eligible for parole
broader growth in controversial police policies to target consideration sooner by completing educational,
misdemeanor crimes under “broken windows” and vocational, treatment, and service programs.
“stop and frisk” strategies.11 James Austin and Michael
Jacobson have argued that “NYPD’s shifting resources Finally, between 2003-2005 the state made substantial
toward misdemeanor arrests as part of the ‘broken revisions to the mandatory sentences stipulated by the
windows’ policing model contributed to the decrease Rockefeller Drug Laws, and in 2009 largely repealed the
in the felony arrests” (emphasis added).12 Given the provisions of the policy. Mandatory minimum terms
disproportionate influence of prison admissions from were eliminated or reduced in 2009, and the revisions
New York City, policing changes in that jurisdiction were made retroactive for persons still incarcerated
under the old law.
played a significant role in the state’s prison decline.

Through a combination of changes
in policy and practice, New York’s
2012 prison population was 26%
smaller than its 1999 peak.

New Jersey reached its peak prison population in 1999,
with 31,493 prisoners, and reduced its size by 26% by
2012. The state downscaled its prisons through both
front-end reforms affecting the number of admissions
and sentence lengths, and back-end reforms that
increased rates of parole and reduced parole revocations.
In 2001, the state settled a lawsuit accusing the Parole
Board of failing to meet deadlines required by state law
to prepare pre-parole reports and hold timely hearings.15
The parole board agreed to conduct more timely hearings
to prevent a future backlog as part of the settlement,
and it enhanced decision-making tools and supervision.
Parole approval rates rose dramatically, from 30.1% in

8	 Id.
9	 James Austin & Michael Jacobson, How New York City Reduced Mass Incarceration: A Model for Change? (Vera Institute of Justice 2012),
available at http://www.brennancenter.org/sites/default/files/publications/How_NYC_Reduced_Mass_Incarceration.pdf.
10	 Computerized Criminal History System: Adult Arrests 2004-2013, New York State Division of Criminal Justice Services (Feb. 25, 2014), http://
www.criminaljustice.ny.gov/crimnet/ojsa/arrests/Allcounties.pdf; Greene & Mauer, supra note 10.
11	 Austin & Jacobson, supra note 9.
12	 Id.
13	 Computerized Criminal History System: Adult Arrests Disposed, New York State Division of Criminal Justice Services (Apr. 22, 2014), http://
www.criminaljustice.ny.gov/crimnet/ojsa/dispos/nys.pdf; Computerized Criminal History System: Adult Arrests Disposed, New York State
Division of Criminal Justice Services (June 23, 2014) (unpublished) (on file with author).
14	 Vincent Schiraldi & Michael Jacobson, Could Less Be More When it Comes to Probation Supervision?, American City & County Viewpoints (June
4, 2014), http://americancityandcounty.com/blog/could-less-be-more-when-it-comes-probation-supervision.
15	 Greene & Mauer, supra note 7.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

6

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

1999 to 51.0% the following year, and have sustained
elevated rates since.16 The state also reduced the rate at
which people who violate the technical terms of their
parole are readmitted to prison.

In order to substantially reduce prison overcrowding,
the California Legislature enacted a “Realignment”
policy (Assembly Bill 109) in October, 2011.18 Key
elements of the legislation included: 1) individuals with
non-violent, non-sex-related, and non-serious (referred
to as “non-non-non”) current and prior convictions
could be incarcerated in county jails but no longer in
state prisons; 2) released prisoners with “non-nonnon” offenses would be supervised for a shorter period
of time and released to county probation supervision
instead of to state parole supervision, and; 3) individuals
who violated the technical terms of their probation or
parole (i.e., did not commit a new crime) could only be
sentenced to jail rather than prison, and for a shorter
length of time. Prior to Realignment, the state had also
passed legislation in 2009 to limit parole supervision for
low-risk individuals, with the intention of reducing the
number of people returning to prison for violating the
technical terms of their parole.19

New Jersey’s drug policy reforms also contributed to its
decarceration. State legislators established a sentencing
commission in 2004 that first investigated the state’s
“drug free zone law,” concluding that the law created
unwarranted racial disparity among people incarcerated
for drug offenses. The New Jersey Office of the
Attorney General issued guidelines to exempt the lowest
level of drug offenders from the law and increase judicial
discretion in sentencing. The state also passed Senate Bill
1866 to give judges discretion to sentence individuals
below the mandatory minimums of the school zone law,
and made this retroactive with a companion bill.

California
Since California reached its peak prison population in
2006, with 173,942 men and women, prisoner counts
have fallen every year. The rate of decline was small at
first: the size receded by 5.6% between 2006 and 2010.
But between 2010 and 2012, the prison population
decreased by 18.3%. This dramatic change was primarily
driven by the state’s efforts to comply with a court order
to reduce prison overcrowding.
In a significant 2011 decision, the U.S. Supreme Court in
Brown v. Plata found the provision of health care in the
California prison system to be constitutionally inadequate
due to the severe overcrowding in the system.17 Noting
that California prisons had been operating at around
200% of their design capacity for at least 11 years, the
Court ruled that the state was required to reduce this
figure to 137.5% of design capacity within two years.
This meant an additional reduction of 37,000 prisoners.
Through “Realignment,” described next, the state has
made significant reductions in its prison population but
has yet to reach the court-stipulated level.

Realignment has increased county jail populations while
reducing the state’s prison population. But the net effect
has been to reduce the total incarcerated population (in
jail and prison combined). The best estimates show that:
“Realignment increased the average daily jail population
by roughly one inmate for every three fewer offenders
going to state prison.”20

THE LIMITED RELATIONSHIP
BETWEEN INCARCERATION AND
CRIME
While it might seem intuitive that reducing prison
populations would negatively impact public safety – or
conversely, that declining crime rates would drive down
levels of incarceration – such a relationship has generally
been shown to be relatively weak. This is because just
as forces beyond crime rates affect incarceration levels,
forces beyond incarceration affect crime.

16	 Id.; New Jersey State Parole Board, 2013 Annual Report (2013), available at http://www.state.nj.us/parole/docs/reports/AnnualReport2013.
pdf.
17	 Brown v. Plata, 131 S. Ct. 1910 (2011).
18	 Joan Petersilia & Jessica Greenlick Snyder, Looking Past The Hype: 10 Questions Everyone Should Ask About California’s Prison Realignment, 5
California Journal of Politics and Policy 266 (2013), http://www.bscc.ca.gov/downloads/Looking_Past_The_Hype_Petersilia.pdf. This is part
of an ongoing series of analyses conducted and produced at Stanford Law School, see Stanford Law Sch., California Realignment, http://www.law.
stanford.edu/organizations/programs-and-centers/stanford-criminal-justice-center-scjc/california-realignment.
19	 CDCR Implements Public Safety Reforms to Parole Supervision, Expanded Incentive Credits for Inmates, Inside CDCR News (Jan. 25, 2010), http://
www.insidecdcr.ca.gov/2010/01/cdcr-implements-public-safety-reforms-to-parole-supervision-expanded-incentive-credits-for-inmates.
20	 Magnus Lofstrom & Steven Raphael, Impact of Realignment on County Jail Populations (Public Policy Institute of California 2013), available
at http://www.ppic.org/content/pubs/report/R_613MLR.pdf.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

7

the

Sentencing
Project
During the near four-decade continuous rise in
incarceration since 1972, crime rates increased in some
periods and declined in others. Most notably, during the
period 1984-1991 the incarceration rate increased by
more than 5% each year, reaching a peak increase of
12.8% in 1989 alone. Yet despite this significant rise in
the number of imprisoned individuals, crime rates also
rose substantially during this time.
Conversely, one might expect that trends in rates of
crime might affect the size of the prison population, but
there is little evidence for this assertion. As described
in the comprehensive 2014 report of the National
Research Council, The Growth of Incarceration in the United
States: Exploring Causes and Consequences:

“Over the four decades when incarceration
rates steadily rose, U.S crime rates showed
no clear trend: the rate of violent crime rose,
then fell, rose again, then declined sharply.
The best single proximate explanation of
the rise in incarceration is not rising crime
rates, but the policy choices made by
legislators to greatly increase the use of
imprisonment as a response to crime.”21

Policy Brief: Fewer Prisoners, Less Crime

Studies consistently find that
expediting prisoners’ release from
prison has no or minimal impact
on recidivism rates.
of linking crime to convicted individuals, but they are
also impacted by changing police and court practices
towards people under parole or probation supervision.
Yet studies quite consistently find that expediting
prisoners’ release from prison has no or a minimal
impact on recidivism rates. This pattern has been
true among federal prisoners whose sentences were
shortened,22 California prisoners re-sentenced under the
state’s reform to the “three strikes and you’re out” law,23
and California prisoners who avoided prison sentences
under Realignment.24
A number of factors are key to understanding why a
declining prison population might not produce higher
rates of crime. These include:

•	 The number of individuals released from prison in
a given year represents a relatively small proportion
Even to the extent that changes in crime rates might
of the overall “at risk” population of young males.
contribute to a rise or decline in prison populations, the
experience of the three states analyzed in this report •	 The crime-reducing effect of incarcerating certain
groups of offenders – particularly for drug offenses
demonstrates that such a relationship is very much a
and youth crimes, which are often committed in
secondary explanation. During the period that the prison
groups – is relatively modest since such offenders
population was declining in these states, crime rates
are frequently replaced on the streets by others
were declining not only in these states but in virtually
seeking to gain income.
all states. Yet despite a slowing of incarceration growth,
most states nevertheless experienced an increase in their
prison populations, and in some cases, very substantial •	 To some extent prison may produce criminogenic
effects; that is, longer stays in prison may lead
increases. Policy decisions, and not levels of crime, have
to higher rates of recidivism, in part due to the
been the main determinant of the scale of incarceration.
challenges of maintaining ties with family and
community. A 1999 meta-analysis of offender
Finally, many studies have asked how one approach
studies over four decades found that longer prison
to decarceration, shortening prison sentences, affects
sentences were associated with a modest increase in
recidivism. Data on recidivism rates have the advantage
21	 National Research Council, The Growth of Incarceration in the United States: Exploring Causes and Consequences 3 (The National
Academies Press 2014), available at http://www.nap.edu/openbook.php?record_id=18613.
22	 U.S. Sentencing Comm’n, Recidivism Among Offenders Receiving Retroactive Sentence Reductions: The 2007 Crack Cocaine
Amendment (2014), available at http://www.ussc.gov/sites/default/files/pdf/research-and-publications/research-projects-and-surveys/
miscellaneous/20140527_Recidivism_2007_Crack_Cocaine_Amendment.pdf.
23	 Proposition 36 Progress Report: Over 1,500 Prisoners Released Historically Low Recidivism Rate, Stanford Law School Three Strikes Project
& NAACP Legal Defense and Education Fund (2014), https://www.law.stanford.edu/sites/default/files/child-page/595365/doc/slspublic/
ThreeStrikesReport.pdf.
24	 Magnus Lofstrom, Steven Raphael, & Ryken Grattet, Is Public Safety Realignment Reducing Recidivism in California? (Public Policy
Institute of California 2014), available at http://www.ppic.org/content/pubs/report/R_614MLR.pdf.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

8

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

recidivism.25 So reductions in the length of prison Italy’s 2006 prison downsizing through commutation
terms may contribute to public safety, or at least proved short-lived.28 The large-scale parliamentary
produce fewer negative consequences.
commutation cut the country’s prison population by
one-third. But the legislation did not reform sentences
for future convictions and in fact enhanced prison terms
for released individuals who recidivated. Consequently,
the country’s incarceration rate returned to its prepardon high within two and half years.
The experience of the three states described in this report
is mirrored in other nations as well, with policymakers
and practitioners abroad enacting a range of measures
that have substantially reduced prison populations. The
Canadian province of Alberta significantly decreased
its prison population in the 1990s. The decline was not
produced by a government committed to a reduced use
The experience in New York, New Jersey, and California
of imprisonment, but rather a newly elected provincial
over more than a decade demonstrates that substantial
premier committed to balancing the budget through
reductions in prison populations can be achieved without
sharp cuts in government expenditures, including
adverse effects on public safety. It is also important
corrections. As a result there was a sharp decline in the
to note that prior to embarking on these population
number of people sentenced to provincial prisons for
reductions these states did not have excessive rates of
less serious crimes (persons convicted of serious crimes
incarceration by U.S. standards. In 1999, New Jersey and
continued to be sentenced to federal prisons). By closing
New York had incarceration rates of 384 and 400 per
two provincial prisons, diverting minor cases from the
100,000 population respectively, compared to a national
justice system, and expanding the use of alternative
rate for all states of 434 per 100,000. California’s rate of
sentencing, the province was able to reduce prison
475 per 100,000 when it began its reduction in 2006 was
admissions by 32% between 1993 and 1997. Researchers
just 7% above the national rate of 445 per 100,000.
have found that the decline was not due to changes in
reported crime and also that reduced incarceration “had In contrast, 14 states had rates of incarceration in
no obvious important negative impacts on offenders.”26 excess of 450 per 100,000 as of 2012. Given the
relatively modest relationship between crime rates and
In Europe, governments in Germany and Finland
incarceration rates we can therefore surmise that the
embarked on ambitious campaigns to reduce prison
degree of “excessive” imprisonment in these states is
populations in the 1960s and 1970s, the effects of
likely to be substantial. Such a finding helps to provide
which can still be seen today.27 In Germany officials
context for recent population reductions in states like
concluded that short prison terms served little crime
Texas. During 2012 the state experienced a reduction of
control purpose, but significantly affected offenders’
nearly 6,000 people in its prison population. This shift
relationships with families and communities, and so
built on bipartisan initiatives designed to reduce parole
substituted a range of community penalties instead.
revocations and enhance treatment programming. But
In Finland policymakers became concerned that
even with this recent population reduction the state’s
though their crime rates were similar to those of other
incarceration rate declined only to 601 per 100,000, a
Scandinavian nations their rate of incarceration was two
dramatic rate of imprisonment even by the standards
to three times higher. Through a series of sentencing
of a nation of mass incarceration. Such an observation
and programmatic shifts over a number of years the
does not diminish the significance of these changes or
country was able to reduce its imprisonment rate to
suggest that changing a political climate on criminal
become comparable to neighboring nations. By contrast,

INTERNATIONAL EXPERIENCE IN
PRISON POPULATION REDUCTION

POTENTIAL FOR SUBSTANTIAL
PRISON POPULATION
REDUCTIONS

25	 Paul Gendreau, Claire Goggin, & Francis T. Cullen, The Effects of Prison Sentences on Recidivism (Public Works and Government Services
Canada 1999), available at http://www.prisonpolicy.org/scans/e199912.htm.
26	 Cheryl Marie Webster & Anthony N. Doob, Penal Reform ‘Canadian style’: Fiscal Responsibility and Decarceration in Alberta, Canada, 16 Punishment
and Soc’y 3, 23 (2014).
27	 Michael Tonry, Thinking About Crime: Sense and Sensibility in American Penal Culture 29-34 (Oxford University Press 2004).
28	 Paolo Buonanno & Steven Raphael, Incarceration and Incapacitation: Evidence from the 2006 Italian Collective Pardon, 103 Am. Econ. Rev. 2437,
2442 (2013).

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

9

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

justice policy is a simple matter, but it does tell us that shown that older offenders have much lower rates of
there is potential for more substantial change in many recidivism than younger ones, and so such limitations on
release both lack compassion and are counterproductive
states.
in allocating public safety resources.
Further, we note that in the three states under review
continuous prison population reductions were achieved
during a mix of Democratic and Republican gubernatorial Address racial/ethnic Disparities in
terms. As can be seen nationally, increasingly issues of prison population
criminal justice reform are being viewed as bipartisan
As policymakers reduce prison populations in the coming
initiatives designed to produce better public safety
years, it will be important to assess how those initiatives
outcomes and reduced reliance on incarceration.
affect the racial composition of incarcerated persons.
Reductions in populations overall may or may not affect
existing disparities in imprisonment depending on the
strategies and criteria employed for such change.

EXPANDING THE AGENDA FOR
PRISON POPULATION REDUCTION

For example, in New York State the prison population
reduction of recent years has also produced a significant
decline in racial disparity among women.29 Most of this
decline has come about through a substantially reduced
number of persons serving sentences for drug offenses.
Focus on long-term prisoners
Since that population was about 90% African American
Much of the reform activity of recent years has centered or Hispanic, the declines almost inevitably led to a
around lower-level drug offenders, with increasing reduction in overall disparity as well.
support for diverting such people to treatment programs But in situations where policymakers restrict sentence
rather than prisons, as well as reducing excessively severe reductions for persons convicted of a serious offense
mandatory minimum sentencing provisions. While these and/or with a prior criminal record, population
initiatives have produced significant results in many cases, reductions may then exacerbate racial disparities. This
they represent only one aspect of a broader strategy is because African Americans in particular are more
for prison population reduction. This can be seen by likely to fall in these categories, either due to greater
examining the composition of prison populations today. involvement in offending and/or greater attention from
In seeking to take advantage of the changing climate for
reform, policymakers would be well advised to prioritize
several goals:

Among the population in state prisons nationally half law enforcement agencies. Unless there is a sustained
(53.5% as of 2011) were incarcerated for a violent focus and attention to this issue, racial disparities may be
offense and a declining proportion, now 16.6%, for a compounded even as overall populations decline.
drug offense (with the remainder having been convicted
of property and public order offenses). While persons
convicted of a violent offense clearly raise significant
Unless there is a sustained focus
concerns for public safety, in far too many cases such
concerns have led to excessively lengthy prison terms.
and attention to this issue, racial
Through policies and practices such as “life means life”
disparities may be compounded
and “no parole for violent offenders,” parole boards and
even as overall prison populations
governors in many states have adopted across-the-board
policies that fail to distinguish among individual offense
decline.
circumstances, accomplishments in prison, or degree
of risk to public safety. Research over many years has
29	 Marc Mauer, The Changing Racial Dynamics of Women’s Incarceration, The Sentencing Project (2013), http://sentencingproject.org/doc/
publications/rd_Changing%20Racial%20Dynamics%202013.pdf.

The Sentencing Project • 1705 DeSales Street NW, 8th Floor • Washington, D.C. 20036 • sentencingproject.org

10

the

Sentencing
Project

Policy Brief: Fewer Prisoners, Less Crime

Reinvest in communities

CONCLUSION

Mass incarceration has been produced by the combined
impact of a broad range of law enforcement, sentencing,
and corrections policies. But ultimately, it stems from
a substantial shift in the balance of approaches to
public safety in disadvantaged communities. Whereas
public safety is produced by a complex mix of family
and community support, education and economic
opportunity, and social interventions to address
individual deficits, as well as criminal justice responses,
over the past several decades policymakers have created
a severe imbalance in these approaches. Rather than
preventing or addressing crime through job creation,
mental health and substance abuse treatment and other
interventions, far too often arrest and incarceration have
become the preferred options.

At least in three states we now know that the prison
population can be reduced by about 25% with little or no
adverse effect on public safety. Individual circumstances
vary by state, but policymakers should explore the
reforms in New York, New Jersey, and California as a
guide for other states.

As a means of remedying this imbalance, savings achieved
through reductions in prison populations should be
targeted to those communities most heavily affected
by mass incarceration. As originally conceptualized in
Justice Reinvestment, targeting such savings to high
incarceration neighborhoods would both address the
harms created by mass incarceration as well as promote
public safety in a proactive manner.30

There is also no reason why a reduction of 25% should
be considered the maximum that might be achieved.
Even if every state and the federal government were
able to produce such reductions, that would still leave
the United States with an incarceration rate of more
than 500 per 100,000 population – a level 3-6 times that
of most industrialized nations.
In recent years a broader range of proposals has
emerged for how to reduce the prison population and
by various scales of decarceration. In a recent right/
left commentary Newt Gingrich and Van Jones describe
how they will “be working together to explore ways to
reduce the prison population substantially in the next
decade.”31 The experiences of New York, New Jersey,
and California demonstrate that it is possible to achieve
substantial reductions in mass incarceration without
compromising public safety.

30	 Susan B. Tucker & Eric Cadora, Ideas for an Open Society: Justice Reinvestment, Open Society Institute (November 2003) http://www.
opensocietyfoundations.org/sites/default/files/ideas_reinvestment.pdf.
31	 Newt Gingrich & Van Jones, Prison System is Failing America, CNN Opinion (May 22, 2014 5:23PM EDT), http://tyger.ac/posts/4036/frame.

This briefing paper was written by Marc Mauer, Executive Director
of The Sentencing Project, and Nazgol Ghandnoosh, Ph.D.,
Research Analyst.
1705 DeSales Street NW, 8th Floor
Washington, D.C. 20036
Tel: 202.628.0871
Fax: 202.628.1091
sentencingproject.org

The Sentencing Project works for a fair and effective U.S. criminal
justice system by promoting reforms in sentencing policy,
addressing unjust racial disparities and practices, and advocating
for alternatives to incarceration.
11