A REPORT
TO THE

ARIZONA LEGISLATURE

Performance Audit Division
Performance Audit

Department of
Corrections—

Prison Population Growth

September • 2010
REPORT NO. 10-08

Debra K. Davenport
Auditor General

The Auditor General is appointed by the Joint Legislative Audit Committee, a bipartisan committee composed of five senators
and five representatives. Her mission is to provide independent and impartial information and specific recommendations to
improve the operations of state and local government entities. To this end, she provides financial audits and accounting services
to the State and political subdivisions, investigates possible misuse of public monies, and conducts performance audits of
school districts, state agencies, and the programs they administer.

The Joint Legislative Audit Committee
Representative Judy Burges, Chair

Senator Thayer Verschoor, Vice Chair

Representative Tom Boone
Representative Cloves Campbell, Jr.
Representative Rich Crandall
Representative Kyrsten Sinema
Representative Kirk Adams (ex officio)

Senator John Huppenthal
Senator Richard Miranda
Senator Rebecca Rios
Senator Bob Burns (ex officio)

Audit Staff
Dale Chapman, Director and Contact Person
Jeremy Weber, Team Leader
Kerry Howell

Copies of the Auditor General’s reports are free.
You may request them by contacting us at:

Office of the Auditor General

2910 N. 44th Street, Suite 410 • Phoenix, AZ 85018 • (602) 553-0333
Additionally, many of our reports can be found in electronic format at:

www.azauditorv.gov

STATE OF ARIZONA
OFFICE OF THE

DEBRA K. DAVENPORT, CPA
AUDITOR GENERAL

MELANIE M. CHESNEY

AUDITOR GENERAL

DEPUTY AUDITOR GENERAL

September 30, 2010

Members of the Arizona Legislature
The Honorable Janice K. Brewer, Governor
Mr. Charles L. Ryan, Director
Department of Corrections
Transmitted herewith is a report of the Auditor General, A Performance Audit of the
Department of Corrections—Prison Population Growth. This report is in response to a
November 3, 2009, resolution of the Joint Legislative Audit Committee. The performance
audit was conducted as part of the sunset review process prescribed in Arizona Revised
Statutes §41-2951 et seq. I am also transmitting within this report a copy of the Report
Highlights for this audit to provide a quick summary for your convenience.
As outlined in its response, the Department of Corrections agrees with all of the findings
and plans to implement all of the recommendations directed at it.
My staff and I will be pleased to discuss or clarify items in the report.
This report will be released to the public on October 1, 2010.
Sincerely,

Debbie Davenport
Auditor General

Attachment

2910 NORTH 44 th STREET • SUITE 410 • PHOENIX, ARIZONA

85018 • (602) 553-0333 • FAX (602) 553-0051

Department of
Corrections—

Prison Population Growth

REPORT

HIGHLIGHTS
PERFORMANCE AUDIT
Our Conclusion
Arizona’s prison population
has grown significantly in
the last 30 years and is
expected to continue
growing. The State has
addressed this growth by
constructing new prison
facilities and contracting
with private prisons,
among other things. The
State has several options
for addressing this growth
in the future. The State
could continue to build
prisons and/or contract for
private prisons. The State
could also consider
diverting more nonviolent,
low-risk offenders from
prison or reducing their
time in prison, and
expanding the use of
nonprison alternatives for
these offenders, such as
home arrest. In addition,
using more nonprison
alternatives for parole
violators could also reduce
the prison population.

Significant growth in prison population and
spending
The State’s population has doubled in
about the last 30 years, but the State’s
prison population has increased tenfold,
from 3,377 inmates in June 1979 to 40,477
inmates in June 2010. Arizona’s prison
growth rate exceeded that of every other
western state between 2000 and 2008.
In 2008, 1 in every 170 Arizonans was in
prison, compared to 1 in 749 in 1980.

The growth in prison population has come
at a substantial cost. The Legislature has
appropriated nearly $949 million in State
General Fund monies to the Department
of Corrections (Department) for fiscal year
2011. This represents 11.2 percent of the
State General Fund budget and trails only
K-12 education and healthcare appropriations.

Expanding the prison system
To address prison population growth, the
State has constructed new prison
facilities, expanded existing prison
facilities by adding new and temporary
prison beds, and contracted with private
prisons for more beds. However, the
Department expects the prison population
to continue to increase, growing to nearly
50,000 inmates by 2016. Although under
revision as of September 2010, the
Department’s plan proposes to add
another 6,500 private prison beds at an
estimated cost of about $640.7 million
through 2017. The plan also calls for more
state construction to add another 2,000

beds at an estimated cost of $334.1
million through 2017.
Private prisons cost slightly more—
According to a 2009 department report,
the State paid more per inmate in private
prisons than for equivalent services in
state facilities. After adjusting costs to
make the expenditures comparable, the
State paid private prisons $55.89 for each
medium-custody inmate per day
compared to a daily cost of $48.13 per
medium-custody inmate in state facilities.
The State also paid private prisons slightly
more for each minimum-custody prisoner.

Alternatives to imprisonment

2010
September • Report No. 10 - 08

State laws largely determine how long an
offender is imprisoned. Before 1978,
judges had broad discretion in sentencing
defendants. However, Arizona’s
presumptive sentencing system requires
judges to impose a “presumptive”
sentence prescribed by statute for a given
offense. The sentence may be increased
or decreased based on mitigating and
aggravating factors.
Further, Arizona began adopting
mandatory sentences in 1978, that require

harsher penalties for certain offenders,
such as repeat or violent offenders.
Arizona also adopted “truth in sentencing”
in 1993, which abolished discretionary
parole and requires all inmates to serve at
least 85 percent of their sentences in
prison. Although truth in sentencing
requires inmates to serve more of their
sentences, other law changes shortened
sentences for some offenders, which has
contributed to some inmates serving less
time in prison.

Expanding diversion—The Legislature could
consider diverting some additional low-risk
offenders from prison to nonprison alternatives.
Statute requires some drug offenders to be
sentenced to probation and treatment instead of
prison, and this approach could be considered for
other nonviolent, low-risk offenders. According to a
2006 Arizona Supreme Court report, diverting 1,072
offenders to probation and treatment in fiscal year
2005 avoided an estimated $11.7 million in net
costs. Depending on how diversion is expanded,
sentencing law changes may be needed.
Expanding early release—Currently, some
nonviolent, low-risk offenders who make satisfactory
progress on their corrections plans, maintain
behavior, and meet other criteria may be released 3
months earlier than their sentences require. During
those 3 months, they receive treatment, transitional
housing, education, and other services. At the end
of the 3 months, they are placed on regular
community supervision. Most inmates successfully
complete the 3-month supervised release.
The Legislature could consider other alternatives for
expanding early release. This could include revising
the truth-in-sentencing laws to reduce the amount of
time nonviolent, low-risk offenders serve. Mississippi
reinstated parole for such offenders and, as a result,
has avoided prison costs of about $37 to $42 per
inmate per day. The Mississippi Department of
Corrections also reported that between January 31,
2009 and January 31, 2010, the state’s prison
population decreased by 1,360 inmates when an
increase of 1,000 inmates was expected. The
Legislature could also authorize earned time credits
for inmates, which reduce inmate sentences. These
credits can be earned for completing education,
vocational training, and/or treatment.
Nonprison alternatives such as drug treatment,
home arrest, and day reporting centers—Another
approach would be to expand drug treatment
alternatives beyond drug court. Some states,
notably Texas, have created secure facilities to
provide treatment to drug offenders. As a result,
Texas has reduced its prison costs.
Arizona law allows home arrest with electronic
monitoring for a small number of nonviolent, firsttime offenders. According to a Florida study, home
arrest costs a fraction of the cost of imprisonment.
Expanding this program in Arizona, which would

Department of
Corrections—

Prison Population Growth

require legislative action, could potentially reduce
prison costs.
Day reporting centers are nonprison alternatives
that blend high supervision levels with intensive
services and programming. A 2005 Georgia State
University study reported that offenders completing
a day reporting center program had a lower
recidivism rate than those not completing or not in
the program. Georgia Department of Corrections
officials reported that its day reporting centers cost
$16.50 daily per inmate as compared to $48 per
inmate, per day in prison. Although a 1999 study
showed that Maricopa County’s day reporting
center program was no more effective at reducing
recidivism for repeat DUI offenders than probation,
it was more cost-effective. Maricopa County ended
its day reporting center program in 2002.
Reducing parole violation revocations—Parolees
returned to prison on revocation typically serve
about 3 months, which costs about $1,222,
compared to $774 for one who remains in the
community. In some cases, the Department uses
graduated sanctions, such as reprimands and
increased supervision, before it revokes parole.
However, it lacks nonprison facilities to also use as
a graduated sanction. Other states use nonprison
facilities to house parole violators, including
residential treatment facilities, day reporting centers,
halfway houses, and assessment centers. Texas
uses secure facilities to provide treatment programs
and confine parole violators. Such facilities cost
about $35 to $41 per offender per day compared to
$47.50 per offender per day in a Texas prison.
Options—The Legislature could:
•

•

•

•

Continue to expand the prison system. If it
decides to expand, the Legislature should
consider directing the Department to further study
state costs for building and operating new prisons
compared to contracting with private prisons.
Consider diverting more nonviolent, low-risk offenders from prison and/or reducing the time they
serve.
Consider directing the Department and/or the
courts to further study the use and costs of nonprison alternatives for nonviolent, low-risk
offenders.
Consider expanding nonprison alternative sanctions for parole violators.

A copy of the full report is available at:
www.azauditor.gov
Contact person:
Dale Chapman (602) 553-0333

REPORT

HIGHLIGHTS
PERFORMANCE AUDIT
September 2010 • Report No. 10 - 08

TABLE OF CONTENTS
Introduction & Scope
Chapter 1: Arizona’s prison population and corrections spending
have grown significantly

1
3

Arizona’s prison population has grown considerably and
may continue growing

3

Arizona has expanded prison system to accommodate
growth

8

Prison population growth results from both policy and
social factors

13

Arizona has increased corrections spending to help
keep pace with growth

14

Chapter 2: Option 1—Expanding prison system
to address prison population growth

17

Continued expansion will require significant spending

17

State should further study cost-effectiveness of privately
operated prisons compared to state-operated prisons

19

Chapter 3: Option 2—Diverting more nonviolent, low-risk
offenders or reducing the time they serve to
address prison population growth

23

Arizona laws largely determine prison sentences

23

continued

Office of the Auditor General
page

v

TABLE OF CONTENTS
Chapter 3 (Continued)
Sentencing laws have affected State’s prison population

25

Other potential sentencing law changes have been under
study for many years

27

Options to divert nonviolent, low-risk offenders from prison or reduce
time they serve could be expanded

28

Permanent sentencing commission could review sentencing
policies and laws

34

Chapter 4: Option 3—Expanding use of nonprison

alternatives to slow or reverse prison population growth
Arizona uses probation as nonprison alternative

37

Nonprison alternatives could take several forms

40

State should further study expansion of nonprison alternatives,
including costs and needed legislative action

45

Chapter 5: Option 4—Reducing revocations from
parole violations

State of Arizona
page

vi

37

47

Most inmates serve part of sentence in community

47

Parole violations contribute to prison population growth

48

Expanding range of nonprison alternatives for parole violators
could help reduce prison population growth

49

Expanding alternatives for parole violations would require action

52

TABLE OF CONTENTS
Chapter 6: Recommendations for legislative and
department consideration

53

Appendix A: Data and methodology

a-i

Appendix B: References

b-i

Agency Response
Tables
1

2

3

4

5

Number and Percentage of Total Inmates by Crime Category
Calendar Years 1989, 1999, and 2009
(Unaudited)

5

Arizona Prison Capacity and Population
June 30, 2010
(Unaudited)

12

Estimated Prison Construction and Privatization
Costs for 8,500 Projected Beds
Fiscal Years 2012 through 2017
(Unaudited)

18

Comparison of Department Calculations
for Actual and Adjusted Private Prison
Per Capita Rates and State Prison Per Capita Costs
Fiscal Year 2009
(Unaudited)

20

Median Sentence Lengths for Admitted Offenders
by Offense Type, in Years
Calendar Years 1990, 1993, 1996, 1999,
2002, 2005, and 2008
(Unaudited)

27

continued

Office of the Auditor General
page

vii

TABLE OF CONTENTS
Figures
1

Arizona Prison Population
Fiscal Years 1979 through 2010
(Unaudited)

4

Comparison of Western States’ Average Annual
Prison Population Growth
December 31, 2000 to December 31, 2008
(Unaudited)

7

3

Arizona Prison Complex Locations
Fiscal Year 2010

9

4

Comparison of State General Fund Expenditures for Fiscal Year 1979
and Appropriations for Fiscal Year 2011
(Unaudited)

2

concluded

State of Arizona
page

viii

16

INTRODUCTION
& SCOPE
The Office of the Auditor General has conducted a performance audit of the
Department of Corrections (Department) pursuant to a November 3, 2009, resolution
of the Joint Legislative Audit Committee. This audit, conducted as part of the sunset
review process prescribed in Arizona Revised Statutes (A.R.S.) §41-2951 et seq.,
focuses on prison population growth and options for addressing this growth. The
Office of the Auditor General will issue two additional reports, one of which will
address the 12 statutory sunset factors.
This report discusses prison population growth in Arizona and the resultant growth
in state spending on corrections (see Chapter 1, pages 3 through 16). It offers
various options for legislative and department consideration to address this growth,
including:

•

Continuing to expand the prison system to address anticipated growth in the
prison population (see Chapter 2, pages 17 through 21);

•

Diverting more nonviolent, low-risk offenders from prison and/or reducing the
time they serve—alternatives that may require changes to the State’s sentencing
laws (see Chapter 3, pages 23 through 35);

•

Expanding the use of nonprison alternatives for nonviolent, low-risk offenders
(see Chapter 4, pages 37 through 46); and

•

Reducing admissions from parole revocations by expanding nonprison options
for responding to offenders who violate the conditions of their community
supervision (see Chapter 5, pages 47 through 52).

This audit was conducted in accordance with generally accepted government
auditing standards. Those standards require that we plan and perform the audit to
obtain sufficient, appropriate evidence to provide a reasonable basis for our findings
and conclusions based on our audit objectives. We believe that the evidence
obtained provides a reasonable basis for our findings and conclusions based on our
audit objectives.
The Auditor General and staff express appreciation to the Department’s Director and
staff for their cooperation and assistance throughout the audit.

Office of the Auditor General
page

1

State of Arizona
page

2

Chapter 1
Arizona’s prison population and corrections
spending have grown significantly
Arizona’s prison population has grown significantly, leading to increased spending
on corrections. Specifically, Arizona’s prison population has grown from 3,377
inmates in fiscal year 1979 to 40,477 inmates in fiscal year 2010 and is expected to
continue growing. Several factors have contributed to Arizona’s prison population
growth, including the State’s general population growth, sentencing policies, and
social factors such as crime and unemployment. As a result of the increase, the State
has expanded its prison system and appropriated a correspondingly greater portion
of State General Fund monies to corrections—11.2 percent in fiscal year 2011,
compared with expenditures of 4.3 percent in fiscal year 1979. This substantial
increase means that less funding is available for other priorities.

Arizona’s prison population has grown considerably and
may continue growing
Arizona has not only experienced significant prison population growth since fiscal
year 1979, but this growth is expected to continue into the future. The growth rate in
Arizona’s prison population has outpaced the growth rate in most other states and,
based on Department of Corrections (Department) and state budget office
projections, is projected to grow annually through 2016 to potentially 49,700 inmates.

Arizona’s prison population has grown by more than 37,000 inmates
since fiscal year 1979—As shown in Figure 1 (see page 4) and according
to department data, the State’s prison population grew from 3,377 inmates as of
June 30, 1979, to 40,477 inmates as of June 30, 2010—an average increase of
approximately 1,200 inmates per fiscal year. According to department data, annual
admissions to Arizona’s prison system have consistently exceeded releases.

From fiscal years 1979
through 2010, Arizona’s
prison population has
increased by
approximately 1,200
inmates annually.

Office of the Auditor General
page

3

Figure 1: Arizona Prison Population

Fiscal Years 1979 through 2010
(Unaudited)

45,000
Number of Inmates
Number of Inmates

40,000
35,000
30,000
25,000
20,000
15,000
10,000
5,000
0

Fiscal
YearYear
Fiscal
Source:

Auditor General staff analysis of the Department’s March 2010 Two-Year Prison Population Trend Report
and the ADC Institutional Capacity Committed Population report for June 30, 2010.

Although the State’s general population has also increased, the State’s prison
population has grown even faster. Specifically, according to Arizona Department of
Economic Security estimates, Arizona’s general population more than doubled
between fiscal years 1980 and 2008. During this same time, the State’s prison
population increased more than tenfold. As a result, while 1 in every 749 persons in
Arizona was in prison as of June 30, 1980, 1 in every 170 Arizonans was in prison as
of June 30, 2008.
In addition to this growth, the demographics of Arizona’s prison population have
changed. Specifically, auditors’ analysis of department annual reports and data
highlighted the following changes in the prison population:

•

Drug offenders accounted
for 20.5 percent of the
prison population as of
December 31, 2009.

State of Arizona
page

4

Various categories of offenders have increased—Although Arizona’s prison
population consists of inmates sentenced to prison for a wide variety of crimes,
as shown in Table 1 (see page 5), certain categories of criminal offense have
increased as a percentage of the prison population. For example, the number
of imprisoned drug offenders increased from 1,975, or 15.6 percent of the
prison population as of June 30, 1989, to 8,271, or 20.5 percent of the prison
population as of December 31, 2009. The number of persons imprisoned for
assaults has also increased, from 989, or 7.8 percent of the prison population
as of June 30, 1989, to 4,875, or 12.1 percent of the prison population as of
December 31, 2009.

Table 1:

Number and Percentage of Total Inmates by Crime Category
Calendar Years 1989, 1999, and 2009
(Unaudited)
June 30, 1989
Number

Crimes Against Persons
Homicide
Kidnapping
Sexual Assault
Robbery
Assault
Property Crimes
Arson
Burglary
Theft/Larceny
Forgery-Fraud
Other1
Morals-Decency Crimes
Drugs
Sex Offenders
Other2

Number

Percent

December 31, 2009
Number

Percent

1,144
276
785
1,170
989
4,364

9.1%
2.2
6.2
9.3
7.8
34.54

2,090
443
1,460
2,014
3,118
9,125

8.1%
1.7
5.7
7.8
12.1
35.34

3,406
1,232
2,151
3,454
4,875
15,118

50
1,899
1,565
459
507

0.4
15.0
12.4
3.6
4.0

69
2,395
2,404
1,000
634

0.3
9.3
9.3
3.9
2.5

88
2,948
4,477
1,610
167

0.2
7.3
11.1
4.0
0.4

4,480

35.44

6,502

25.24

9,290

23.0

1,975
641
99
2,715

15.6
5.1
0.8
21.5

5,575
1,286
197
7,058

21.6
5.0
0.8
27.34

8,271
1,906
427
10,604

20.5
4.7
1.1
26.3

621
280
901

4.9
2.2
7.1

1,238
808
2,046

4.8
3.1
7.9

2,135
2,803
4,938

5.3
6.9
12.2

180

1.4

1,103

4.3

390

1.0

Public Order Crimes
DUI
Other3

Miscellaneous Crimes
Total Crimes

Percent

June 30, 1999

12,640

100.0%4

25,834

100.0%

40,340

8.4%
3.1
5.3
8.6
12.1
37.5

100.0%

1 Other Property Crimes can include criminal damage, criminal littering or pollution, and unlawful failure to return rented property.

2 Other Morals-Decency Crimes can include domestic violence, child or adult abuse, prostitution, and public display of obscene
materials.
3 Other Public Order Crimes can include disorderly conduct, stalking, rioting, smuggling, and weapons offenses.
4 Amounts do not total due to rounding.
Source: Auditor General staff analysis of department annual reports for fiscal years 1989 and 1999 (data as of June 30) and prison
population data obtained from the Department’s Adult Inmate Management System as of December 31, 2009.

Office of the Auditor General
page

5

•

Violent and nonviolent offenders—The percentage of prison admissions for
violent offenses has remained at about 24 percent (see textbox for definitions of
violent and nonviolent offenses). However, the percentage of inmates
incarcerated for violent crimes has increased from 41 percent as of June 30,
1995, to approximately 49 percent as of December 31, 2009. An additional 10.6
percent were incarcerated for nonviolent crimes but had at least one prior violent
offense.

Violent and Nonviolent Offenses
Violent—Arizona Revised Statutes (A.R.S.) §13-901.03(B) defines violent
offenses as offenses that include any criminal act that results in death or
physical injury, or any criminal use of a deadly weapon or dangerous
instrument. For purposes of auditors’ analysis, the following crimes,
among others, were defined as violent: assault, homicide, kidnapping,
robbery, sex offenses (except indecent exposure and voyeurism), and
weapons offenses.
Nonviolent—The Bureau of Justice Statistics defines nonviolent offenses
as property, drug, and public order offenses that do not involve a threat
of harm or an actual attack upon a victim. For purposes of auditors’
analysis, the following crimes, among others, were defined as nonviolent:
drug crimes, driving under the influence, forgery and fraud, property
damage, and theft.
Source:

Auditor General staff analysis of A.R.S. §13-901.03(B), the Bureau of Justice Statistics Web site,
and department data.

Arizona’s prison population has grown faster than most states’
prison populations—Arizona’s prison population has grown at a faster rate

Many states experienced
a decline in their prison
populations in 2009.

than most other states’ since at least 2000. According to a 2010 federal Bureau of
Justice Statistics report, Arizona ranked third nation-wide and, as illustrated in
Figure 2 (see page 7), first among western states in its average annual prison
population growth rate between 2000 and 2008.1,2 Further, this report indicated
that prison populations in many states decreased in 2009. Specifically, 24 states,
including 6 western states, experienced a decline in their prison populations,
resulting in a 0.2 percent nation-wide decline in the number of state prisoners.
According to the report, Arizona’s prison population grew by an average annual
increase of 5.1 percent between 2000 and 2008, but grew by just 2.6 percent
between 2008 and 2009. However, Arizona’s percentage increase in 2009 was still
higher than most other states’, including all western states except Alaska.

1 See West, 2010
2 According to report data, Arizona experienced the largest average annual growth in its prison population among

western states between December 31, 2000 and December 31, 2008, in terms of both actual and percentage growth.

State of Arizona
page

6

Figure 2: Comparison of Western States’ Average Annual

Prison Population Growth
31, 2000
December
31,Average
2008 Annual Prison
Figure 3:December
Comparison
of to
Western
States’
(Unaudited)
Population Growth
December 31, 2000, to December 31, 2008
(Unaudited)
Arizona

Colorado
Arizona

4.1%

Oregon
Colorado

5.1%

3.7%
4.1%

Idaho
Oregon

3.5%
3.7%

Nevada
Idaho

3.0%
3.5%

Wyoming
Nevada

2.7%
3.0%

Alaska1
Wyoming

1

2.7%
2.3%

Alaska1
New
Mexico

2.3%
2.3%

New
Mexico
Washington

2.3%
2.3%

1

1

Washington
Hawaii1

2.3%
2.1%

1

Hawaii1
Utah

2.1%
1.9%

Montana
California

5.1%

1.7%
0.8%

1 Numbers for Alaska and Hawaii include total jail and prison populations because they form one integrated

system in these states.

Source: Auditor General staff analysis of state prison population data reported in the Bureau of Justice Statistics’
Prisoners at Yearend 2009—Advanced Counts.

A 2010 Pew Center on the States (Pew) report described the reasons for many
states’ 2009 prison population decline.1,2 According to the report, an important
contributor to prison population declines nation-wide was that “states began to
realize they could effectively reduce their prison populations and save public
funds, without sacrificing public safety. In the past few years, several states,
including those with the largest population declines, have enacted reforms
designed to get taxpayers a better return on their public safety dollars.” However,
the report cautioned that it is too soon to say whether the 2009 decline will be
temporary or the beginning of a downward trend.

Arizona’s prison population expected to grow—Both

state budget
offices—the Joint Legislative Budget Committee (JLBC) and the Governor’s Office
of Strategic Planning and Budgeting (OSPB)—and the Department have projected
that Arizona’s prison population will continue growing based upon historical
growth trends. According to OSPB’s General Fund Executive Budgets and JLBC’s

1 Pew, 2010
2 The Pew Center on the States is a division of the Pew Charitable Trusts, a nonprofit organization that seeks to identify

and advance solutions to critical issues facing states. According to the Pew report, 26 states experienced a decline in
their prison populations in 2009.

Office of the Auditor General
page

7

Appropriations Reports for fiscal year 2010, the prison population was expected to
grow by 150 or 151 inmates per month, respectively, in fiscal year 2010. The OSPB
and JLBC reports projected growth of 114 or 126 inmates per month, respectively,
in fiscal year 2011. The Department previously projected growth of 151 inmates
per month for November 2009 through December 2016, but it has revised its
projections downward to reflect the OSPB projected growth of 114 inmates per
month beginning in August 2010. However, none of these projections predicted
the significant slowing in prison population growth for fiscal year 2010. According
to department records, the State’s prison population experienced a net increase
of only 65 inmates in fiscal year 2010, growth that fell substantially below
projections. Department staff reported that this less-than-expected increase in the
prison population is based on decreased prison admissions from Maricopa
County, although they have been unable to determine the exact cause for this
decrease. The Department is continuing to research this unexpected small
increase in the State’s prison population for fiscal year 2010 to determine whether
this was a 1-year anomaly or whether it should revise its longer-term growth
forecasts. If the growth that occurred in fiscal year 2010 is an anomaly and the
previously projected growth at 114 inmates per month resumes, this would result
in a state prison population of nearly 49,700 inmates by December 31, 2016.

Arizona’s prison
population may grow to
nearly 49,700 inmates by
the end of 2016.

Arizona has expanded prison system to accommodate
growth
The State has significantly expanded its prison system to accommodate the growth
in the prison population. As of June 30, 2010, the State operated 10 prison complexes
with a total capacity of more than 33,400 beds and contracted with 5 in-state private
prisons and 1 out-of-state private prison for 7,440 additional beds (see Figure 3,
page 9, for a map of the prison locations in Arizona). These 40,840 beds represent
nearly a nine-fold increase from the approximately 4,730 beds the Department
operated prior to 1980.
The State has expanded the prison system in the following ways:

•

Arizona has constructed several new prison complexes adding thousands
of beds—According to information provided by Department of Corrections and
Arizona Department of Administration (ADOA) staff, the State added six new
prison complexes and expanded its four existing prison complexes between
1981 and 2004. These new and expanded complexes cost the State at least
$561 million to build and added more than 22,100 beds to the state system.1
The Department gained an additional 300 beds in February 2010 when the
Eagle Point facility, which is part of the Lewis prison complex and formerly
housed juveniles under the jurisdiction of the Arizona Department of Juvenile

1 The $561 million figure does not include the costs to build six units within the prison complexes because these costs

were not available from ADOA.

State of Arizona
page

8

Figure 3: Arizona Prison Complex Locations
Fiscal Year 2010

Page

Grand Canyon
National Park
Window Rock

Kingman

Winslow
Flagstaff

Lake Havasu City

Prescott
Show Low

Phoenix
West

Alpine

Eyman
Florence

Phoenix

Perryville

Central Arizona
Correctional Facility

Lewis
Gila Bend

Florence
West
Safford

Yuma
Marana
Tucson

Wilcox

Bensen

Legend
State-Operated Prisons
Private Prisons

Douglas
Nogalas

Source: Auditor General staff depiction of information from Department’s Web site.

Office of the Auditor General
page

9

Corrections, was transferred to its control. The Department reported that it spent
more than $107,100 preparing the Eagle Point facility for its use.
According to the ADOA, construction was completed in early 2010 on another
4,000 beds in new buildings at existing prison complexes as authorized by Laws
2007, Ch. 261. The Department received funding in fiscal year 2011 to begin
filling these beds. According to the ADOA, this expansion cost almost $194
million. Although the new buildings were designed to house 4,000 inmates,
according to department officials, the necessary infrastructure was included to
accommodate an additional 1,000 beds should they be needed.

•
The Department
contracted for 5,680 beds
in 5 private prisons as of
June 30, 2010.

Arizona has contracted for thousands of private prison beds—The
Department began contracting for beds in private prisons in fiscal year 1994
and, as of June 30, 2010, contracted for a total of 5,680 beds in 5 private prisons
in Arizona. The State plans to expand its use of in-state private prison beds.
Specifically, Laws 2009, 3rd S.S., Ch. 6, §37, requires the Department to contract
for an additional 5,000 private prison beds. Although the Department had
issued a request for proposals for these beds, according to department officials,
as of September 2010, the request for proposals had been canceled and was
in the process of being revised for re-issuance.
The Department has also used private facilities in other states, but this policy is
changing. As of December 31, 2009, the Department had contracted for nearly
4,500 beds at three privately operated facilities in Colorado and Oklahoma.
However, the State has decided to discontinue out-of-state prison contracts. As
a result, the Department began moving prisoners housed at the out-of-state
facilities back into the State in March 2010. As of June 30, 2010, there were still
1,765 inmates housed at a private facility in Oklahoma, but the Department
plans to return all of these prisoners to in-state facilities by November 2010.
From fiscal year 1993 through fiscal year 2010, the Department reported that it
spent more than $731.5 million to contract for private prison beds.1

•

Department has added temporary beds to existing prison facilities—
Despite the extensive expansion of Arizona’s prison system, the State has been
unable to keep pace with prison population growth. According to a department
official, the Department first used temporary beds—that is, beds in excess of
what a facility is designed or rated to house—in July 1982 when the prison
population exceeded the rated bed capacity. The Department has added these
temporary beds by double bunking occupied single cells, adding more beds to
occupied dormitories, and adding beds in prison spaces not designed to house
inmates. For example, at the Eyman prison complex, the Department has
added double bunks to maximum security cells originally designed for single
occupancy and has expanded lower custody units that were designed for 24

1 This total does not include private prison contract costs for fiscal years 2002 through 2004 and includes only a part of

those costs for fiscal year 2005 because the JLBC Appropriations Reports that the Department used to compile this
information did not separately account for these costs in those years.

State of Arizona
page

10

inmates to hold 48 beds. At the Perryville prison complex, which houses most of
the State’s female inmates, the Department temporarily converted serving
kitchens to 56-bed dorms and former programming rooms into cells that hold 8
to 10 inmates in bunk beds. According to department officials, alternative
spaces are only occasionally used for temporary beds and on an emergency
basis, although they have been used for extended periods of time.
Department officials also reported that, although the use of temporary beds is
less costly than constructing new ones, it carries several disadvantages.
According to the Department, the primary costs associated with temporary beds
are for providing food and healthcare to the inmates. However, adding temporary
beds increases capacity beyond what industry standards have deemed safe.1
This increase, in turn, can create overcrowded conditions at prisons, which can
lead to additional stress for staff, inmates, and the physical plant facilities. For
instance, according to department officials, the kitchen and restroom facilities
and the state prisons’ electricity, water, and sewer systems were built to
accommodate only these prisons’ design capacities. When capacity is
exceeded, problems can arise with these systems. For example, according to
the Department, the growing inmate population at the Perryville prison complex
increased demand for food service from the complex’s central kitchen, which
was originally designed in 1981 to supply 3,600 meals per day. A department
official explained that, as of July 2010, the Perryville prison’s central kitchen was
producing 10,200 meals per day for the Perryville complex alone and an
additional 2,100 meals for the Phoenix prison complex. The official further stated
that the prolonged overuse of Perryville’s kitchen has resulted in the need for
significant upgrades to the facility’s physical structure as well as the plumbing,
electricity, and other equipment in order to be in compliance with building and
health codes.

Use of temporary beds
can stress the prison
system.

Table 2 (see page 12) shows the number of beds each prison complex was rated to
accommodate, the total bed capacity, and the inmate population as of June 30,
2010, at both state-operated and privately operated facilities. For example, the
Florence prison complex was rated to hold 3,692 inmates, but had a total operating
capacity of 4,439 inmates as of June 30, 2010. The total operating capacity includes
temporary beds. However, as of June 30, 2010, the Florence prison complex had
4,495 inmates.2

1 The American Correctional Association (ACA) sets prison capacity design standards to safeguard the life, health, and

safety of staff and offenders. Although the State’s prisons are not certified by the ACA, the Department reported that it
builds prisons with these standards in mind.

2 As indicated in Table 2 (see page 12), the Florence prison complex inmate population as of June 30, 2010, included

inmates placed in special use beds that are not reflected in the prison’s total operating capacity, but that the Department
uses for temporary placements due to sickness and other reasons.
Office of the Auditor General
page

11

Table 2:

Arizona Prison Capacity and Population
June 30, 2010
(Unaudited)
Rated
Capacity

State Prisons
Douglas
Eyman
Florence
Lewis
Perryville4
Phoenix4
Safford
Tucson
Winslow
Yuma
State Bed Totals
Private Prisons
Central Arizona
Correctional Facility
Florence West
Great Plains-Cornel, OK
Kingman
Marana
Phoenix West
Private Bed Totals
Total Prison System

Total
Operating
Capacity1

Inmate
Population2

% Total
Operating
Capacity
Reached3

2,055
4,588
3,692
4,952
3,002
563
1,486
3,932
1,754
2,448

2,684
5,252
4,439
5,356
3,958
714
1,934
4,572
1,890
2,604

2,663
5,530
4,495
5,256
3,454
539
1,934
4,608
1,896
2,695

99%
105
101
98
87
75
100
101
100
103

28,472

33,403

33,070

99

1,000

1,280

1,272

99

600
1,760
2,542
450
400

750
1,760
2,650
500
500

743
1,765
2,650
490
487

99
100
100
98
97

6,752

7,440

7,407

100

35,224

40,843

40,477

99

1 The total operating capacity includes temporary beds, but does not include special use beds, which are

beds where inmates stay temporarily for various reasons such as detention or sickness. The Department
does not include these beds in its operating capacity because there must be general population beds
available for these inmates when released from detention or the sick ward.

2 The inmate population as of June 30, 2010, includes inmates housed in special use beds, as described

in footnote 1, as well as inmates who are under the jurisdiction of the Department but were outside of the
prisons because of reasons such as a court date or hospital stay.

3 For prisons where the total inmate population exceeds total operating capacity, some inmates were in

special use beds.

4 The operating capacities of the Perryville and Phoenix prisons include beds used during the intake process.

All prison admissions must pass through one of these two facilities before being assigned a permanent
bed. Because of these intake beds, the percentage of capacity reached appears lower than actual
conditions at the facility.

Source: Auditor General staff analysis of the Department’s ADC Institutional Capacity Committed Population
report for June 30, 2010.

State of Arizona
page

12

Prison population growth results from both policy and
social factors
Various factors contribute to growth in prison populations. According to an August
2005 Vera Institute of Justice report (Vera report) that studied the impact of state-level
sentencing and corrections policies between 1975 and 2002, these policies and
social factors affect states’ incarceration rates.1,2 The incarceration rate is the number
of inmates per 100,000 residents. An increase in the incarceration rate would indicate
a growing prison population. The Vera report identified a number of social factors
associated with the size or growth of incarceration rates. For example, states with
larger minority populations, more state revenue per capita, a higher rate of arrests for
drug offenses, and more law enforcement personnel per capita had higher
incarceration rates, while states with higher personal income per capita and more
generous welfare benefits had lower incarceration rates. States with higher property
crime rates experienced larger growth in incarceration rates. In addition, the Vera
report found that higher levels of unemployment, greater increases in unemployment,
higher levels of income inequality, and larger youth populations were also associated
with larger growth in incarceration rates, but the size of minority populations was not
related to growth. The Vera report also found that some state sentencing policies can
affect incarceration rates. For example, states with more provisions for increasing
sentences for drug offenses (such as drug sales near a school, offenses involving
minors, or weapon use), had higher incarceration rates, as did states with more
mandatory sentencing laws (laws requiring courts to impose incarceration for a
specific offense and/or a longer prison term). See Chapter 3 (pages 23 through 35)
for more information on Arizona’s sentencing laws and their effect on the State’s
incarceration rate and prison population.
Further, Arizona’s incarceration rate has continued to increase despite the fact that
Arizona’s crime rate has generally declined since the mid-1990s. According to crime
rate data compiled by the Bureau of Justice Statistics, while Arizona had 1 crime for
every 12 residents in 1995, the figure had dropped to 1 for every 22 residents in 2008.
A 2010 Arizona Prosecuting Attorneys’ Advisory Council report suggested the drop
in the crime rate could be due to the State’s increased imprisonment rate.3 However,
literature auditors reviewed indicates that the effect of incarceration on crime is
limited compared to the combined effect of other factors (such as increased law
enforcement, employment, and education) and diminishes as prison populations
grow.4 In addition, although Arizona’s crime rate has dropped, the State has one of
the highest reported crime rates in the nation despite also having one of the highest
incarceration rates (see textbox, page 14).

Arizona’s crime rate has
generally declined since
the mid-1990s.

1 See Stemen, Rengifo, & Wison, 2005
2 The Vera Institute of Justice (Vera) is a nonpartisan, nonprofit center for justice policy and practice. According to its

Web site, Vera combines expertise in research, demonstration projects, and technical assistance to help leaders in
government and civil society improve the systems people rely on for justice and safety.

3 See Fischer, 2010
4 See Stemen, 2007; Liedka, Piehl, & Useem, 2006

Office of the Auditor General
page

13

Arizona’s incarceration rate
A 2009 Pew report reviewed state incarceration rates for 2007
and reported that Arizona had the highest incarceration rate
among western states and tied with South Carolina to rank
ninth nation-wide, behind the District of Columbia and seven
other states.
Source:

Auditor General staff review of the Pew Center on the State’s 1 in 31: The Long
Reach of American Corrections report.

According to auditors’ analysis of Federal Bureau of investigation (FBI) data, Arizona
had one of the top five highest reported crime rates among all 50 states, the District
of Columbia, and Puerto Rico in 2006 through 2008.1

Arizona has increased corrections spending to help keep
pace with growth
Regardless of the reasons for the increased prison population, the growth has led to
substantially increased corrections spending, which accounted for $1 in every $8.77
of State General Fund estimated operating expenditures in fiscal year 2010. To
accommodate the growth in Arizona’s prison population, the Legislature has
significantly increased State General Fund spending on corrections operations. In
fact, the Legislature has appropriated nearly $949 million in State General Fund
monies to the Department for fiscal year 2011, a significant increase from the $41.4
million spent in fiscal year 1979 for corrections. However, the Department has
implemented several cost-saving measures to keep per-inmate costs low, helping to
avoid even greater correctional expenses.

Department operations compose 11.2 percent of State General
Fund appropriations—The Legislature has significantly increased the
For fiscal year 2011, the
Legislature has
appropriated nearly $949
million in State General
Fund monies to the
Department.

amount of State General Fund monies it spends on department operations.
According to JLBC data, State General Fund corrections operating expenditures
totaled more than $41.4 million in fiscal year 1979. For fiscal year 2011, the
Legislature has appropriated nearly $949 million in State General Fund monies to
the Department, including $58 million in startup and operational costs, which will
cover the first year of operations for the 4,000 new inmate beds the State added
in 2010.
1 Auditors compared estimated state crime rates reported by the FBI in its annual Crime in the United States publications

for 2006 through 2008. The FBI develops the estimated crime rates based on crimes reported as part of the FBI’s
Uniform Crime Reporting (UCR) Program. The UCR Program collects crime statistics on eight crime categories: murder
and nonnegligent manslaughter, forcible rape, robbery, aggravated assault, burglary, larceny-theft, motor vehicle theft,
and arson (however, arson is not included in the estimated state crime rates for 2006 through 2008 because of
insufficient data). Arizona ranked second after the District of Columbia in 2006, third in 2007, and fifth in 2008.

State of Arizona
page

14

According to JLBC reports and as shown in Figure 4 (see page 16), this increase
has meant that spending on the Department constitutes a greater portion of
available State General Fund monies, thus impacting monies that are available for
other state priorities. Specifically, the Department’s fiscal year 2011 State General
Fund appropriation accounts for 11.2 percent of all State General Fund
appropriations for the fiscal year, which is more than double the 4.3 percent of
State General Fund operating monies spent on corrections in fiscal year 1979. For
fiscal year 2011, corrections will be the third largest State General Fund operating
expense, trailing only K-12 education and health. By contrast, university spending
has decreased from nearly 19.1 percent of State General Fund operating
expenditures in fiscal year 1979 to only about 10.5 percent of State General Fund
appropriations in fiscal year 2011. In fiscal year 2011, for every State General Fund
operating dollar appropriated to the Department, $0.94 was appropriated to the
universities.

Department has kept per-inmate daily costs low—Even

though state
spending on the Department’s operations has increased significantly, the
Department has taken steps to keep the per-inmate daily cost (per capita rate) low.
In fact, although the per capita rate increased from $42.46 per day in fiscal year
1986 (the earliest year data was available) to $64.96 per day in fiscal year 2009, it
actually decreased to $32.98 per day when adjusted for inflation. A department
letter prepared in response to a request for information from the Commission on
Privatization and Efficiency—a commission the Governor established to identify
state services and agencies whose functions can be eliminated, consolidated,
streamlined, or outsourced to achieve greater operational efficiency—noted
several ways in which the Department has kept per capita rates low. According to
this letter, the Department has contracted for services (such as food, health, and
work-based education) with private organizations and community colleges;
downsized administrative office staff; placed responsibility for more costs on the
inmates; taken advantage of volunteer support; replaced typical mattresses with
ones made from recycled materials; and used inmate labor and inmate-produced
products whenever feasible, among numerous other efficiencies. The Department’s
ability to keep its per-inmate costs low has helped the State to avoid even higher
spending on department operations in light of the significant prison population
growth.

Prison per capita costs
have decreased when
adjusted for inflation.

Office of the Auditor General
page

15

Figure 4: Comparison of State General Fund Expenditures for Fiscal Year 1979
and Appropriations for Fiscal Year 2011
(Unaudited)
Fiscal Year 1979 Expenditures
4.30%
4.84%

19.06%
45.86%

Department of
1
Corrections11
Total Health Services
2
(includes AHCCCS) 2
Universities/Regents
All Other Agencies

25.93%

Department of Education
(K-12)

Fiscal Year 2011 Appropriations
10.50%
11.19%
41.19%

15.69%

21.42%
1 In fiscal year 1979, incarcerated juveniles not convicted as adults were housed under the Department of Corrections; in

fiscal year 2011, they are housed in the Arizona Department of Juvenile Corrections. The adult corrections portion of fiscal
year 1979 was less than 4.3 percent of the State General Fund.

2 The Arizona Health Care Cost Containment System (AHCCCS) is the State’s Medicaid agency.

Source:

State of Arizona
page

16

Auditor General staff analysis of the JLBC General Fund Operating Budget Spending Fiscal Years 1979 -2011 report and
the fiscal year 2011 Appropriations Report.

Chapter 2
Option 1—Expanding prison system to address
prison population growth
One option to address Arizona’s prison population growth is to continue expanding
the prison system. Specifically, the Legislature could consider constructing new
prison facilities and/or contracting for more private beds. Based on the Department
of Corrections’ (Department) proposed plan for expanding the prison system to meet
expected growth using a combination of state and private facilities, this option could
cost an estimated $975 million between fiscal years 2012 and 2017, and actual costs
could be higher. If the Legislature decides to continue expanding the prison system,
it should consider directing the Department to further study and analyze the costs for
the State to build and operate prison facilities compared to contracting with private
prisons to determine which option would be more cost-effective while still ensuring
public safety.

Continued expansion will require significant spending
As discussed in Chapter 1, the Department has projected that the State’s prison
population could reach nearly 50,000 inmates by December 31, 2016, based on a
growth rate of 114 inmates per month (see Chapter 1, page 8). Based on the
projected growth, the Department estimates that the State will need 8,500 new
beds—in addition to the 4,000 new beds that became operational in fiscal year
2011—and has developed a plan to meet this demand. The proposed plan
recommends adding both state-operated and private beds because statute requires
the Department to consider contracting for private prisons before expanding or
constructing new minimum- or medium-security prison facilities for certain offenders.
As illustrated in Table 3 (see page 18), the plan could cost approximately $975 million
for construction and operating costs between fiscal years 2012 and 2017 and
includes the following:

•

Expanding the prison
system to meet expected
prison population growth
could cost approximately
$975 million between fiscal
years 2012 and 2017.

Private prison beds—The Department’s plan recommends an additional 6,500
private prison beds for minimum- and medium-custody level male inmates. This

Office of the Auditor General
page

17

Table 3:

Estimated Prison Construction and Privatization
Costs for 8,500 Projected Beds
Fiscal Years 2012 through 2017
(Unaudited)

(
Custody
Level

Ownership

)

Gender

Beds

Construction
Costs1

Start-Up
Costs2

Operating
Costs
2012-20173

Total Costs

Units to be Built in Fiscal Year 2012
Minimum
Medium

Private
Private

Male
Male

2,000

N/A4

N/A4

$213,422,880

$213,422,880

3,000

4

4

371,210,880

371,210,880

N/A

N/A

Units to be Built in Fiscal Year 2013
Minimum

State

Female

Maximum

State

Male

500

$23,500,000

$1,762,975

48,318,900

73,581,875

1,000

93,000,000

3,525,950

125,348,589

221,874,539

500

N/A4

N/A4

24,182,593

24,182,593

500

23,500,000

1,762,975

13,421,100

38,684,075

1,000

4

N/A

4

31,902,920

31,902,920

8,500

$140,000,000

$7,051,900

$827,807,862

$974,859,762

Units to be Built in Fiscal Year 2015
Medium

Private

Male

Units to be Built in Fiscal Year 2016
Minimum
Medium

State
Private

Totals

Female
Male

N/A

1 Amounts estimated by the Arizona Department of Administration (ADOA) are based on previous construction costs for adding similar types

of units at existing prison complexes and construction-specific inflation rates as of the second quarter of 2010. According to ADOA, the
estimates also include some infrastructure improvements beyond what exists at existing complexes. Actual costs may differ based on the
sites chosen for each facility and the utility and infrastructure costs. No land costs are included.

2 Amounts are based on the Department’s estimated start-up costs for its 2010 4,000-bed expansion project. The 4,000-bed project was for

minimum- and medium-custody level beds. Start-up costs for higher custody level beds could vary.

3 Amounts are derived by multiplying the number of beds by a per capita rate. Per capita rates are taken from the Department’s Fiscal Year

2009 Operating Per Capita Cost Report. The private minimum-custody level rate was calculated by auditors by averaging all private
minimum-custody level per capita rates, weighted by number of prisoners in each facility. Per capita rates may fluctuate between fiscal years
2012 and 2017, which would affect these results.

4 Private prison construction and start-up costs are covered under the per capita rate paid to private prison contractors included in the

Operating Costs column.

Source:

Auditor General staff analysis of ADOA prison construction cost estimates, ADOA-reported actual costs for the 2010 4,000-bed
expansion project, the Department’s Fiscal Year 2011 Budget Request Decision Package, and the Department’s Fiscal Year 2009
Operating Per Capita Cost Report.

number includes the 5,000 private prison beds required by Laws 2009, 3rd S.S.,
Ch. 6. In addition to these beds, the Department proposes to contract for an
additional 1,500 private prison beds in fiscal years 2015 and 2016. Contracting
for these 6,500 total private beds could cost the State an estimated $640.7
million through fiscal year 2017 based on the average rate paid to private
prisons in fiscal year 2009. However, the actual per capita rates for future private
prison beds could be higher than the 2009 rates used to develop the estimates.

State of Arizona
page

18

•

State-operated beds—The Department’s plan also recommends constructing
additional facilities, either by expanding existing prison facilities or by constructing
new facilities, to add 2,000 beds, including 1,000 maximum-custody level beds
for male inmates and 1,000 minimum-custody level beds for female inmates.
According to the Department’s plan, these beds would become available in
fiscal years 2013 and 2016. Adding these beds could cost at least $334.1 million
to construct and operate through fiscal year 2017. Again, however, actual costs
could be higher. For example, both Arizona Department of Administration and
department officials reported that the estimated construction costs for the
proposed state-operated facilities—estimated to be $140 million of the $334.1
million—are conservative because they are based on estimated 2010 costs,
and actual costs will depend on when and where the facilities will be built. Costs
are highly dependent on the location chosen for the facilities, and it is possible
that additional monies could be needed to account for higher construction costs
in various parts of the State, land costs, or costs to expand waste and water
treatment facilities at existing prison complexes.

Although auditors used different per capita rates for the private and state-operated
beds in developing the cost estimates, the stated costs represent different custody
level beds and bed activation years and should not be used to compare the costs of
private versus state-operated prisons. Moreover, these estimates are largely based
on the number of needed beds identified in the Department’s bed plan and operating
costs reported in its Fiscal Year 2009 Operating Per Capita Cost Report. As of
September 2010, department officials reported that they were in the process of
updating both of these documents. Consequently, the projected bed need and cost
estimates could change

Projected bed need and
estimated costs for these
beds could change.

State should further study cost-effectiveness of privately
operated prisons compared to state-operated prisons
Part of the deliberations about adding capacity is determining whether the State
should contract with private prisons for additional beds or construct and operate its
own prisons. As discussed in Chapter 1 (see pages 3 through 16), the State has
pursued both of these options to help meet the prison housing demands that the
State’s growing prison population has required. Although statute requires the
Department to consider contracting for private prisons before expanding or
constructing prison facilities for certain offenders and allows the Department to enter
into private prison contracts, statute also stipulates that such contracts offer "cost
savings" to the State. However, department analysis of private prison and state prison
costs indicated that it may be more costly to house inmates in private prisons.
Specifically, according to the Department’s Fiscal Year 2009 Operating Per Capita
Cost Report, the State paid private prisons a higher per inmate rate than it spent on
equivalent services at state-operated facilities in fiscal year 2009. After adjusting state
and private rates to make them more comparable, the Department’s study found that
rates paid to private facilities were higher for both minimum- and medium-custody
Office of the Auditor General
page

19

beds—the two categories of beds for which the Department contracts (see Table
4).1,2 Specifically:

Table 4:

Comparison of Department Calculations
for Actual and Adjusted Private Prison
Per Capita Rates and State Prison Per Capita Costs
Fiscal Year 2009
(Unaudited)

Actual per capita rate
Minimum-custody inmates
Medium-custody inmates
Adjusted per capita rate1
Minimum-custody inmates
Medium-custody inmates

Cost Savings From Use
of Private Prisons
Per Inmate Per Inmate
Per Day
Per Year

StateOperated
Facilities

Privately
Operated
Facilities

$58.80
$60.51

$54.78
$63.52

$4.02
-$3.01

$1,468
-$1,099

$46.81
$48.13

$47.14
$55.89

-$0.33
-$7.76

-$121
-$2,834

1 Includes adjustments for healthcare costs, depreciation costs, and costs for functions provided only
by the State to make private and state prison comparisons more comparable.
Source: Auditor General staff analysis of the Department’s Fiscal Year 2009 Operating Per Capita Cost
Report.

•

Minimum-custody beds—The Department’s study found that although the
State paid less per inmate, per day to private prisons than the cost to house a
minimum-custody inmate at a state-operated facility ($54.78 vs. $58.80), state
costs were not directly comparable to the private prison rate because private
prisons do not have all the same responsibilities and costs as state-operated
facilities. For example, according to the report, private prisons do not accept
inmates in need of more serious medical care, nor do they intake and classify
inmates because the Department does this. After removing dissimilar costs and
adding a depreciation cost to the state rate to mirror construction costs captured
in the private prison rate, the Department found that the State paid private
prisons a slightly higher rate than it spent to house minimum-custody inmates
in state-operated facilities ($47.14 vs. $46.81).

•

Medium-custody inmates—For these inmates, both the actual and adjusted
rates paid to privately operated facilities were higher, according to the
Department’s study. Specifically, the per day rate paid to privately operated
facilities was $3.01 higher than the cost at state-operated facilities ($63.52 vs.
$60.51). Again, however, state costs and private-prison rates were not directly

1 Auditors assessed the reliability of the unadjusted per capita costs reported by the Department and concluded that

they were based on a reasonable method and appeared reasonably accurate. However, auditors did not assess the
Department’s method for adjusting the per capita rates for comparing state-operated and private prison costs.

2 As of September 2010, department officials reported that they were in the process of updating its Fiscal Year 2009

Operating Per Capita Cost Report, and its comparison of private and state-operated prison operating per capita costs
could change pending completion of the revision.

State of Arizona
page

20

comparable for the reasons described above. After making the adjustments, the
difference grew to $7.76 ($55.89 vs. $48.13).
Other studies auditors reviewed were consistent with the Department’s analysis.
These studies indicated that costs savings from contracting with private prisons in
place of state-operated prisons are not guaranteed. For example, a 2009 University
of Utah review of eight studies comparing private and state prison costs found that
results were mixed. Specifically, four studies identified private prison cost savings
ranging from 4.6 percent to 15.2 percent, two studies found no difference in costs,
and two studies—including a 2006 study the Department commissioned—found
that costs of private prisons were 10.0 to 14.2 percent higher.1 In addition to prison
operational costs, consideration should be given to whether the State can construct
new prisons at a lower cost. According to a 2001 U.S. Department of Justice report,
evidence suggested that private companies can construct new facilities faster and
cheaper than the public sector.2 Additionally, this report noted that there is no
consensus among academics and professionals in the field regarding the potential
cost savings that private prisons can offer. Therefore, if the Legislature decides to
expand the prison system, it should consider directing the Department to further
study and analyze the costs for the State to build and operate prison facilities
compared to contracting with private prisons to determine which option would be
more cost-effective while still ensuring public safety.

Studies have indicated that
contracting with private
prisons does not
guarantee cost savings.

1 See Lundahl et al., 2008
2 See Austin and Coventry, 2001

Office of the Auditor General
page

21

State of Arizona
page

22

Chapter 3
Option 2—Diverting more nonviolent, low-risk
offenders or reducing the time they serve to
address prison population growth
A second option for addressing projected prison population growth is to divert more
nonviolent, low-risk offenders from prison or reduce the time they serve—alternatives
that may require changes to the State’s sentencing laws. Arizona’s sentencing laws
largely dictate prison sentences and have contributed to the growth that has
occurred to date in the prison population. The Legislature has studied changing
these laws several times, and an ad hoc committee in the House of Representatives
is addressing the subject again in 2010. Although some steps have been taken to
divert nonviolent, low-risk offenders from prison and reduce the time they serve, the
Legislature could consider expanding these efforts. Establishing a permanent
sentencing commission to periodically review Arizona’s sentencing laws and help
monitor the State’s prison population would be a way to provide ongoing attention
to this area.

Arizona laws largely determine prison sentences
Arizona’s sentencing laws largely determine prison sentences. Since 1978, Arizona
has enacted several sentencing laws to provide equity in the sentencing process and
harsher penalties for certain crimes, and to ensure that offenders serve most of the
sentence imposed. These laws include presumptive sentencing, which requires that
judges impose certain sentences based on the felony offense; mandatory
sentencing, which provides for harsher penalties for certain offenses; and truth in
sentencing, which dictates how long a sentenced offender must serve. Specifically:

•

Presumptive sentencing—Arizona’s presumptive sentencing system, which
became effective in 1978, requires judges to impose a statutorily defined
sentence for a given offense. Prior to this change, judges had broad discretion
in determining sentences, which resulted in sentencing disparities for similar
crimes. Presumptive sentencing was adopted to provide more equitable

Office of the Auditor General
page

23

punishment for similar offenders who commit similar crimes. Under the State’s
presumptive sentencing system, felony offenses are assigned to one of six
classes depending on their seriousness, with class 1 being the most serious
and class 6 the least serious. Judges are required to impose a recommended,
or “presumptive,” sentence for a given offense class, but may give shorter or
longer sentences within a statutorily defined range based on mitigating or
aggravating circumstances. According to a 2005 Vera Institute of Justice report
(Vera report), nine states, including Arizona, adopted some form of presumptive
sentencing between 1975 and 2002.1 In contrast, between 1980 and 2002, 17
other states adopted sentencing guidelines rather than presumptive sentencing.2
Although presumptive sentencing does not preclude judges from sentencing
eligible offenders to probation rather than prison time, it largely shifts discretion
in determining sentence lengths to the Legislature, which determines sentence
length in statute, and to prosecutors, who determine which violations to charge.
Prosecutors can also offer plea bargains that reduce the seriousness or number
of charges against the defendant in exchange for a guilty plea.

Presumptive sentencing
shifts discretion from
judges to the Legislature
and prosecutors.

•

Mandatory sentences—Arizona also began adopting mandatory sentence
provisions in 1978 that provide for harsher penalties for certain groups of
offenders, such as repeat offenders, violent offenders, sex offenders, and
certain DUI and drug offenders. Mandatory sentencing provisions require the
judge to send the offender to prison (i.e., make the offender ineligible for
probation) and/or to lengthen the presumptive sentence for the offense (see the
textbox below for an example of how mandatory sentences can affect
sentencing). However, the judge can apply a mandatory sentence only when the
prosecutor presses charges that require a mandatory penalty. According to the

Example of how mandatory sentencing can affect sentence length:
A person convicted of robbery, a class 4 felony offense, can either be
sentenced to probation or sent to prison for 2.5 years, the presumptive
sentence for a class 4 felony. However:
• If the offender had a prior felony conviction (regardless of what it was),
the prosecutor could press charges that invoke the repetitive offender
mandatory sentence depending on when the prior offense occurred. If
proven, the offender would be ineligible for probation and the imposed
presumptive sentence would increase from 2.5 to 4.5 years.
• If the offender used or threatened to use a gun during the robbery, the
prosecutor could invoke the dangerous offender mandatory sentence. If
proven, the offender would be ineligible for probation and the imposed
presumptive sentence would increase from 2.5 to 6 years.

1 See Stemen, Rengifo, & Wilson, 2005
2 According to the Vera report, sentencing guidelines are a system of multiple recommended sentences and dispositions

and a set of procedures designed to guide judicial sentencing decisions and sentencing outcomes that account for the
severity of the offense and prior criminal history. Guidelines can be presumptive, which requires judges to impose a
sentence within a range or provide written justification for imposing some other sentence (which sentence can be
appealed), or voluntary, which does not require judges to impose the sentence recommended by the guidelines.

State of Arizona
page

24

Vera report, all 50 states had adopted various mandatory sentencing laws by
2002.

•

Truth in sentencing—In 1993, Arizona adopted truth-in-sentencing laws that
abolished discretionary release by a parole board for any offense committed
after 1993 and require offenders to serve at least 85 percent of their sentences
before becoming eligible for community supervision (the 85 percent requirement
applies to both violent and nonviolent offenders).1 Prior to this change, prisoners
were required to serve at least 67 to 75 percent of their sentences (depending
on the offense), but typically became eligible for parole after serving one-half or
two-thirds of their sentences.2 Truth in sentencing was adopted to promote truth
and accountability in sentencing by requiring offenders to serve the majority of
their sentence. According to the Vera report, 17 states had abolished
discretionary parole release by 2002, while 33 states still had it.3 In addition,
according to a 1999 Bureau of Justice Statistics report, most states had laws
requiring offenders to serve a specific percentage of their sentences.4 According
to this report, by 1998, 27 states required violent offenders to serve at least 85
percent of their sentences. However, the report noted that only a few states—
such as Florida, Mississippi, and Ohio—also required nonviolent offenders to
serve a substantial portion of their sentences, similar to Arizona.

According to a 1999
Bureau of Justice Statistics
report, only a few states
required nonviolent
offenders to serve a
substantial portion of their
sentences, similar to
Arizona.

Sentencing laws have affected State’s prison population
Arizona’s sentencing laws affect the State’s prison population by determining who
goes to prison and how long they stay. Prison population growth is essentially a
function of prison admissions and length of stay. In Arizona, the combination of
sentencing laws has contributed to increased prison admissions, but the actual time
inmates served has decreased.
Specifically, consistent with national trends, Arizona’s imprisonment rate has steadily
increased since adopting presumptive sentencing. However, the rate of increase
slowed after abolishing discretionary parole when the State made various sentencing
changes in 1993. These results are consistent with the Vera report, which found that
states that controlled sentencing decisions through presumptive sentencing but did
not control release decisions by abolishing discretionary parole release—such as
Arizona between 1978 and 1993—had higher incarceration rates.5 In contrast, the
1 Parole is a period of conditional supervised release outside of prison before an entire prison term is completed. It is

granted by the Arizona Board of Executive Clemency after the inmate has served a portion of his or her sentence and
has applied for release on parole. Parole eligibility dates are calculated in accordance with the provisions of the
committing offense and the laws in effect at the time the offense was committed. Only inmates who committed offenses
before January 1, 1994, are eligible for parole. Community supervision is a portion of a felony sentence and is served
consecutive to the inmate’s period of imprisonment. The term of community supervision is a period equal to 1 day for
every 7 days of the sentence and is imposed on the convicted person by the court at the time of sentencing.
Community supervision replaced parole after truth in sentencing was adopted.

2 Alternatives to Sentencing Workgroup, 2005
3 See Stemen, Rengifo, & Wilson, 2005
4 See Ditton & Wilson, 1999
5 See Stemen, Rengifo, Wilson, 2005

Office of the Auditor General
page

25

report found that states that controlled both sentencing and release decisions
through presumptive sentencing and abolishing discretionary parole, such as
Arizona after 1993, had lower incarceration rates and smaller growth in incarceration
rates.
The Vera report also found that states with more mandatory sentences have higher
incarceration rates. The report noted that mandatory sentencing laws may not lead
directly to increased incarceration, but likely act as proxies for a state’s general
approach to sanctioning offenders. Specifically, mandatory sentences may lead to
higher incarceration rates because they can provide prosecutors additional leverage
in plea bargains.1 Literature further suggests that mandatory sentences are applied
only in a few cases and are instead used by prosecutors to obtain a conviction
through plea bargaining that does not have a mandatory penalty attached.2 In
Oregon, for example, incarceration rates have increased, but more offenders were
convicted for nonmandatory offenses while convictions under mandatory sentencing
statutes have declined.3 Although a 1992 Department of Corrections (Department)
study reported that mandatory sentencing had caused a buildup of longer-term
offenders in the prison system, a 2010 Arizona Prosecuting Attorneys’ Advisory
Council report indicated that only 25 percent of the prison population was
incarcerated with mandatory or flat term sentences as of September 30, 2009.4,5,6

Since adopting truth in
sentencing, the median
time violent offenders
serve in prison is only a
few months longer than
the median time
nonviolent offenders serve
in prison.

Finally, although it was thought that truth in sentencing would require inmates to
serve longer in prison, other changes instituted when Arizona adopted truth in
sentencing have diminished its expected impact. Specifically, when Arizona adopted
the 85 percent truth-in-sentencing requirement in 1993, it also shortened sentences
for nondangerous offenders without two or more prior felony convictions. As shown
in Table 5 (see page 27), auditors’ analysis of department admissions data indicates
that median sentence lengths have generally decreased for several nonviolent and
violent crimes (except for homicide, manslaughter, and sexual assault) since
adopting truth in sentencing. However, although median sentence lengths for several
nonviolent crimes have decreased, truth in sentencing has meant that actual time
served in prison for nonviolent offenses has not changed appreciably. According to
auditors’ analysis of inmates released from prison between 1990 and 2009, inmates
sentenced for nonviolent offenses committed before 1994 served a median of 2
years in prison, while inmates sentenced for nonviolent offenses committed in or after
1994 served a median of 1.9 years. In contrast, the actual time served for violent
offenses has decreased from a median of 4.8 years for offenses committed before
1994 to a median of 2.6 years for offenses committed in or after 1994. Thus, for
offenders sentenced under truth in sentencing, the typical violent offender spends
only a few months longer in prison than the typical nonviolent offender.

1 Merritt, Fain, & Turner, 2006
2 Ulmer, Kurlychek, & Kramer, 2007; Tonry, 2009
3 Merritt, Fain, & Turner, 2006
4 See Fischer & Thaker, 1992
5 See Fischer, 2010
6 Flat term sentences are a form of mandatory sentencing that require the offender to serve 100 percent of the imposed

sentence.

State of Arizona
page

26

Table 5:

Median Sentence Lengths for Admitted Offenders
by Offense Type, in Years1
Calendar Years 1990, 1993, 1996, 1999,
2002, 2005, and 2008
(Unaudited)
1990

1993

1996

1999

2002

2005

2008

5.00

5.00

3.50

3.50

3.00

3.00

3.50

Homicide
Manslaughter
Robbery
Sex Offender
Sexual Assault
Weapons Offense

15.00
7.50
7.00
10.00
10.00
4.00

15.00
9.00
7.00
10.00
10.00
4.00

17.25
10.00
5.00
10.00
7.50
2.50

16.00
10.50
5.00
6.00
10.00
2.50

16.00
10.50
5.00
5.00
8.00
2.50

14.50
10.50
5.00
6.00
10.00
2.50

16.00
10.50
7.00
5.00
10.00
2.50

Nonviolent Crimes
Burglary
Drugs
DUI
Forgery-Fraud
Theft

5.00
5.00
0.50
5.00
4.00

5.00
4.00
0.50
5.00
5.00

3.00
2.50
0.33
2.50
2.75

3.00
2.50
0.33
2.50
2.50

2.50
3.00
0.33
2.50
2.25

2.50
2.50
0.33
2.50
1.75

2.50
2.50
0.33
2.50
1.75

Violent Crimes
Assault

1 The analysis includes admissions resulting from new court commitments, except for inmates

sentenced to life in prison.

Source: Auditor General staff analysis of department prison admissions data.

Other potential sentencing law changes have been
under study for many years
The Legislature has been studying whether to change Arizona’s sentencing laws as
far back as 1991, and an ad hoc committee in the House of Representatives (House)
studied this issue in 2009 and 2010. Various organizations have issued reports
calling for changes that would address the State’s growing prison population and set
up other ways to equitably hold offenders accountable for their crimes. For example:

•

In a 1991 report that the Arizona Legislative Council commissioned, the Institute
for Rational Public Policy recommended that the Legislature repeal all mandatory
sentencing provisions, replace presumptive sentences with presumptive
guidelines, and create a sentencing commission to establish the guidelines.1

•

In 2004, Families Against Mandatory Minimums recommended that the
Legislature give judges authority to set aside mandatory prison sentences,
make drug court an option for all nonviolent offenders with underlying substance
abuse problems, provide alternatives to prison for drunk drivers, make probation

1 Institute for Rational Public Policy, 1991

Office of the Auditor General
page

27

a sentencing option for more nonviolent offenders, and use alternatives to
prison for probation violators.1,2

•

A House of
Representatives ad hoc
committee reviewed
Arizona’s sentencing laws
during 2009 and 2010.

In 2005, the Alternatives to Sentencing Workgroup formed by the House and
comprising nine house members made similar recommendations, which
included allowing judges to ignore mandatory penalties for some offenders,
creating a sentencing commission, reforming truth-in-sentencing laws for
nonviolent first-time offenders, and expanding the use of alternatives to prison
such as treatment, drug courts, and electronic monitoring.3,4

The House is again reviewing the need for sentence reform. In 2009, it formed an ad
hoc Committee on Sentencing comprising six house members to “review and
assess Arizona’s sentencing laws and evaluate the purpose, history, and evidence
of effectiveness of the laws regarding criminal sentencing.” The committee is
charged with reporting its findings and recommendations for improving Arizona’s
sentencing laws to the Speaker of the House on or before November 1, 2010. Since
its formation, the committee has held two meetings, one in December 2009 and the
other in May 2010. At these meetings, the committee has received testimony from
various stakeholders regarding Arizona’s prison population, the impact of mandatory
sentences and truth in sentencing, and areas in which the Legislature could make
changes to reduce the prison population. Members of the committee have also met
with stakeholder groups and received and reviewed applicable literature.

Options to divert nonviolent, low-risk offenders from
prison or reduce time they serve could be expanded
Although Arizona voters and the Legislature have taken steps to divert some
nonviolent, low-risk offenders from prison or reduce the time served in prison, the
Legislature could consider expanding these efforts to further address prison
population growth, some of which may require changes to the State’s sentencing
laws. Specifically, Arizona voters passed Proposition 200 in 1996, which requires
some nonviolent drug offenders to be sentenced to probation and treatment rather
than prison. The Legislature could consider expanding this program by diverting
more nonviolent, low-risk offenders to nonprison alternatives (see Chapter 4, pages
37 through 46, for more information). Additionally, the Legislature established an
early release program in 2003 for nonviolent, low-risk offenders and could consider
other early release options, such as reducing the time served requirement for
nonviolent, low-risk offenders and establishing earned time credits.
1 See Greene & Pranis, 2004
2 Families Against Mandatory Minimums is a nonpartisan, nonprofit organization that promotes judicial sentencing

discretion.

3 Alternatives to Sentencing Workgroup, 2005
4 With regard to its recommendation to reform truth-in-sentencing laws for nonviolent first-time offenders, the Workgroup’s

report did not include specific criteria for defining nonviolent first time offenders, although it stated the recommendation
particularly applied to women offenders.

State of Arizona
page

28

If the Legislature expands diversion or early release options, it also should consider
taking the following steps:
1.

Further defining diversion and/or early release eligibility criteria for
other nonviolent, low-risk offenses in statute; and/or

2.

Ensuring the use of valid and reliable risk assessment tools in
determining offender eligibility for these options.

Diversion opportunities could be expanded to more nonviolent, lowrisk offenders—In 1996, Arizona voters passed Proposition 200, adding
Arizona Revised Statutes (A.R.S.) §13-901.01, which requires nonviolent persons
convicted of a first or second offense for the personal possession or use of drugs
to be sentenced to probation and mandatory treatment. According to a 2004 Vera
report, Arizona’s law was the nation’s first successful effort to replace incarceration
with treatment for some substance-abusing offenders.1 According to a 2006
Arizona Supreme Court report, an estimated 1,072 offenders were diverted to
probation and treatment in fiscal year 2005, resulting in an estimated $11.7 million
in net costs avoided.2,3 The Legislature could consider this same approach for
other nonviolent, low-risk offenders, particularly those whose crimes are related to
substance abuse. As illustrated in Table 1 (see page 5), offenders convicted of
property crimes, such as burglary, theft, and fraud, made up 23 percent of the
inmate population as of December 31, 2009, and these crimes are often associated
with drugs.

Arizona law diverts some
nonviolent drug offenders
to probation and
treatment.

Expanding diversion may require sentencing law changes—In order
to divert more nonviolent, low-risk offenders from prison, the Legislature may need
to consider revising some of the State’s sentencing laws. Specifically, the
Legislature could consider revising some sentencing laws to allow nonviolent, lowrisk offenders to be diverted to nonprison alternatives. Revising these laws would
allow judges to consider individual cases in determining whether prison or some
alternative such as treatment would be more appropriate for the offender. Other
states have taken steps to divert a wider range of nonviolent, low-risk offenders
from prison. For example:

•

Similar to Arizona, Hawaii passed legislation in 2002 requiring probation with
treatment for first-time, nonviolent offenders convicted of drug possession or
use. However, Hawaii later expanded the availability of diversion to treatment
for first-time, nonviolent drug offenders with prior nondrug convictions in 2004
and for first-time property offenders whose offense was committed in response
to substance abuse problems in 2006.

1 See Wool & Stemen, 2004
2 Arizona Supreme Court, Administrative Office of the Courts, Adult Probation Services Division, 2006
3 The estimate is based on marginal costs (department costs to house, feed, and supervise one additional inmate) for

state-operated prisons, the full per capita costs for private prisons, and probation costs. It assumes that one-third of
the diverted offenders would have been sent to state-operated prisons and two-thirds would have been sent to private
prisons. The estimated net costs avoided had all diverted offenders been sent to state-operated prisons was more than
$1.4 million. The estimated net costs avoided had they all been sent to private prisons was approximately $16.9 million.
Office of the Auditor General
page

29

•

Florida passed legislation in 2009 that requires courts to sentence nonviolent,
low-risk offenders to a new diversion program unless they pose a risk to the
public. This law is not limited to drug offenders.

•

New York passed legislation in 2009 that reformed its drug laws. These reforms
included providing discretion to judges to place convicted drug offenders into
treatment; diverting people who have committed crimes other than a drug
offense but whose crime was related to substance abuse; and providing
diversion eligibility to some second-time repeat offenders.

If the Legislature decides to divert more offenders to alternatives other than prison,
it could consider expanding the availability of these alternatives. Chapter 4
discusses potential nonprison alternatives that the Legislature might consider in
more detail (see pages 37 through 46).

Legislature could consider expanding early release alternatives—The
Legislature enacted Laws 2003, Ch. 256, which established an early release
program for nonviolent, low-risk drug offenders who make satisfactory progress in
their individualized corrections plans, maintain civil behavior, and meet other
criteria. Inmates who participate in this program are released 3 months earlier than
they otherwise would be under truth in sentencing and receive a variety of services,
including substance abuse treatment, transitional housing, education and
employment services, help in accessing social services, and mentoring
relationships. At the end of the 3 months, they are placed on regular community
supervision to complete the remaining 15 percent of their sentence. According to
department reports, the number of participating inmates grew from approximately
500 offenders in calendar year 2005 to approximately 1,000 offenders in fiscal year
2009, and most participants successfully complete the 3-month early supervised
release and continue their term of community supervision (896 inmates successfully
completed the supervised early release during fiscal year 2009, while only 72
inmates did not complete it). In addition, A.R.S. §31-285 requires the program to
result in a cost reduction of at least $17 per inmate, per day.
The Legislature enacted a
3-month early release
program for certain
nonviolent, low-risk
offenders.

In April 2010, the Legislature amended A.R.S. §31-281 to expand program
eligibility to include all nonviolent, low-risk offenders who have not been convicted
of a sexual offense, arson, or driving under the influence. The Department expects
the number of eligible offenders to double based on this law change.
The Legislature could consider other alternatives for expanding early release to
reduce the amount of time nonviolent, low-risk offenders serve in prison. There is
always a risk that an offender will commit new crimes once released from prison.
However, according to a 2002 Bureau of Justice Statistics study, the deterrent
effect of incarceration on recidivism is mixed.1 The study, which measured
recidivism rates for prison inmates from 15 states who were released in 1994,
found that recidivism rates (measured as rearrest within 3 years of release) did not

1 See Langan & Levin, 2002

State of Arizona
page

30

differ significantly for anyone serving less than 5 years in prison.1 However, there
was a significant reduction in recidivism for inmates serving more than 5 years. The
study also found no evidence that spending time in prison raises the recidivism
rate. A 2008 National Council on Crime and Delinquency review of more than
12-peer reviewed articles or reports on early release programs and their effect on
recidivism found no significant difference in rates of recidivism among accelerated
release and full-term prisoners, and, in some cases, early release prisoners had
lower recidivism rates.2 These results have led at least one group of advocates to
conclude that modest changes in the length of stay, such as reducing it by a few
months, likely would have no impact on recidivism rates or aggregate level crime
rates.3
Additional alternatives that the Legislature might consider for early release include:

•

Reducing time served requirements for nonviolent offenders—The
Legislature could consider revising truth-in-sentencing laws to reduce the
amount of time that nonviolent, low-risk offenders must serve in prison. As
previously discussed (see page 25), truth-in-sentencing laws that most states
adopted generally focused on violent offenders, and only a few states required
both violent and nonviolent offenders to serve a substantial portion of their
sentences, similar to Arizona. At least one of these states has revised its policy
for nonviolent offenders. Like Arizona, Mississippi abolished discretionary
parole as a release mechanism and required all offenders to serve 85 percent
of their sentences when it adopted truth in sentencing in 1995. However, in
2001, it reinstated parole eligibility for first-time, nonviolent offenders and, in
2008, expanded eligibility to all offenders never convicted of a violent crime or
a crime with an enhanced penalty regardless of the number of prior convictions.
These inmates are eligible for parole after serving 25 percent of their sentence
(although there are minimum-time-served requirements for inmates with
shorter sentences). The law took effect in July 2008, and was applied
retroactively for nonviolent offenses committed after June 30, 1995. According
to a 2010 Pew Center on the States report, 3,076 prisoners were released
between July 2008 and August 2009.4 These inmates served a median of 13
months’ less time in prison than they would have under the previous truth-insentencing requirements. According to Mississippi Department of Corrections
staff, this has resulted in costs savings—its parole costs were $1.55 per day in
fiscal year 2009 while its prison costs were between $36.67 and $41.61 per
day. Moreover, the Mississippi Department of Corrections reported that
between January 31, 2009 and January 31, 2010, the state’s prison population
decreased by 1,360, although it had been expected to increase by 1,000
before revising the truth-in-sentencing laws.

Mississippi has revised its
truth-in-sentencing laws
for nonviolent offenders.

1 The analysis included only offenders leaving prison for the first time since the beginning of their sentence. It excluded

offenders who left prison in 1994 but who had previously been released under the same sentence and had returned to
prison for violating the conditions of release.

2 See Guzman, Krisberg, & Tsukida, 2008
3 Austin & Fabelo, 2004
4 See Pew Center on the States, 2010

Office of the Auditor General
page

31

Arizona could potentially realize a reduction in its prison population and
corrections spending if it similarly revised its truth-in-sentencing requirements.
The Legislature could reinstate discretionary parole release for nonviolent
offenders similar to Mississippi. However, as noted previously, the 2005 Vera
report suggested that reestablishing discretionary parole could lead to higher
incarceration rates.1 Alternatively, the Legislature could reduce the 85 percent
time-served requirement for nonviolent, low-risk offenders. For example, if the
Legislature reduced the time served requirements for nonviolent, low-risk
offenders to 50 percent instead of 85 percent, actual time served would be
reduced by 10.5 months based on a 2.5-year sentence. For every day that an
inmate spends on community supervision (parole) rather than prison, the State
would save an estimated $4.62, which represents the difference between the
daily marginal cost of housing an inmate in a state-operated prison compared
to supervising an inmate on parole in fiscal year 2009.

•

According to the National
Conference of State
Legislators, 31 states offer
earned time credits that
reduce the time inmates
who participate in
rehabilitative and
productive activities serve
in prison.

Creating earned time credits—Another approach would be for the Legislature
to authorize earned time credits for inmates. According to a 2009 National
Conference of State Legislatures report, earned time credits are reductions to
the portion of a sentence that must be served in prison that inmates can earn
for completing education, vocational training, treatment, or work programs or
participating in other productive activities.2 Earned time credits are different
from, and can be offered in addition to, “good time” credits that are given for
following prison rules (such as the credits that allow Arizona offenders to serve
15 percent of their sentences on community supervision). According to the
report, at least 31 states offer earned time credits to inmates. Earned time
credits are usually made available to lower-risk offenders, and the typical range
for a credit is between 30 and 120 days. For example, Kansas offers a 60-day
earned time credit for successfully completing substance abuse treatment, a
general education diploma, a technical or vocational training program, or any
program that the secretary of corrections believes will reduce an inmate’s risk
of violating release conditions. Texas offers 10 to 30 days for each month an
inmate works or participates in educational, vocational, or rehabilitation
programs while in prison.

Legislative options may require further action—If the Legislature expands
diversion or early release options, it should also consider taking the following
steps:

•

Defining offender eligibility—Similar to A.R.S. §13-901.01, which requires
diversion for first or second drug use or possession offenses, the Legislature
could consider further defining diversion and/or early release eligibility criteria
for other nonviolent, low-risk offenses in statute. Other states have made other
crimes eligible for diversion, including first-time property offenders whose
offense involved substance abuse, low-level offenders whose drug and

1 See Stemen, Rengfio, & Wilson, 2005
2 See Lawrence, 2009

State of Arizona
page

32

alcohol use contributed to the criminal activity, and other nonviolent, low-risk
offenders.

•

Ensuring the use of valid and reliable risk assessment tools—In addition
to or instead of defining offender eligibility in law, the Legislature could consider
ensuring the use of a valid and reliable risk assessment tool to determine
offender eligibility for diversion and/or early release. According to a 2007 report
prepared for the Crime and Justice Institute and the National Institute of
Corrections, assessing an offender’s level of risk and criminogenic needs
through both the use of an actuarial tool and sound professional judgment is
important for determining an offender’s suitability for diversion.1 Additionally,
according to the Justice Policy Institute, a growing number of states are
beginning to use actuarial risk and needs assessments in various parts of the
criminal justice system.2 For example, according to a 2007 evaluation, Virginia
uses an actuarial risk assessment in conjunction with the state’s voluntary
sentencing guidelines to divert 25 percent of nonviolent, prison-bound
offenders with the lowest risk of recidivating.3 Finally, both the county probation
departments and the Department have developed risk assessment tools. The
county probation departments use the offender screening tool (OST) to assess
a defendant’s risk of reoffending and criminogenic needs and make diversion
recommendations in the pre-sentencing reports prepared for the judges. They
also use the field reassessment offender screening tool (FROST) to reassess
probationers. The Department has developed a tool for assessing an offender’s
risk of recidivism that it uses to determine initial supervision levels for inmates
released to the community, to help determine eligibility for the 3-month early
release program, and to rank offenders for in-prison programming (see textbox, page 34). The Department also uses the FROST to determine community
supervision levels. According to Administrative Office of the Courts and
department officials, these tools may be appropriate for determining offender
eligibility for expanded diversion and/or early release options.
Based on the Department’s assessment of offender risk, auditors identified a
number of inmates in the prison population on December 31, 2009, who
appear to be at a lower risk for committing new felony offenses. Specifically,
auditors identified 3,538 inmates (nearly 9 percent of the prison population)
who have never been sentenced to prison for a violent offense, whose
recidivism risk scores are 3 or less (considered lower risk by the Department),
and who were incarcerated for less serious offenses (offense classes 4, 5, or
6). Crimes for which these offenders were convicted include drug offenses,
burglary, driving under the influence, forgery/fraud, and theft. Although the
number of offenders is a relatively small percentage of the overall prison
population, consistently diverting even a small number of low-risk offenders
from prison could have a significant impact on the prison population over time.
It could also reduce prison costs. Estimated marginal prison costs for the

Auditors identified 3,538
nonviolent inmates who
appear to be at a lower
risk for committing new
felony offenses.

1 See Warren, 2007
2 Justice Policy Institute, 2010
3 See Kleiman, Ostrom, & Cheesman, 2007

Office of the Auditor General
page

33

Department risk assessment instrument
The instrument is composed of two assessments, one that assesses an offender’s general risk of
committing new felony offenses and one that assesses the risk of committing new violent felony level
offenses. The general risk score ranges from G1 (lowest risk) to G8 (highest risk), and the violence risk
score similarly ranges from V1 to V9. Offenders with both general and violence risk scores of 3 or less
are considered lower risk. The scores are based on the following risk factors:Age at most recent
commitment and current age
• Number of prior adult felony convictions and incarcerations
• Number of prior juvenile felony adjudications
• History of addictive drug use (primarily heroin, cocaine, and methamphetamine)
• Affiliation with a prison or street gang
• Gender
• Current or most recent prison custody level (minimum, medium, close, or maximum)
• History of felony violence or other serious offenses

3,538 offenders total approximately $26.9 million for the total time they are
required to serve in prison. Further, more savings could be realized if enough
inmates were diverted to close a prison unit. However, a significant number of
inmates would need to be diverted before a prison unit would close. As of
August 31, 2010, the Department had over 4,800 temporary beds in the prison
system, and department officials indicated that the Department would likely
stop using these beds before closing a unit.

Permanent sentencing commission could review
sentencing policies and laws
Finally, the Legislature could consider establishing a permanent sentencing
commission to assist in reviewing and recommending changes to the State’s
sentencing laws. According to a 2009 Vera report, sentencing commissions are
typically neutral permanent bodies that analyze data and research to inform
sentencing and corrections policies.1 The report noted that while sentencing
commissions were often established to develop sentencing guidelines, many states
are creating or expanding commissions to address broader criminal justice policy
agendas. As of April 2010, 19 states, the District of Columbia, and the federal
government had sentencing commissions that were members of the National

1 See Scott-Hayward, 2009

State of Arizona
page

34

Association of Sentencing Commissions, and at least 4 other states had adopted
sentencing commissions as well.
Sentencing commissions in other states perform a variety of functions. For example,
Virginia tasked its sentencing commission with developing discretionary sentencing
guidelines; developing an offender risk assessment instrument for assessing felons’
risk to the public; preparing guidelines for sentencing courts to use in determining
appropriate candidates for alternative sanctions; monitoring sentence lengths, crime
trends, correctional facility population trends, and correctional resources; and
making recommendations regarding correctional capacity and resource needs.
Washington’s sentencing commission is similarly responsible for evaluating
sentencing policies and practices and recommending modifications to the governor
and legislature. It conducts ongoing research of recidivism, disparities in sentencing,
prison and jail capacity, deterrence, drug policy, and sentence enhancements for
weapons-related crime.
Arizona has made some efforts to establish a sentencing commission in the past.
Laws 2002, Ch. 311, created a temporary sentencing commission charged with
reviewing the State’s sentencing practices and making recommendations for
changes by December 31, 2003. The commission was also charged with reviewing
class 6 felonies and considering whether any should be repealed or reclassified. This
commission comprised 27 members representing all three branches of government,
various members of the criminal justice system, and the community. However, the
commission disbanded before submitting its final report. The Legislature establishing
a permanent sentencing commission (with a similar membership to ensure all
criminal justice stakeholders’ participation) is one option for taking long-term action
in this area. Possible functions or responsibilities of a state sentencing commission
could include reviewing and recommending changes to the State’s sentencing laws,
determining eligibility criteria for diversion, recommending guidelines for determining
appropriate candidates for alternative sanctions, and monitoring reform results to
ensure they are having the intended effect.

Office of the Auditor General
page

35

State of Arizona
page

36

Chapter 4
Option 3—Expanding use of nonprison
alternatives to slow or reverse prison population
growth
A third option for addressing prison population growth is to expand the use of
nonprison alternatives for nonviolent, low-risk offenders—a step that, similar to
revising sentencing laws, could limit or reverse growth in the number of inmates. Like
all states, Arizona uses probation as an alternative to prison. Auditors identified both
counties and other states that have expanded their nonprison alternatives to include
forms of substance abuse treatment (in addition to drug court), home arrest with
electronic monitoring, or day reporting centers. Arizona could use similar alternatives
in lieu of prison sentences or in conjunction with earlier release from prison. Although
the Department of Corrections (Department) and the courts have statutory authority
to establish nonprison alternatives, the Legislature could consider directing the
Department and/or courts to further study the use and costs of nonprison alternatives
to identify the right mix of these alternatives for Arizona. Additionally, depending on
whether the Legislature provides funding for expanded nonprison alternatives and
which alternatives are expanded, some statutes will need to be revised, such as the
home arrest statute.

Arizona uses probation as nonprison alternative
Similar to other states, many Arizona felony offenders are sentenced to probation in
lieu of prison or as part of their sentence. Although probation serves as an alternative
to prison, probationers who violate the terms of their probation can be sent to prison
or jail, and, according to the Department, approximately 14 percent of fiscal year
2010 prison admissions were probation violators. Additionally, in some cases,
probationers also participate in drug court programs, which provide monitoring and

Approximately 14 percent
of prison admissions in
fiscal year 2010 were
probation violators.

Office of the Auditor General
page

37

drug treatment services, and impose other requirements on offenders charged with
or convicted of drug- or alcohol-related crimes.

Probation serves as nonprison alternative—Like all states, Arizona uses
probation as an alternative to prison. Probation is a criminal sentence in which an
offender agrees to fulfill certain court-mandated conditions and remains in the
community under supervision instead of serving time in jail or prison. These
conditions typically include a probation officer’s supervision, fines, participation in
treatment programs, community service, restitution, or other activities. Arizona has
three types of probation: standard, intensive, and administrative (see textbox). The
State’s adult probation system is decentralized and operates under the jurisdiction
of Arizona counties, each of which maintains a separate probation department.

Probation types
Standard Probation—Offenders placed on standard probation are under the care and
control of the court and are supervised by probation officers. Offenders typically must report
to a probation officer, maintain employment, pay fees or fines, and pay restitution.
Intensive Probation—Offenders placed on intensive probation are those who would
otherwise have been sent to prison at initial sentencing or for a technical violation of standard
probation. Offenders must comply with strict control, surveillance, and supervision of their
movement and activities in the community.
Administrative Probation—Offenders on administrative probation are not directly supervised
by probation officers because of incarceration in jail or prison, having absconded or been
deported, or being placed on unsupervised probation at sentencing.

Arizona courts sentence the majority of felony offenders to probation for at least part
of their sentence, and many offenders sentenced to probation also serve some time
in jail or prison. A felony offender is eligible for probation if a sentence of probation
is not prohibited by law. According to the Arizona Superior Court’s case activity report
for fiscal year 2009, approximately 44 percent of the State’s felony cases resulted in
a sentence of only probation, 20 percent resulted in probation with jail time, and 4
percent resulted in probation with prison time. As of May 31,
2010, almost 84,500 offenders were involved with the
Probation Populations as of May 31, 2010
probation system (see textbox).
Standard
Intensive
Administrative
Total
Source:

Auditor General staff analysis of Fiscal Year 2010 Monthly
Statistics, Statewide Population obtained from the
Administrative Office of the Courts Web site; see
http:/www.azcourts.gov/apsdMonthlyStatistics.aspx

State of Arizona
page

38

44,189
2,226
38,053
84,468

Probation violations can lead to jail or prison—If
a probationer violates the court’s conditions, the court can
revoke probation and impose a new sentence sending the
probationer to jail or prison. According to the Department,
probation violators accounted for approximately 14 percent
of prison admissions in fiscal year 2010, and, based on
auditors’ analysis, almost 8 percent of the prison population

as of December 31, 2009, consisted of inmates who had violated the terms of their
probation.
In 2008, the Legislature took action to reduce the number of probation revocations
by providing Arizona counties with a financial incentive to keep offenders on
probation. Laws 2008, Ch. 298, requires the Legislature to appropriate funding to
counties that reduce their total probation revocations and probation revocation
rates for new felony offenses using fiscal year 2008 revocation rates as a
benchmark. Specifically, the law requires the Legislature to appropriate 40 percent
of any cost savings resulting from reduced probation revocations and reduced
probation revocation rates for new felony offenses to eligible counties beginning in
fiscal year 2011. Statute also requires counties to use these monies to provide
substance abuse treatment and other risk reduction programs and interventions
for probationers, and to provide grants to nonprofit victims’ services groups that
partner with the county probation departments to assist victims and increase
restitution payments collected from probationers. In fiscal year 2009, according to
the Joint Legislative Budget Committee, eight counties met both reduction
requirements resulting in a cost savings of just over $6 million, of which
approximately $2.4 million should be allocated to the counties for fiscal year 2011.
However, because of budget considerations, the Legislature passed Laws 2010,
7th S.S., Ch. 6, §29, which suspends the requirement to allocate the funding to the
counties in fiscal year 2011.

Probation can involve drug court—Some

probationers also participate in
drug court programs. Drug courts are voluntary programs that combine the efforts
of judges, probation departments, and treatment providers into a coordinated
intervention for offenders charged with or convicted of drug-related crimes. Drug
courts provide supervision, drug testing, and treatment services such as
counseling and education, and participants must follow certain rules such as
abstaining from drugs and alcohol. Arizona has adult drug courts in nine counties,
and the Superior Court in which the programs operate determines program
eligibility.1 According to the Fiscal Years 2009-2011 Master List of State Government
Programs, there were 1,154 adult participants sentenced to drug court and 389
drug court graduates in fiscal year 2009. Some counties also have similar
programs for offenders convicted of driving under the influence or domestic
violence or who have mental health issues. The literature on the overall effectiveness
of drug courts is mixed, suggesting that its success may be limited to specific
interventions, outcomes, or participants.2 However, researchers have evaluated
individual drug courts in Coconino and Yuma Counties and identified several
positive outcomes, including lower recidivism and drug-use rates. Additionally, a
2005 U.S. Government Accountability Office evaluation of eight drug court
programs, including Maricopa County’s drug court, found that while recidivism
was reduced, other results, such as treatment outcomes and cost reductions,

Drug courts are voluntary
programs that coordinate
interventions for offenders
charged or convicted of
drug-related crimes.

1 These counties are Cochise, Coconino, Gila, Maricopa, Navajo, Pima, Pinal, Yavapai, and Yuma.
2 See Banks & Gottfredson, 2003; Gottfredson, Najaka, & Kearley, 2003; Shaffer, 2006; Wilson et al., n.d.

Office of the Auditor General
page

39

were mixed.1 In addition, this evaluation found that drug court programs tend to
be more expensive than conventional case processing.

Nonprison alternatives could take several forms
In addition to probation, auditors identified a number of counties and other states
that have implemented other nonprison alternatives in an effort to reduce their prison
populations. Advocates suggest that providing nonprison alternatives provides
several benefits. This section discusses three of these alternatives and any
recognized costs and benefits: (1) expanding treatment alternatives in conjunction
with or beyond drug courts, (2) expanding home arrest with electronic monitoring,
and (3) establishing day reporting centers. These or other alternatives could be used
in lieu of prison sentences or in conjunction with earlier release from prison.

Nonprison alternatives may offer benefits other than reducing prison
populations—In addition to potentially reducing or slowing prison population

Some states are using
nonprison alternatives,
especially for low-risk
offenders.

growth, nonprison alternatives may provide other benefits. Specifically, keeping
offenders in the community rather than behind bars allows them to maintain family
ties, be employed, and perhaps regain their place in the community.2 Advocates
of alternative sanctions also argue that nonprison alternatives are an effective
solution that reduces crime and recidivism and are a better investment for taxpayer dollars.3 However, the scientific research does not fully support this claim.
Specifically, testimony presented on July 10, 2009, before the United States
Sentencing Commission on the effectiveness of alternative sanctions concluded
that no definitive statements can be made on the comparative effectiveness of
alternative sanctions to incarceration, but a recent meta-study suggested there is
promise.4 Despite the mixed effects for alternatives, some states are using
alternative sanctions, especially for low-risk offenders.

Substance abuse treatment could be expanded—Some states, such as
Texas, Virginia, Vermont, and Kansas, have expanded substance abuse treatment
alternatives beyond drug courts in an effort to reduce prison admissions and
prevent recidivism. For example:

•

Texas—Texas has received national attention for its efforts to divert some
offenders from prison to treatment alternatives. According to a 2007 Council of
State Governments Justice Center report, in 2007, the state projected a need
for 14,000 additional prison beds by 2012.5 Building and operating new prison
facilities to meet this growth was estimated to cost $523 million for fiscal years
2008 and 2009 alone. Instead of building new prison facilities, the Texas

1 U.S. Government Accountability Office, 2005
2 Demlietner, 2005
3 Drake, Aos, & Miller, 2009
4 See Byrne, 2009; Also see Drake, Aos, & Miller, 2009
5 Council of State Governments Justice Center, 2007

State of Arizona
page

40

Legislature appropriated $241 million to expand substance abuse treatment
and prison diversion programs in order to address two of the state’s
contributing factors to prison population growth—substance abuse and
probation and parole revocations. According to a 2009 Council of State
Governments Justice Center report, this expansion included 800 residential
treatment beds and 3,000 outpatient treatment slots for probationers; 2,200
treatment slots for jail and prison inmates; and 1,250 transitional treatment
center beds for offenders transitioning from institutional treatment programs.1
In addition to these programs, Texas also funded 1,500 beds as part of a
residential treatment program for probationers and parolees who violate the
conditions of their supervision because of substance abuse problems. This
program includes 6 months of treatment in a secure facility, followed by 3
months in a transitional treatment center, and 3 to 9 months of outpatient
counseling.

Texas has expanded
substance abuse
treatment and prison
diversion programs to
address prison population
growth.

These treatment and diversion programs have helped Texas reduce its prison
population. According to a March 2010 Council of State Governments Justice
Center presentation to the Texas House Corrections Committee, Texas’ yearend 2009 prison population was about 1,000 inmates fewer than in September
2007 and about 9,000 inmates fewer than it had been projected to be.
Moreover, according to prison population projections Texas’ Legislative Budget
Board released for fiscal years 2010 through 2015 in June 2010, the state’s
prison population is expected to remain below its prison operational capacity
through fiscal year 2015, assuming no additional changes to its treatment and
diversion programs. Because of the reduced prison population, Texas has
been able to cancel contracts with county jails to house prisoners, which has
resulted in annual savings of approximately $36 million. Additionally, according
to the March 2010 presentation, offenders diverted from prison represent $292
million in avoided annual incarceration costs; about 2,000 more low-risk
offenders had been released on parole one year after the reform, but parole
revocations had declined by 27 percent since 2006; and the felony probation
population has increased by about 8 percent since before the reform, but the
yearly probation revocation rate has stayed about the same at 7.5 percent.
Arizona could consider expanding substance abuse treatment alternatives, either
by expanding the use of drug courts and/or establishing additional substance
abuse treatment alternatives. These additional alternatives could include counseling
services, in-patient beds, and secure residential treatment beds.

Home arrest with electronic monitoring could be expanded—The use
of home arrest with electronic monitoring for nonviolent, low-risk offenders is
almost nonexistent in Arizona (see textbox for definitions, page 42). Although
Arizona Revised Statutes (A.R.S.) §41-1604.13 allows the Board of Executive
Clemency to release certain nonviolent, first-time offenders to home arrest with
electronic monitoring, statute limits its use to persons who committed offenses

1 Council of State Governments Justice Center, 2009.

Office of the Auditor General
page

41

Home Arrest—A sanction that requires inmates to remain at their place of
residence at all times except for movement outside the residence for mandated
reasons, such as employment or drug screening. In Arizona, home arrest is
conditioned on electronic monitoring surveillance, participation in employment
and other beneficial activities, and submission to alcohol and drug tests.
Electronic Monitoring (EM)—A sanction that typically requires offenders to wear
a wrist or ankle bracelet or global positioning unit, use voice verification systems,
or wear or breathe into alcohol testing devices. EM may be passive, such as
requiring an offender to answer the telephone and verify his/her presence at
home, or active, emitting a constant signal to a home-monitoring device that
communicates the inmate’s location to a central computer. Although often used
with home arrest, EM is also used with other forms of offender supervision such
as probation or parole.
Source:

Auditor General staff review of A.R.S. §41-1604.13(D) and the National Law Enforcement and Corrections
Technology Center October 1999 bulletin.

before 1994. According to a department official, only three inmates were serving
time on home arrest as of June 30, 2010.
Some states that use home arrest with electronic monitoring have reduced
corrections spending without negatively impacting recidivism. For example:

•
According to the Florida
Department of
Corrections, Florida’s
home arrest program is
less expensive than
prison.

Florida—Florida began using home arrest with electronic monitoring in 1983
as an alternative to parole for offenders who needed more intensive supervision
after prison as well as for offenders who violated conditions of their probation
and would otherwise be sent to prison. According to the Florida Department of
Corrections (FDOC), the program is less expensive than prison and allows
offenders to remain active community members who can work to assist their
families and pay victim restitution. Further, according to a 2000 FDOC study,
offenders on home arrest did not pose any greater risk to the community than
probationers—both groups had reoffense rates of about 15 percent after 2
years.
In an effort to reduce state spending on prison costs, a 2009 Florida Office of
Program Policy Analysis and Government Accountability (OPPAGA) report
recommended the state expand various alternative sanctions, including home
arrest. The report estimated that about 350 additional offenders projected to
be sentenced to prison for probation violations might be eligible for home
arrest with electronic monitoring and other nonprison sanctions. OPPAGA said
if this were the case, the state would save $5.7 million annually because the
average cost to incarcerate an inmate in Florida is $16,410 per year, whereas
the cost to monitor an inmate on home arrest in the state is only $2,730 to
$5,285 per year, depending on the type of electronic monitoring device used.

State of Arizona
page

42

•

Mississippi—A judge can sentence offenders in Mississippi to home arrest
with electronic monitoring, or the Mississippi Department of Corrections
(MDOC) can place offenders on this sanction after they are sentenced to
prison. In fiscal year 2009, nearly 1,200 offenders were serving time on home
arrest in the state at a cost savings of nearly $29 per offender per day (prison
costs an average of $40.68 per day whereas home arrest costs $11.74 per day
in Mississippi). MDOC officials believe home arrest is a safe and effective
program and told auditors they are actively trying to expand the program to
alleviate pressure on the prison system.

Expanding home arrest in Arizona has been considered in recent years. The
January 2009 Appropriations Chairmen Budget Options report for fiscal years 2009
and 2010 proposed revising eligibility criteria to expand the State’s home arrest
program. The Department reported in April 2009 that almost 2,500 offenders—935
inmates who were incarcerated at the time, plus an additional 1,522 offenders the
Department estimated would be admitted in fiscal year 2010—would likely be
eligible for home arrest based on the proposed criteria at the time. The actual
number of eligible inmates would depend on the criteria adopted. In addition to
the offense date requirement, statute limits eligibility to inmates who have served
at least 6 months of their sentence; were convicted of a class 4, 5, or 6 felony that
did not involve intentionally or knowingly inflicting serious physical injury or using
or exhibiting a deadly weapon or dangerous instrument; were not convicted of a
sexual offense; have not previously been convicted of a felony; have violated
parole by committing a violation that was not chargeable or indictable as a criminal
offense; and are eligible for work furlough or parole.

Expanding home arrest in
Arizona has been
considered.

Placing more inmates on home arrest with electronic monitoring could reduce
costs. According to department estimates, home arrest could cost approximately
$19 per inmate, per day, including $7.50 for monitoring equipment. Although this
estimate is higher than the marginal cost of $12.60 to house, feed, and supervise
an additional inmate in prison, statute requires offenders to pay some home arrest
costs. Specifically, A.R.S. 41-1604.13(D) requires that offenders on home arrest
pay an electronic monitoring fee of between $1 per day and the total cost of
electronic monitoring and a home arrest supervision fee of at least $65 per month
if they have the ability to pay these fees. Depending on the amount offenders pay,
the daily cost of the program could be less than the marginal cost of a day in
prison. In addition, more savings could be realized if enough inmates were
diverted to close a prison unit. However, a significant number of inmates would
need to be diverted before a prison unit would close. As of August 31, 2010, the
Department had over 4,800 temporary beds in the prison system, and department
officials indicated that the Department would likely stop using these beds before
closing a unit.

Day reporting centers could be used—Many states and/or counties use day
reporting centers to reduce jail and prison populations and associated costs (see

Office of the Auditor General
page

43

textbox for definition).1 Georgia reported lower recidivism rates for offenders who
participated in day reporting centers rather than being incarcerated. Specifically:
Day reporting centers—A nonprison alternative that blends high levels of supervision
with intensive services and programming. Offenders typically report to the centers
during the day but sleep at home and are responsible for providing their own meals
and means of transportation to and from the centers. According to a 1995 National
Institute of Justice report that surveyed 114 day reporting centers in 22 states (all of the
centers that existed at the time), the centers have strict requirements for monitoring the
whereabouts and behavior of participating offenders, and most centers’ surveillance
policies include graduated phases of supervision, frequent on-site contact, close
monitoring of offenders when off-site, and vigilant surveillance of certain behaviors such
as drug use.
Although day reporting centers are commonly used for nonviolent offenders who need
substance abuse treatment, some states have also used them for arson, sex or
weapons offenses, and other violent offenses.
Source:

Auditor General review of the literature on day reporting centers; see Parent et al., 1995; Parent and Corbett,
1996; Craddock, 2000; Martin, Lurigio, and Olson, 2003.

•

Georgia—Georgia, which began using day reporting centers in 2001, had 13
day reporting centers state-wide as of August 2010. According to a Georgia
Department of Corrections (GDOC) official, a judge can sentence offenders to
these centers at initial sentencing or if they violate the conditions of their
probation. A 2005 Georgia State University study on a day reporting center in
Atlanta reported that offenders who completed the day reporting center program
had a recidivism rate of 9 percent compared to a rate of 31 percent for offenders
who did not complete the program and 20 percent for a comparison group of
released parolees.2,3 Further, Georgia has experienced a significant per-inmate
cost reduction by using the centers. According to GDOC officials, Georgia’s day
reporting centers only cost $16.50 per inmate, per day compared to the $48-perinmate-per-day cost of prison. GDOC officials were unable to give estimates for
the start-up costs associated with the centers because they already owned many
of the buildings and furniture used for the centers prior to their opening. However,
these officials did mention that construction and start-up costs are relatively low,
especially in comparison to prison construction and start-up costs because the
facilities are “basically a blend of office and classroom space,” which “are
relatively inexpensive to accommodate with furniture, fixtures, and equipment,
unlike [their] incarceration facilities.”

1 According to a 1995 report by the National Institute of Justice, there were 114 day reporting centers in 22 states as of

mid-1994. Although auditors did not determine whether these states continued to have day reporting centers, auditors
identified at least 10 states that had day reporting centers as of May 2010: Georgia, Illinois, Kansas, Nebraska, New
Jersey, Oregon, Pennsylvania, South Carolina, Utah, and West Virginia: see Parent et al., 1995.

2 See Finn, 2005
3 Recidivism numbers are based on a review of offenders discharged from either the day reporting center or the comparison

parole center between April 2001 and April 2003, beginning from their referral date through September 15, 2004.

State of Arizona
page

44

In Arizona, day reporting centers have been used at the county level. According
to a 1995 National Institute of Justice (Institute) report, Maricopa County
(County) began using day reporting centers in 1992 after a federal court
ordered the County to reduce its jail population.1 Although the Institute noted
that no formal impact evaluation had been completed on the County’s day
reporting center program, it reported that the return rate for participants
returned to jail for serious rule violations was quite low, especially compared to
the return rate for intensive supervision programs. Further, according to the
report, county staff estimated that the program had saved the equivalent of
35,426 days in jail between the time it opened the centers in 1992 and the time
the report data was collected in 1994. Based on county-reported per diem
costs for jail and day reporting centers ($37 and $16, respectively), the Institute
suggested that day reporting centers represented a significant potential cost
savings, but that evaluative research was needed to draw any conclusions
about program effectiveness. A 1999 study on the effectiveness of the County’s
day reporting center program in reducing DUI recidivism for repeat DUI
offenders found that, although the program was no more effective at reducing
recidivism than a standard probation program, it was more cost-effective and
helped reduce pressure on the county jail system.2 However, the study noted
that the day reporting center program offered general substance abuse
treatment rather than alcohol-specific treatment, which could have affected the
results. According to Maricopa County probation staff, the County stopped
using its centers in May 2002. Staff report that the program was discontinued
because county judges and prosecutors deemed fewer and fewer probationers
eligible for the program and because of budget considerations.

Maricopa County used
day reporting centers from
1992 to 2002.

State should further study expansion of nonprison
alternatives, including costs and needed legislative
action
Although the Department and the courts have statutory authority to establish
nonprison alternatives, further study should be conducted to identify the best mix of
alternatives. A.R.S. §41-1613 authorizes the Department to establish and operate
community correctional centers to provide housing, supervision, counseling, and
other correctional programs for persons in prison or on community supervision.
According to department officials, these centers could include day reporting centers,
work release centers, residential treatment centers, and halfway houses. Similarly,
A.R.S. §12-299.01 authorizes the county courts to establish and operate community
punishment programs for probationers that include noncustodial programs such as
house arrest, electronic monitoring, and drug and alcohol outpatient treatment;
residential programs such as restitution centers, halfway houses, and inpatient drug
or alcohol treatment; and individualized services, such as counseling and education.
1 See Parent et al., 1995
2 Jones and Lacey, 1999

Office of the Auditor General
page

45

Further study could
identify the best mix of
nonprison alternatives for
Arizona.

The Legislature could consider directing the Department and/or the courts to
conduct a study to identify the best mix of nonprison alternatives for Arizona, which
could be used in lieu of prison or in conjunction with earlier release, as well as
develop recommendations for nonprison alternatives. For example, before
Washington invested in nonprison alternatives, it commissioned a study of numerous
evidence-based programs for adult corrections, juvenile corrections, and crime
prevention to determine what evidenced-based alternatives could be used to reduce
the need for prison but still be fiscally sound and reduce future crime.1 The study
recommended that Washington adopt a moderate-to-aggressive portfolio of
evidenced-based options, which, if successfully implemented, would reduce future
prison construction significantly. Specifically, the study projected that Washington
could save about $2 billion and crime rates would be reduced. Cost benefits are a
mix of savings from lower incarceration rates and cost avoidance based on fewer
crimes committed in the future. The Legislature could also consider directing the
Department and the courts to monitor the cost and impact of any or all of the new
programs.
The costs of implementing alternatives would also need to be considered, including
startup costs. However, these costs could be lessened by requiring offenders to pay
some costs, especially for alternatives that allow them to work, such as home arrest
with electronic monitoring, and by using other funding sources or cost-saving
measures. For example, according to the National Institute of Justice’s 1995 review
of Maricopa County’s day reporting center program, the County reallocated
resources and developed new funding options to pay for the program.2 Specifically,
Maricopa County raised the charge for housing a federal inmate from $38 per day
to $78 per day and received legislative approval to use funds from a 1986 bond
issue for day reporting center facility acquisition. In addition, $150,000 in Bureau of
Justice Assistance money was also applied to the project. Since the 1986 bond
issue could be used only to improve the physical plant and not to support the
program, the County offered free rent in their buildings to treatment providers in
exchange for slots in the treatment programs for day reporting center offenders.
Depending on whether the Legislature provides funding for expanded nonprison
alternatives and which alternatives are expanded, some statutes will need to be
revised. For example, the Legislature would need to revise statute to expand eligibility
for the home arrest program.

1 See Aos, Miller, & Drake, 2006
2 See Parent et al., 1995

State of Arizona
page

46

Chapter 5
Option 4—Reducing revocations from parole
violations
A fourth option available for reducing prison population growth is to reduce prison
admissions that result from offenders who violate the terms of their community
supervision (commonly referred to as parole). After serving at least 85 percent of their
sentences in prison, most Arizona inmates are conditionally released to the
community under Department of Corrections (Department) supervision. Released
inmates spend a median of about 5 months on community supervision. However,
their parole can be revoked and they can be returned to prison for violating the
conditions of their release, such as missing appointments with parole officers, using
illegal substances, or engaging in criminal behavior. These violations accounted for
15 percent of the State’s prison admittances in fiscal year 2010. Although the
Department has developed policies and procedures to address parole violations,
expanding the Department’s alternatives for responding to them may help reduce
prison admissions and associated costs. These include nonprison alternatives, such
as those mentioned in Chapter 4 (see pages 37 through 46), residential treatment
facilities, or other secure facilities. The Department has authority to establish
alternative sanctions for parole violators and is in the process of studying potential
options. Once it completes its study, the Department should present its findings to
the Governor and Legislature for consideration and expand its use of nonprison
sanctions in accordance with the direction it receives from state policymakers.

Most inmates serve part of sentence in community
Most inmates serve about 15 percent or less of their sentence in the community
under department supervision. As discussed in Chapter 3 (see pages 23 through
35), Arizona abolished discretionary parole as a release mechanism for offenses
committed after 1993 and requires offenders to serve at least 85 percent of their
sentence before becoming eligible for release to the community. Although release
eligibility depends on prisoner behavior and other factors, department staff indicated

Most Arizona inmates
serve about 15 percent or
less of their sentence on
community supervision.

Office of the Auditor General
page

47

that most inmates are released after serving the minimum requirement. After release,
department community corrections personnel (parole officers) evaluate released
inmates’ risk to the community and assign a level of supervision. In addition, inmates
must agree in writing to follow several conditions of supervision and release (see
textbox for examples). According to the Department, there were approximately 7,500
inmates on community supervision each month in fiscal year 2010. Based on
auditors’ analysis of department data, inmates serve a median of about 5 months on
parole.

Examples of community release conditions
•
•
•
•
•
•
•

Maintaining contact with parole officers
Living in approved housing
Securing and maintaining employment
Abstaining from alcohol and drugs and submitting to drug tests
Obeying all laws
Not engaging in violent or threatening behavior
Not possessing or using firearms or dangerous weapons

Source:

Auditor General staff review of the Department’s community supervision agreement.

Parole violations contribute to prison population growth

Parole revocations
accounted for about 15
percent of prison
admissions in fiscal year
2010.

Offenders can have their parole revoked and be sent back to prison for violating the
conditions of their release. According to the Department, parole revocations
accounted for about 15 percent of prison admissions in fiscal year 2010 (up from 14
percent in fiscal year 2009). Parole can be revoked for any number of violations,
including absconding, avoiding the parole officer, committing new crimes
(misdemeanors or felonies), failing drug tests, carrying a dangerous weapon, or
failing to maintain employment.1,2 According to department information for calendar
year 2009, the most common violations are absconding and substance abuse. The
Department initiates the revocation process by issuing a warrant for a parole
violator’s arrest. When arrested, the offender is either returned to prison or, in some
cases, may remain in the community to await a parole revocation hearing with the
Board of Executive Clemency (Board; see textbox, page 49). According to
department policies, department staff may allow a parole violator to remain in the
community if he/she does not pose a threat to self or others and will likely appear at
the hearing. The Board is responsible for determining whether offenders who have
violated parole should finish serving their sentences in prison or remain in the
community.
1 Absconding is where an inmate’s location is unknown and/or the inmate fails to maintain contact with his/her parole

officer.

2 Although community supervision can be revoked for absconding or committing new crimes, the 15 percent of prison

admissions from community supervision revocations in fiscal year 2010 does not include absconders still at large or
supervised offenders convicted of new felonies and returned to prison.

State of Arizona
page

48

Arizona revokes a relatively low
percentage of parolees compared
to the national percentage.
Formerly the Board of Pardons and Paroles,
According to a Bureau of Justice
the Board of Executive Clemency is a fiveStatistics report, approximately 25
member body appointed by the Governor
percent of offenders exiting parole
that has exclusive power to pass upon and
nationally in 2008 (excluding federal
recommend reprieves, commutations,
parolees) returned to incarceration
paroles, and pardons for persons who have
with parole revocations; the rate for
committed offenses prior to 1994. The
Board also has authority to revoke
Arizona was 15.4 percent.1 Arizona’s
community supervision for offenders who
significant absconder population,
have violated the terms of their release.
which the Department tracks
separately and according to the
Source: Auditor General staff review of A.R.S. §§31-401 and
same Bureau of Justice Statistics
31-402 and the Board of Executive Clemency
report was the third highest in the
Web site.
country in 2008, may partially
explain why its revocation rate is
lower than the national rate. Although Arizona revokes a relatively small percentage
of its parolees, these revocations impose costs to the State. Based on auditors’
analysis of department data, inmates who had their parole revoked in 2008 spent a
median of more than 3 months in prison, which cost approximately $6,300 per
revoked inmate (or approximately $1,222 per revoked inmate based on marginal
costs) compared with approximately $774 per offender on community supervision.2
Parolees who were returned to prison for parole violations in 2008, but subsequently
released back to the community by the Board of Executive Clemency, spent a
median of more than 2 months in prison before they were re-released.

Board of Executive Clemency

Expanding range of nonprison alternatives for parole
violators could help reduce prison population growth
The Department provides guidance to its parole officers on what actions to take to
address parole violations and when to return a parolee to prison. However, its
methods for dealing with violators are limited by a lack of nonprison alternatives to
confine parolees who face revocation. Other states use nonprison alternatives to
address parole violations.

Department provides guidance on when to return parole violators to
prison or use other sanctions—Department policies and procedures
guide parole officers on when to initiate the revocation process or use other
sanctions in response to violations. Similar to some other states, the Department
can use a variety of sanctions to address violations. These sanctions, commonly
called graduated or intermediate sanctions (see textbox, page 50), include verbal
1 See Glaze and Bonczar, 2009
2 These cost estimates are based on the Department’s reported 2009 per capita costs.

Office of the Auditor General
page

49

or written reprimands, increased supervision or
programming, increased drug testing, community
service, curfews, or referrals to the Department’s
Community Accountability Pilot Program (see textbox).
According to department guidelines, when and how
these alternative sanctions are used depends on the
seriousness of the violations, the offender’s level of
supervision, and the number of violations an offender
has committed. However, department policy requires
parole officers to request warrants for numerous specified
violations. These include refusal to sign conditions of
supervision and release, failure to contact the parole
office within one working day of release to the community,
absconding, violations involving firearms or dangerous
Pew Center on the States, Policy Framework to Strengthen
weapons, verified personal injury to another person or
Community Corrections, 2008.
threat of violence, arson, sex offense behavior, and all
new felony arrests. Although the Department has a range
of sanctions to address parole violations, it has no nonprison alternative facilities
available, either for use as a graduated sanction or to hold offenders once it
begins the parole revocation process.

Graduated sanctions include a wide range of
actions that can be taken to swiftly respond to
violations without returning parole violators to
prison. They include, but are not limited to,
electronic supervision tools; drug and alcohol
testing or monitoring; day or evening reporting
centers; restitution centers; forfeiture of earned
compliance credits; rehabilitative interventions
such as substance abuse or mental health
treatment; reporting requirements to supervision
officers; community service or work crews; secure
or unsecure residential treatment facilities or
halfway houses; and short-term or intermittent
incarceration.
Source:

Community Accountability Pilot Program (CAPP)
Laws 2004, Ch. 204, required the Department to establish the CAPP for
eligible offenders. Eligible offenders must be either nonviolent but at a
high risk of reoffending, or a lower-risk individual with a history of mental
health or substance abuse issues and must be referred to the program
by the Department. Offenders referred to this program are placed on
electronic monitoring and provided life skills training, substance abuse
education, and help finding employment. They may participate in the
program for up to 90 days and remain on parole under the
Department’s supervision. As required by law, the Department contracts
with a vendor to provide the CAPP services. The CAPP is only available
to offenders in Maricopa County, and, according to the Department, 96
offenders were admitted to the program in fiscal year 2010.
Source: Auditor General staff analysis of Laws 2004, Ch. 204, and department eligibility criteria.

Other states have adopted nonprison alternatives for parole
violators—Some states have established various facilities to house parole
violators instead of returning them to prison, including residential treatment
facilities, day reporting centers, halfway houses, and assessment facilities. For
example:

State of Arizona
page

50

•

•

As discussed in Chapter 4 (see pages 37 through 46), Texas has expanded its
use of substance abuse treatment and prison diversion programs, which has
helped reduce its prison population and the number of parole revocations.
According to a 2009 Council of State Governments Justice Center report, this
expansion of programs accompanied an emphasis on releasing more eligible
inmates on parole and allocating resources toward addressing their needs
once on parole.1 This included expanding Texas’ use of intermediate sanction
facilities for parole violators. Intermediate sanction facilities are secure facilities
used to sanction parole violators instead of revoking them to prison. Parole
violators are confined in these facilities for an average of 60 days. According
to the Texas Legislative Budget Board Criminal Justice Uniform Cost Report,
Fiscal Years 2006—2008, in fiscal year 2008, intermediate sanction facilities
cost between $35.45 per day (privately contracted) and $41.29 per day (staterun), compared with the average cost of $47.50 per day for incarceration in a
Texas state prison.
New Jersey uses nonprison facilities to house parole violators awaiting parole
revocation hearings. According to a 2010 Sentencing Project report, in July
2008, New Jersey began using regional assessment centers for parole violators,
which are designed to confine up to 45 people at a time for 15 to 30 days of
lockdown.2,3 During their confinement, violators are assessed for their mental
health, social, familial, and economic needs, and risk to reoffend. These
assessments help the parole board make more informed decisions, which has
resulted in fewer revocations. The 2010 Sentencing Project report also noted
that of the 810 parolees assessed at these centers by February 2009, only 46
percent were returned to prison, compared to a return rate of 81 percent before
their use. Additionally, according to the New Jersey State Parole Board’s fiscal
year 2009 annual report, an additional review of 181 parolees who continued on
parole after being assessed in the centers found that 73.4 percent of them either
successfully completed parole or, if still on parole, had not committed additional
violations at the time of the review. The remainder eventually returned to prison.
Finally, according to this annual report and a New Jersey parole official, the
centers saved New Jersey an estimated $10 million in fiscal year 2009 because
of decreased parole revocations and its use of regional assessment centers to
hold parole violators rather than county jails.

New Jersey uses regional
assessment centers for
parolees facing revocation.

Arizona could use similar facilities as a graduated sanction or as holding facilities for
parole violators awaiting a revocation hearing. In addition, the nonprison alternatives
discussed in Chapter 4 (see pages 37 through 46), such as day reporting centers
and home arrest with electronic monitoring, could also be used for parole violators.

1 Council of State Governments Justice Center, 2009
2 See Greene & Mauer, 2010
3 The Sentencing Project is a criminal justice policy research and advocacy firm.

Office of the Auditor General
page

51

Expanding alternatives for parole violations would require
action

The Department is
studying options for
expanding alternative
sanctions for parole
violators.

As discussed in Chapter 4 (see pages 37 through 46), statute authorizes the
Department to establish and operate community correctional centers to provide
housing, supervision, counseling, and other correctional programs for offenders on
community supervision. According to department officials, the Department previously
used such centers but stopped doing so in the 1980s because of funding limitations.
Department officials indicated that they are in the process of studying potential
options for expanding the use of nonprison sanctions for parole violators. The
Department should complete this study and present its findings to the Governor and
Legislature for consideration. The Department should then expand its use of
nonprison sanctions in accordance with the direction it receives from state
policymakers.
Finally, if additional nonprison sanctions are implemented, the Department should
incorporate their use in its community supervision policies and procedures. Other
states have incorporated the use of formal sanction grids into their parole procedures.
For example, Ohio has used its Progressive Sanction Grid to provide guidance in
imposing sanctions based on offender risk and violation severity since 2005.1 In
2008, California began using the Parole Violation Decision Making Instrument, a
computer-based instrument that identifies a range of recommended responses to
each parole violation based on the offender’s risk level and the severity of the
violation.2

1 See Martin & Van Dine, 2008
2 See Murphy & Turner, 2010

State of Arizona
page

52

Chapter 6
Recommendations for legislative and department
consideration
The Legislature could consider a number of options for addressing Arizona’s growing
prison population. These options are not mutually exclusive and include the following:

•

Option 1: The Legislature could continue to expand the prison system, either by
constructing new prison facilities and/or contracting for more private beds. If the
Legislature decides to expand the prison system, it should consider directing
the Department of Corrections (Department) to further study and analyze the
costs for the State to build and operate prison facilities compared to contracting
with private prisons to determine which option would be more cost-effective
while still ensuring public safety.

•

Option 2: The Legislature could consider diverting more nonviolent, low-risk
offenders from prison and/or reducing the time they serve—alternatives that
may require changes to the State’s sentencing laws. Specifically:

°

Similar to Arizona Revised Statutes (A.R.S.) §13-901.01, which requires
nonviolent persons convicted of a first or second offense for the personal
possession or use of drugs to be sentenced to probation and mandatory
treatment, the Legislature could consider revising statute to expand
diversion opportunities to other nonviolent, low-risk offenders, particularly
those whose crimes are related to substance abuse. In order to divert more
nonviolent, low-risk offenders from prison, the Legislature may need to
consider revising some of the State’s sentencing laws.

°

The Legislature could consider expanding early release options, such as
reducing the time served requirement for nonviolent, low-risk offenders and
establishing earned time credits. These options would also require changes
to the State’s sentencing laws.

°

If the Legislature expands diversion or early release options, it should also
consider taking the following steps:

Office of the Auditor General
page

53

°

•

·

Further defining diversion and/or early release eligibility criteria for other
nonviolent, low-risk offenders in statute, and/or

·

Ensuring the use of valid and reliable risk assessment tools to determine
offender eligibility for diversion and/or early release.
The Legislature could consider establishing a permanent sentencing
commission to assist in reviewing and recommending changes to the
State’s sentencing laws. Other possible functions this commission could
perform include determining eligibility criteria for diversion, recommending
guidelines for determining appropriate candidates for alternative sanctions,
and monitoring reform results to ensure they are having the intended effect.
If the Legislature establishes a sentencing commission, it should consider
including representatives from all criminal justice system stakeholders.

Option 3: The Legislature could consider using more nonprison alternatives for
nonviolent, low-risk offenders. This could include:

°

Expanding substance abuse treatment alternatives by expanding the use of
drug courts and/or establishing additional substance abuse treatment
alternatives. This might include providing additional counseling services,
in-patient beds, and secure residential treatment beds.

°

Expanding the use of home arrest with electronic monitoring.

°

Establishing day reporting centers.

These or other alternatives could be used in lieu of prison sentences or in
conjunction with earlier release. The Legislature could consider directing the
Department and/or the courts to further study nonprison alternatives and
develop recommendations for expanding their use, which should include an
evaluation of the costs of these alternatives. Additionally, the Legislature could
direct the Department and the courts to monitor the cost and impact of any
nonprison alternatives established. Depending on whether the Legislature
provides funding for expanded nonprison alternatives and which alternatives are
expanded, some statutes will need to be revised, such as the home arrest
statute.

•

Option 4: Expanding nonprison alternatives for parole violators would require
the following actions:

°

State of Arizona
page

54

The Department should complete its study of potential options for expanding
the use of nonprison alternatives for parole violators and present its findings
to the Governor and Legislature for consideration. The Department should
then expand its use of nonprison sanctions in accordance with the direction
it receives from state policymakers.

°

If nonprison alternatives or sanctions are implemented, the Department
should incorporate the use of these additional sanctions in its community
supervision policies and procedures.

Office of the Auditor General
page

55

State of Arizona
page

56

APPENDIX A
Data and methodology
Auditors used various methods to study the issues addressed in this report. These
methods included interviewing Department of Corrections (Department) officials and
staff, Arizona Department of Administration (ADOA) staff, Joint Legislative Budget
Committee (JLBC) and Governor’s Office of Strategic Planning and Budget staff, and
various stakeholders; reviewing JLBC reports; reviewing statutes, department
orders, director’s instructions, department guidelines regarding supervising parole
violators, and other department documentation; reviewing Arizona crime data for
1960 through 2008 obtained from the Bureau of Justice Statistics Web site, national
crime data for 2006 through 2008 obtained from the Federal Bureau of Investigation
Web site, and Arizona population estimates from the Arizona Department of
Economic Security Web site; calculating Arizona’s imprisonment rate for fiscal years
1980 through 2008; and attending and/or reviewing the minutes of the December
2009 and May 2010 Arizona House of Representatives Study Committee on
Sentencing. In addition, auditors used the following data and methods:

•

•

Data Sources—Auditors obtained the following data downloads from the
Department’s Adult Inmate Management System (AIMS):

°

One-day census of all Arizona prison inmates as of December 31, 2009;

°

All inmates admitted to Arizona prisons between January 1, 1985 and June
30, 2009; and

°

All Arizona prison inmates released between January 1, 1990 and December
31, 2009.

Data validation—To validate the AIMS data, auditors assessed the Department’s
internal controls by reviewing applicable policies and procedures and interviewing
various staff and management responsible for the data. Auditors also tested a
sample of records in the data against inmates’ hard copy files to validate
specific fields used in auditors’ analyses. This test work included 10 inmate

Office of the Auditor General
page

a-i

records from the one-day census data and 27 records from release data.
Auditors found some errors in some of the data fields. However, auditors either
did not use those fields or determined that these errors would not be material to
overall conclusions. In general, auditors concluded that the AIMS data was
sufficiently reliable for audit purposes

State of Arizona
page

a-ii

•

Data analysis—Auditors analyzed the AIMS data to determine the demographic
makeup of the prison population as of December 31, 2009; identify nonviolent
inmates as of December 31, 2009, who appeared to be at lower risk for
committing new felony offenses; review sentence lengths for admitted offenders
over time; determine the length and percentage of time served by violent and
nonviolent inmates released before and under Arizona’s truth-in-sentencing
laws; determine the percentage of violent and nonviolent prison admissions
over time; determine the length of time offenders spend on community
supervision; and determine the length of time offenders who have had their
parole revoked spend in prison.

•

Review of department reports (unaudited)—To obtain and review historical
Arizona State prison population, cost, and sentencing guideline information,
auditors reviewed several department reports. These reports included per
capita reports, bed plan reports, a 1992 sentencing report, a 2006 recidivism
report, annual reports from fiscal years 1983 through 2003, daily count sheets,
Corrections at a Glance reports, inmate admittance and release reports, and a
2010 prison population trend report. Auditors also assessed the reliability of the
per capita costs reported in the Department’s Fiscal Year 2009 Operating Per
Capita Cost Report by reviewing the report for mathematical accuracy and
internal consistency and reconciling a sample of reported costs to the Arizona
Financial Information System. Auditors concluded that the report is based on
reasonable methodology, appears to be materially mathematically accurate and
internally consistent, and overall appears reasonably accurate. The test work
was limited to the unadjusted per capita costs.

•

Observation of prison facilities—Auditors conducted observations at the
following three prison facilities: Arizona State Prison Complex (ASPC) Perryville
in Goodyear, which holds most of the State’s female prisoners; Phoenix West, a
privately operated specialty DUI prison for minimum-security male inmates
located in Phoenix; and ASPC Eyman in Florence, which holds male prisoners
in medium, maximum, and close custody.

•

Analysis of prison construction costs—Auditors developed cost estimates for
expanding the State’s prison system using a combination of state and private
facilities to meet projected prison population growth between fiscal years 2012
and 2017. This included an analysis of ADOA prison construction cost estimates,
ADOA-reported actual costs for the 2010 4,000-bed expansion project, the

Department’s Fiscal Year 2011 Budget Request Decision Package, and the
Department’s Fiscal Year 2009 Operating Per Capita Cost Report.

•

Literature review—Auditors reviewed literature and other reports on prison
population growth and its causes, privatization of prisons, incarceration and
crime, nonprison alternatives, and other states’ efforts to address prison
population growth. Auditors also reviewed prior studies and reports on Arizona’s
prison population and sentencing policies. See Appendix B, pages b-i through
b-iv, for references cited in the report.

•

Other state information—In addition to reviewing literature and other reports
regarding other states’ efforts to address prison population growth, auditors
also interviewed representatives, obtained information, and/or reviewed the
Web sites of nine states. These states were selected based on actions taken to
address prison population growth. States reviewed were Florida, Georgia,
Hawaii, Mississippi, New York, New Jersey, Texas, Vermont, and Virginia.

Office of the Auditor General
page

a-iii

State of Arizona

APPENDIX B
References
Alternatives to Sentencing Workgroup. (2005). Final report. Phoenix, AZ: Arizona House of
Representatives.
Aos, S. Miller, M, & Drake, E. (2006). Evidence-based public policy options to reduce
future prison construction, criminal justice costs, and crime rates. Olympia, WA:
Washington State Institute for Public Policy.
Arizona Supreme Court, Administrative Office of the Courts, Adult Probation Services
Division. (2006). Drug Treatment and Education Fund: Report detailing fiscal year
2005. Phoenix, AZ: Author.
Austin, J., & Coventry, G. (2001). Emerging issues on privatized prisons. Washington, D.C.:
U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Assis
tance.
Austin, J., & Fabelo, T. (2004). The diminishing returns of increased incarceration: A blue
print to improve public safety and reduce costs. Washington, D.C.: The JFA
Institute.
Banks, D., & Gottfredson, D.C. (2003). The effects of drug treatment and supervision on
time to rearrest among drug treatment court participants. Journal of Drug Issues,
385-412.
Bryne, J.M. (2009, July 10). A review of the evidence on the effectiveness of alternative
sanctions and an assessment of the likely impact of Federal Sentencing Guideline
reform on public safety [Written Testimony before the United States Sentencing
Commission]. New York, NY: Regional Hearings, U.S. Court of International
Trade Ceremonial Courtroom.
Council of State Governments, Justice Center. (2009). Justice reinvestment in Texas: As
sessing the impact of the 2007 Justice Reinvestment Initiative. Lexington, KY:
Author.
Council of State Governments, Justice Center. (2007). Justice reinvestment state brief:
Texas. Lexington, KY: Author.
Craddock, A. (2000). Exploratory analysis of client outcomes, costs, and benefits of day
reporting centers-Final Report. Terre Haute, IN: Indiana State University, Department
of Criminology.

Office of the Auditor General
page

b-i

Demleitner, N.V. (2005). Smart public policy: Replacing imprisonment with targeted non
prison sentences and collateral sanctions. Stanford Law Review, 58, 339-360.
Ditton, P., & Wilson, D.J. (1999), Bureau of Justice Statistics special report: Truth in
sentencing in state prisons [NCJ 170032]. Washington, D.C.: U.S. Department of
Justice, Office of Justice Programs, Bureau of Justice Statistics.
Drake, E.K., Aos, S. & Miller, M.G. (2009). Evidence-based public policy options to reduce
crime and criminal justice costs: Implications in Washington State. Victims and Of
fenders, 4, 170-196. doi: 10.1080/15564880802612615
Finn, M.A. (2005). Atlanta Day Reporting Center outcome evaluation. Atlanta, GA: Georgia
Day Reporting Center Executive Committee.
Fischer, D. R. (2010). Prisoners in Arizona: A profile of the inmate population. Phoenix, AZ:
Arizona Prosecuting Attorneys' Advisory Council.
Fischer, D.R., & Thaker, A. (1992). Mandatory sentencing study. Phoenix, AZ: Arizona De
partment of Corrections.
Glaze, L.E., & Bonczar, T.P. (2009). Probation and parole in the United States, 2008. Wash
ington, D.C.: U.S. Department of Justice, Office of Justice Programs, Bureau of
Justice Statistics.
Gottfredson, D.C., Najaka, S.S., & Kearley, B. (2003). Effectiveness of drug treatment
courts: Evidence from a randomized trial [Electronic version]. Criminology & Public
Policy, 2(2), 171-196.
Greene, J., & Mauer, M. (2010). Downscaling prisons: Lessons from four states. Washing
ton, D.C.: Justice Strategies & The Sentencing Project.
Greene, J., & Pranis, K. (2004). Arizona prison crisis: A call for smart on crime solutions.
Washington, D.C.: Families Against Mandatory Minimums.
Guzman, C., Krisberg, B., & Tsukida. (2008). Accelerated release: A literature review.
Focus: Views from the National Council on Crime and Delinquency.
Retrieved July 21, 2010 from, http://www.nccd-crc.org/nccd/pubs/2008_focus_ac
celeratedRlease.pdf
Institute for Rational Public Policy, Inc. (1991). Arizona criminal code and corrections study:
Final report to the Legislative Council. Phoenix, AZ: State of Arizona, Legislative
Council.
Jones, R.K., & Lacey, J.H. (1999). Final report: Evaluation of a day reporting center for
repeat DWI offenders [DOT HS 808 989]. Washington, D.C.: U.S. Department of
Transportation National Highway Traffic Safety Administration.
Justice Policy Institute. (2010). How to safely reduce prison populations and support
people returning to their communities. Retrieved July 21, 2010, from,
http://www.justicepolicy.org/images/upload/10-06_FAC_ForImmediateRelease_PSAC.pdf
Kleiman, M, Ostrom, B.J., & Cheesman, F.L. (2007). Using risk assessment to inform sen
encing decisions for nonviolent offenders in Virginia. Crime & Delinquency, 53(1),
106-132.

State of Arizona
page

b-ii

Langan, P.A., & Levin, D.J. (2002). Bureau of Justice Statistics special report: Recidivism of
prisoners released in 1994 [NCJ 193427]. Washington, D.C.: U.S. Department of
Justice, Office of Justice Programs, Bureau of Justice Statistics.
Lawrence, A. (2009). Cutting corrections costs: Earned time policies for state prisoners.
Denver, CO: National Conference of State Legislatures.
Liedka, R.V., Piehl, A.M., & Useem, B. (2006). The crime-control effect of incarceration:
Does scale matter? [Electronic version]. Criminology & Public Policy, 5(2), 245- 276.
Lundahl, B.W, Kunz, C., Brownell, C., Harris, N., & Van Vleet, R. (2009). Prison
privatization: A meta-analysis of cost and quality of confinement indicators.
Research on Social Work Practice, 19(4), 383-394.
Martin, B., & Van Dine, S. (2008). Examining the impact of Ohio's progressive sanction grid:
Final report. Ohio Department of Rehabilitation and Correction, Bureau of
Research, Office of Policy and Offender Reentry.
Martin, C., Lurrigio, A.J., & Olson, D.E. (2003). An examination of rearrests and
reincarcerations among discharged day reporting center clients. Federal Probation
67(1), 24-30.
Merritt, N., Fain, T., & Turner, S. (2006). Oregon's get tough sentencing reform: A lesson in
justice system adaptation. Criminology and Public Policy, 5(1), 5-36.
Murphy, A., & Turner, S. (2010). Parole violation decision-making instrument (PVDMI)
process evaluation. University of California, Irvine, Center for Evidence-Based
Corrections.
Parent, D., Byrne, J., Tsarfaty, V., Valade, L., & Esselman, J. (1995). Day reporting centers:
Volume 1. Washington, D.C.: U.S. Department of Justice, Office of Justice Pro
grams, National Institute of Justice.
Parent, D. G., & Corbett, R.P. (1996). Day reporting centers: An evolving intermediate
sanction. Federal Probation, 60(4), 51-55.
Pew Center on the States. (2010). Prison count 2010: State population declines for the
first time in 38 years. Washington, D.C.: Pew Charitable Trusts.
Pew Center on the States. (2009). One in 31: The long reach of American corrections.
Washington, D.C.: Pew Charitable Trusts.
Pew Center on the States. (2008). Policy framework to strengthen community corrections.
Washington, D.C.: Pew Charitable Trusts.
Scott-Hayward, C.S. (2009). The fiscal crisis in corrections: Rethinking policies and prac
tices. New York: Vera Institute of Justice.
Shaffer, D.K. (2006). Reconsidering drug court effectiveness: A meta-analytic review. Un
published doctoral dissertation, University of Cincinnati.
Stemen, D. (2007). Reconsidering incarceration: New directions for reducing crime. New
York: Vera Institute of Justice.
Stemen, D., Rengifo, A., Wilson, J. (2005). Of fragmentation and ferment; The impact of
state sentencing policies on incarceration rates. New York: Vera Institute of Justice.

Office of the Auditor General
page

b-iii

Tonry, M. (2009). The mostly unintended effects of mandatory penalties: Two centuries of
consistent findings. Crime and Justice, 38, 65-114.
Ulmer, J.T., Kurlychek, M.C., & Kramer, J.H. (2007). Prosecutorial discretion and the impo
sition of mandatory minimum sentences. Journal of Research in Crime and Delin
quency, 44(4), 427-458. doi: 10.1177/0022427807305853.
United States Government Accountability Office. (2005). Adult drug courts: Evidence
indicates recidivism reductions and mixed results for other outcomes. Washington,
D.C.: Author.
Warren, R.K. (2007). Evidence-based practice to reduce recidivism: Implications for state
judiciaries. Washington, D.C.: United States Department of Justice, National
Institute of Corrections.
West, H.C. (2010). Prisoners at yearend 2009—Advanced counts [NCJ 230189].
Washington, D.C.: U.S. Department of Justice, Office of Justice Programs,
Bureau of Justice Statistics.
Wilson, D.B., Mitchell, O., MacKenzie, D. (n.d.) Drug court effects on recidivism [in-press
draft]. Administration of Justice, George Mason University, Manassas, VA.
Wool, J., & Stemen, D. (2004). Changing fortunes or changing attitudes? Sentencing and
corrections reforms in 2003. New York: Vera Institute of Justice.

State of Arizona
page

b-iv

AGENCY RESPONSE

Performance Audit Division reports issued within the last 24 months
08-05
08-06
09-01

09-02
09-03
09-04
09-05
09-06
09-07

09-08

Arizona Biomedical Research
Commission
Board of Podiatry Examiners
Department of Health Services,
Division of Licensing Services—
Healthcare and Child Care
Facility Licensing Fees
Arizona Department of Juvenile
Corrections—Rehabilitation and
Community Re-entry Programs
Maricopa County Special Health
Care District
Arizona Sports and Tourism
Authority
State Compensation Fund
Gila County Transportation
Excise Tax
Department of Health Services,
Division of Behavioral Health
Services—Substance Abuse
Treatment Programs
Arizona Department of Liquor
Licenses and Control

09-09

09-10
09-11
10-01
10-02
10-03
10-04
10-05
10-06
10-07

Future Performance Audit Division reports
Office of Pest Management—Regulation

Arizona Department of Juvenile
Corrections—Suicide Prevention
and Violence and Abuse
Reduction Efforts
Arizona Department of Juvenile
Corrections—Sunset Factors
Department of Health Services—
Sunset Factors
Office of Pest Management—
Restructuring
Department of Public Safety—
Photo Enforcement Program
Arizona State Lottery
Commission and Arizona State
Lottery
Department of Agriculture—
Food Safety and Quality
Assurance Inspection Programs
Arizona Department of Housing
Board of Chiropracitc Examiners
Department of Agriculture—
Sunset Factors