R E S E A R C H
R E P O R T
April 2004

The New Landscape of
Imprisonment:
Mapping America’s
Prison Expansion
Sarah Lawrence
Jeremy Travis

URBAN INSTITUTE

Justice Policy Center

URBAN INSTITUTE

Justice Policy Center
2100 M Street NW
Washington, DC 20037
(202) 833-7200
www.urban.org
© 2004 Urban Institute

The views expressed in this report are those of the authors and should not be attributed to the
Urban Institute, its trustees, or its funders. The Justice Policy Center carries out nonpartisan
research to inform the national dialogue on crime, justice, and community safety.
JPC Publication: CPR04 0121

About the Authors
Sarah Lawrence is an independent consultant and former research associate at the Urban Institute’s
Justice Policy Center. Ms. Lawrence’s professional experiences focus on linking criminal justice research
with policymakers and practitioners. Her primary areas of interest include prisoner reentry, corrections
reform, and community supervision. She is co-author of several publications, including “Beyond the
Prison Gates: The State of Parole in America” and “Prison Programming: What It Can Do and Why It Is
Needed.” She is currently serving as a consultant to the Governor’s Commission on Corrections Reform
in the State of Massachusetts. Ms. Lawrence holds a Masters in Public Policy from the University of
California, Berkeley and a Bachelor of Science in Engineering from Cornell University.
Jeremy Travis is a senior fellow at the Urban Institute, affiliated with the Justice Policy Center. He is
developing research and policy agendas on understanding crime within the context of the community,
new concepts of the agencies of justice, sentencing and prisoner reentry, and international crime. Mr.
Travis is co chair of the Reentry Roundtable, a group of prominent academics, practitioners, service
providers, and community leaders working to advance policies and innovations on prisoner reentry that
reflect solid research. Before joining the Urban Institute, he directed the National Institute of Justice (NIJ),
the research arm of the U.S. Department of Justice, from 1994 to 2000. Prior to his service at NIJ, he was
deputy commissioner for legal matters of the New York City Police Department, chair of the New York
City Chancellor's Advisory Panel on School Safety, chief counsel to the Subcommittee on Criminal
Justice of the House of Representatives Committee on the Judiciary, and special advisor to the mayor of
New York City. Mr. Travis earned his JD, cum laude, from the New York University School of Law and
also holds an MPA from the New York University Wagner Graduate School of Public Service and a BA in
American Studies, cum laude, from Yale College.

The New Landscape of Imprisonment

i

Contents
Acknowledgments ....................................................................................................iv
Executive Summary...................................................................................................1
1

Introduction.....................................................................................................2
Research Question and Data Sources ........................................................................................ 4
Types of Correctional Facilities.................................................................................................... 6
Counting Prisoners in the Decennial Census .............................................................................. 6

2

Overall Growth in the Number of Prisons ....................................................8

3

Changes in the General Population ............................................................11

4

Prison Growth by State ................................................................................13
Prison Distribution within a State ............................................................................................... 13
Prison Location by County Classification................................................................................... 15
Prisoner Location by County Classification ............................................................................... 16

5

Mapping County-Level Changes in the Top 10 States ..............................19
Prison Expansion in California................................................................................................... 20
Prison Expansion in Colorado ................................................................................................... 21
Prison Expansion in Florida ....................................................................................................... 22
PRison Expansion in Georgia.................................................................................................... 23
Prison Expansion in Illinois ........................................................................................................ 24
Prison Expansion in Michigan.................................................................................................... 25
Prison Expansion in Missouri..................................................................................................... 26
Prison Expansion in New York .................................................................................................. 27
Prison Expansion in Ohio .......................................................................................................... 28
Prison Expansion in Texas ........................................................................................................ 29
Summary of Prison Expansion in the Top 10 States ................................................................. 30

6

Effects on Small Communities ....................................................................31

7

Mapping County of Sentencing versus County of Incarceration .............33
California ................................................................................................................................... 33
Florida ....................................................................................................................................... 34
Georgia...................................................................................................................................... 35
Ohio........................................................................................................................................... 35
Texas......................................................................................................................................... 35
Summary of Sentencing Locations versus Incarceration Locations........................................... 36

8

Conclusion ....................................................................................................42

Bibliography.............................................................................................................44
Appendices ..............................................................................................................48
Appendix A. Types of facilities in state and federal correctional systems.................................. 49
Appendix B. Share of counties and share of residents in non-metro counties by state ............. 50
Appendix C. State-level data profiles......................................................................................... 51

ii

The New Landscape of Imprisonment

Table of Figures
Figure 1. Primary data sources........................................................................................... 5
Figure 2. Defining a “metro” county .................................................................................... 5
Figure 3. Growth of state and federal prison populations, 1980, 1990, 2000 ..................... 7
Figure 4. Number of state prisons, 1923–2000................................................................... 8
Figure 5. Percent of growth in number of prisons nationwide, 1979–2000 ......................... 9
Figure 6. States with highest growth in number of prisons ................................................. 10
Figure 7. Percent of counties that are non-metro counties ................................................. 12
Figure 8. Percent of counties with at least one prison by state........................................... 14
Figure 9. Number of counties with at least one prison, 10-state total ................................. 15
Figure 10. Percent of prisons by type of county, 10-state total ........................................... 16
Figure 11. Percent of population in non-metro counties, 10-state total .............................. 17
Figure 12. Number and percentage of prisoners by county type, 10-state total ................. 18
Figure 13. Prison expansion in California by county, 1979–2000....................................... 20
Figure 14. Prison expansion in Colorado by county, 1979–2000 ....................................... 21
Figure 15. Prison expansion in Florida by county, 1979–2000 ........................................... 22
Figure 16. Prison expansion in Georgia by county, 1979–2000 ......................................... 23
Figure 17. Prison expansion in Illinois by county, 1979–2000 ............................................ 24
Figure 18. Prison expansion in Michigan by county, 1979–2000........................................ 25
Figure 19. Prison expansion in Missouri by county, 1979–2000......................................... 26
Figure 20. Prison expansion in New York by county, 1979–2000 ...................................... 27
Figure 21. Prison expansion in Ohio by county, 1979–2000 .............................................. 28
Figure 22. Prison expansion in Texas by county, 1979–2000 ............................................ 29
Figure 23. Number of counties by percent of population in prison, 2000............................ 31
Figure 24. California prisoners by counties of sentencing and imprisonment, 2000........... 37
Figure 25. Florida prisoners by counties of sentencing and imprisonment, 2000............... 38
Figure 26. Georgia prisoners by counties of sentencing and imprisonment, 2000 ............. 39
Figure 27. Ohio prisoners by counties of sentencing and imprisonment, 2000 .................. 40
Figure 28. Texas prisoners by counties of sentencing and imprisonment, 2000 ................ 41

.

The New Landscape of Imprisonment

iii

Acknowledgments
We would like to thank the JEHT Foundation for supporting this project.
We would like to express our gratitude to Erin Neel, of the Massachusetts Institute of Technology, who
was instrumental in the design and development of the maps in this report.
Several individuals assisted with the preparation of data sets and reviewed earlier drafts and we are
grateful for their efforts. Special thanks to Meagan Funches, Vera Kachnowski, Deborah Kaye, Barbara
Parthasarathy, and Peter Tatian of the Urban Institute; Beau Kilmer of the Kennedy School of
Government at Harvard University; and Michelle Waul of the National Center for Victims of Crime.

iv

The New Landscape of Imprisonment

Executive Summary
In recent decades, growth in the number of people in U.S. prisons has been the largest in history—the
prison population increased by more than one million between 1980 and 2000. To accommodate this
growth, corrections officials have pursued a variety of strategies, including greatly expanding the network
of prisons. The number of state prison facilities increased from about 600 prisons in the mid-1970s to
over 1,000 prisons by the year 2000. Because the Census Bureau counts prisoners where they are
incarcerated in the decennial census, the locations of prisons may have significant implications for state
and federal funding allocations, as well as political representation.
Despite this tremendous growth, the prison construction boom has received relatively little attention. It is
remarkable that a public undertaking as far-reaching as the American prison expansion, which affects
millions of incarcerated individuals, influences millions more family and community members, and
consumes billions of public dollars, would receive so little empirical analysis and public scrutiny. This
report contributes to the limited knowledge base by developing an empirical understanding of the
geographic locations of prison facilities—and therefore prisoners—following this record-level expansion
over the past two decades. Prison expansion is examined from national, state, and county-level
perspectives, and in terms of the extent to which prisons were located in “metro” counties or “non-metro”
counties. This report focuses on 10 states that experienced the largest growth in the number of prisons
during the 1980s and 1990s.
Several themes emerge from the analyses presented in this report. First is the pervasiveness of prison
growth. The prison construction boom of the last two decades was not concentrated in a few states or in
certain regions of the country, but occurred in states across the country. Prison systems also expanded
within states, as new prisons were more geographically dispersed. The share of counties in the 10 study
states that were home to at least one prison increased from 13 percent of counties in 1979 to 31 percent
of counties in 2000. In addition, the number of prisons increased significantly in both metro and non-metro
counties, challenging the notion that prison expansion has primarily taken place in non-metro counties.
A second theme to emerge is that in a select number of smaller communities, prison expansion has
significantly impacted the total population. In each of the 10 study states there were several counties
where a notable share of the total population was incarcerated. Thirteen counties in the 10 study states
had 20 percent or more of the resident population imprisoned in 2000. All 10 states had least five
counties where 5 percent or more of the population was imprisoned. Not surprisingly, most of these
counties, but not all, were non-metro counties. Analyses presented in this report show that the share of
prisoners who resides in non-metro counties is greater than the share of the general population who
resides in non-metro counties, and that this has been the case for at least the last two decades.
A third theme of this report is the mismatch between the places prisoners consider home and the places
prisoners serve their time. A series of maps illustrates large disparities between the sentencing counties
and the counties of imprisonment.
Issues related to prison expansion of the 1980s and 1990s are numerous and complex. We hope that this
report 1) provides a better understanding of this expansion in terms of spatial distribution, 2) challenges
some commonly held ideas about prison growth, and 3) highlights issues that deserve additional
attention. Our primary goal, however, is to use empirical analyses to ground the debate surrounding
prison expansion and to lay the foundation for future studies.

The New Landscape of Imprisonment

1

1

Introduction

Highlights: Growth in the number of prison facilities during the 1980s and 1990s was historically
high. Research on where prisons were located is lacking, and therefore, the extent to which issues
related to prison expansion can be debated is limited. This report aims to shed light on the prison
expansion phenomenon by addressing the following question: What did the prison expansion of the
1980s and 1990s look like in terms of the locations of new prisons?

The growth in the number of U.S. prisoners in recent decades is the largest in history. The
incarceration rate has seen a four-fold increase over the last 25 years. Over 1.3 million people
were in state or federal prisons in 2000, up from 218,000 in 1974 (Sourcebook of Criminal
Justice Statistics 2000, table 6.27). To accommodate this growth, corrections officials have
pursued a variety of strategies, including filling prisons beyond capacity, converting buildings to
prisons, and constructing new facilities. As a consequence, the number of “confinement
facilities” operated by departments of corrections increased. Over the last 25 years, the number
of state facilities increased from just fewer than 600 to over 1,000 in the year 2000, an increase
of about 70 percent. In other words, more than 40 percent of state prisons in operation today
opened in the last 25 years.
This report examines one fundamental dimension of prison expansion: the locations of new
prisons. The location of a prison has significant consequences for prisoners and their families.
Prisons built in communities far away from prisoners’ homes make visitation more difficult.
Prison location can also affect the distribution of political power, the allocation of government
resources, and the local economies of the communities in which new institutions are built and the
communities from which prisoners are drawn.
Increases in the number of prisoners have received considerable attention by policymakers,
researchers, and advocates. The expansion of prison facilities, however, has received very little
attention. It is remarkable that a public undertaking as far-reaching as the American prison
expansion—which involves 50 state governments and the federal government, affects millions of
incarcerated individuals, influences millions more family and community members, and
consumes billions of public dollars—would receive so little empirical analysis and public
scrutiny.
Journalistic accounts and single-state studies have recently commented on possible ramifications
of the locations of prisons in America. For example:
One New York Times article states, “As communities become more and more
familiar with the benefits that prisons bring, they are also becoming increasingly
adept at maximizing their windfall through collecting taxes and healthy public
service fees” (Kilborn 2001).
From The Washington Post: “Call it salvation through incarceration—a prisonbased development strategy that small towns all over America are pursuing, and
changing economically and culturally because of it” (Duke 2000).

2

The New Landscape of Imprisonment

According to a report by the Prison Policy Initiative, “These prisoners [from New
York City] then swell the size of the rural prison communities at the expense of the
communities from whence they came” (Wagner 2002b).
In smaller communities, and particularly those with higher than average rates of poverty and
unemployment, opening a new prison is believed to be an economically beneficial endeavor.
Indeed, local “campaigns” have played a role in determining where a prison is located. However,
the few studies on the local economic impacts of prisons to date have not found significant
positive impacts. For example, a study by the Sentencing Project challenges the notion that a
new prison brings economic benefits to smaller communities. Using 25 years of data from New
York State rural counties, the authors looked at employment rates and per capita income and
found “no significant difference or discernible pattern of economic trends” between counties that
were home to a prison and counties that were not home to a prison (King, Mauer, and Huling
2003). According to a recent study by Iowa State University, many towns that made sizeable
investments in prisons did not reap the economic gains that were predicted (Besser 2003).
Another analysis in Texas found no impacts as measured by consumer spending in nearly threefourths of the areas examined (Chuang 1998).
The economic benefits of new prisons may come from the flow of additional state and federal
dollars. In the decennial census, prisoners are counted where they are incarcerated, and many
federal and state funding streams are tied to census population counts. According to the U.S.
General Accounting Office (2003), the federal government distributes over $140 billion in grant
money to state and local governments through formula-based grants. Formula grant money is in
part based on census data and covers programs such as Medicaid, Foster Care, Adoption
Assistance, and Social Services Block Grant (U.S. General Accounting Office 2003). Within a
state, funding for community health services, road construction and repair, public housing, local
law enforcement, and public libraries are all driven by population counts from the census.
Every dollar transferred to a “prison community” is a dollar that is not given to the home
community of a prisoner, which is often among the country’s most disadvantaged urban areas.
According to one account, Cook County Illinois will lose nearly $88 million in federal benefits
over the next decade because residents were counted in the 2000 Census in their county of
incarceration rather than their county of origin (Duggan 2000). Losing funds from the
“relocation” of prisoners is also an issue for New York City, as two-thirds of state prisoners are
from the city, while 91 percent of prisoners are incarcerated in upstate counties (Wagner 2002a).
The effect of prisoner location on population counts may also influence the allocation of political
representation and, therefore, political influence (Haberman 2000). In Wisconsin, the number of
state prisoners who were housed in other states (known as interstate transfers) caused concern
because these prisoners would be counted in the decennial census in the states where they were
incarcerated. In 1999, U.S. Representative Mark Green introduced a bill (unsuccessfully) that
proposed changes to the census policy so Wisconsin prisoners held in other states would be
counted as Wisconsin residents.
In order for these issues to be adequately addressed, we first need to develop an empirical
understanding of the locations of prisons—and therefore prisoners—following a record-level
expansion over the past two decades. This report uses quantitative data analysis and mapping to
document the expansion of prison systems during the 1980s and 1990s. The report looks at
The New Landscape of Imprisonment

3

prison expansion from national, state, and county-level perspectives and analyzes the expansion
along an important dimension, namely whether the prison is located in a county classified as
“metro” or “non-metro.”
Section 1 of the report presents the primary research question, data sources, and policy context
for this analysis. Section 2 uses national-level data, as well as data from 10 states, to assess the
magnitude of prison growth. Section 3 creates a framework for these prison-related changes by
describing shifts in the general resident population. In section 4, the geographic dispersion of
prison facilities, the mix of prisons by type of county (i.e., metro or non-metro), and the mix of
prisoners by type of county are discussed. Section 5 uses maps to show prison system changes at
the county level, and section 6 compares prisoner populations with total populations at the
county level. Sentencing counties are compared with counties of incarceration in section 7.
RESEARCH QUESTION AND DATA SOURCES

In an attempt to ground debate on prison expansion in empirical analysis, this report addresses
the following question:
What did the prison expansion of the 1980s and 1990s look like in terms of
the locations of new prisons?
Focusing on this question will allow us to address other questions such as: Where were facilities
located in terms of metro versus non-metro areas? Are there communities in which the prison
population makes up a notable share of the total population? To what extent are prisoners’
“home communities” the same as the communities where they are incarcerated?
The primary data sources for this analysis are the decennial census, which is conducted by the
U.S. Census Bureau, and the Census of State and Federal Correctional Facilities, which is
conducted by the Bureau of Justice Statistics (BJS).1 (See figure 1 for more details.) Numerous
publications on demographic shifts and housing patterns are produced by the U.S. Census
Bureau after each decennial census. BJS publishes summary reports after each census of state
and federal correctional facilities. To our knowledge, these data sources have not been used to
examine the geographic dispersion of state and federal prisons and prisoners. For much of this
report, the analyses focus on a set of 10 states that experienced significant growth.

1

The software programs SPSS, Excel, ArcView GIS, and Adobe Illustrator were used to compile and analyze the
data used in this report.

4

The New Landscape of Imprisonment

Figure 1. Primary data sources
Census of State and Federal Correctional Facilities, Bureau of Justice Statistics
The Census of State and Federal Correctional Facilities is a complete enumeration of state and
federal adult correctional facilities. The census of correctional facilities was conducted by the
Census Bureau for the BJS in 1974, 1979, 1984, 1990, 1995, and 2000.2 Data from the 1979
and 2000 censuses were used in this analysis.
Decennial Census, U.S. Census Bureau
Every 10 years, the U.S. Census Bureau undertakes an enumeration of every person in the
country. County-level data from the 1980 and 2000 censuses were used in this analysis. Census
2000 is the first year in which state and federal prisoners are itemized and can be distinguished
from other types of correctional populations.
Data from Individual States
Reports, quantitative data, and online information published by the 10 study states were used
where appropriate. Annual reports, online facility profiles, and prior research reports
contributed to the development of state profiles and were often used as cross-checks with other
data sources.

For our purposes, an important characteristic of prisons is the type of area in which they are
situated. Specifically, to what degree are prisons located in more populous, metropolitan areas
compared with less populous, non-metropolitan areas? Although some data are presented at the
national and state levels, counties are our primary unit of analysis. After each decennial census,
the U.S. Census Bureau publishes a list of counties that are classified as “metropolitan”
according to a standard definition determined by the U.S. Office of Management and Budget
(figure 2). Two sets of “metro” counties were used in this analysis; counties that were classified
as metro based on the 1980 decennial census were paired with the 1979 correctional facility data,
and counties that were classified as metro based on Census 2000 data (published June 6, 2003)
were paired with 2000 correctional facility data. All counties not classified as “metro” counties
in the two years of interest (1979 and 2000) are referred to as “non-metro” counties.
Figure 2. Defining a “metro” county
Every 10 years the Census Bureau publishes an updated list of Metropolitan Statistical Areas
(MSAs), which is comprised of counties and based on the most recent decennial census data. The
lists are typically published three years after a census is conducted. For example, MSAs from the
1980 Census were published in 1983.
The Office of Management and Budget determines the standards for MSAs and its component
counties. For Census 2000, OMB defined an MSA as “a large population nucleus, together with
adjacent communities, having a high degree of social and economic integration with that core”
(U.S. Office of Management and Budget 2000).
For the purpose of this report, the set of metro counties based on 1980 Census data and published
in 1983 are applied to the 1979 data on correctional facilities. The updated list of metro counties
based on Census 2000 data, which was published on June 6, 2003, was used in this report. All
counties not classified as “metro” counties are referred to as “non-metro” counties.

2

Facility-level data are not publicly available for the 1974 count. While aggregate data for 2000 are available,
facility-level, electronic data for this year are not yet publicly available. Facility-level data for the ten states of
interest in 2000 are based on previous counts of correctional facilities and state-specific research by Urban Institute
researchers.
The New Landscape of Imprisonment

5

TYPES OF CORRECTIONAL FACILITIES

Because we are interested in the locations where prisoners are incarcerated, our primary interest
is in confinement facilities and, therefore, not all types of correctional facilities are appropriate
for this analysis.3 Only facilities that meet the following criteria are included: 1) facilities that are
operated by a state government, the federal government, or by a private company that is a
contractor for a state or federal government; 2) facilities that house adult offenders sentenced to
one year or more; and 3) facilities that are classified by BJS as “confinement,” meaning less than
50 percent of the prisoners are “regularly permitted to depart unaccompanied.” The third
criterion generally excludes facilities that allow residents to spend significant amounts of time
outside of the facility, such as halfway houses and work release programs.4
Because standard definitions for correctional housing do not exist, each correctional system has
its own system and vocabulary for classifying facilities. For example, Missouri’s system includes
correctional centers, diagnostic centers, treatment centers, and community release centers. Local
jails, juvenile facilities, community corrections centers, and work release centers are some
examples of types of facilities that have been excluded. Throughout this paper, we use the terms
“prisons” and “confinement facilities” to refer to facilities that meet the above-listed criteria. The
types of facilities in the 10 study states (see below) and in the federal system and that are
included in this analysis are listed in appendix A.
COUNTING PRISONERS IN THE DECENNIAL CENSUS

The U.S. Census Bureau counts prisoners as residents of the county where they are incarcerated,
not as residents of the county where they lived prior to incarceration. From the first U.S.
decennial census in 1790 through Census 2000, the guiding principle for determining the
location in which people are enumerated has been to count individuals in their place of “usual
residence.” Usual residence is defined as the place where a person lives and sleeps most of the
time, and it can differ from a person’s legal or voting residence (U.S. Census Bureau 2000).
When this policy is applied to state and federal prisoners, they are counted where they are
incarcerated. In Census 2000, 1.3 million individuals were counted in the locations of their
confinement. As the prison population increased dramatically over the last 25 years, this Census
policy has applied to larger correctional populations. As shown in figure 3, the number of
prisoners increased from approximately 316,000 to over 1.3 million during the 1980s and
1990s—an increase of 318 percent. Stated differently, the policy of “usual residence” applied to
approximately 1 million more state and federal prisoners in 2000 than it did in 1980.
On an interstate level, this growth, in and of itself, does not raise concerns in terms of census
enumeration. Very few prisoners are transferred to other states.5 Yet, when census-based

3

The measure of incarceration used here does not include individuals in local jails, military correctional facilities,
and halfway houses. The total number of individuals that are truly “incarcerated” would be higher than the figures
used in this report.
4
Only 3 percent of state and federal prisoners are housed in nonconfinement facilities according to BJS.
5
According to BJS, in the ten states of interest less than one-half of 1 percent of state prisoners are held out of state.
The national average for prisoners held out of state is 0.9 percent (U.S. Department of Justice 2001, table 4).
6

The New Landscape of Imprisonment

resources are distributed within a state, the movements of large numbers of prisoners between
counties may have significant consequences. Specifically, if prison populations rose at the same
time that the location of prisoners within a state changed, then intrastate distributions of
financial resources may be affected.

Figure 3. Growth of state and federal prison populations,
1980, 1990, 2000
1980

1990

2000

1980 – 2000

Prison population

315,974

739,980

1,321,137

+ 1,005,163

Percent change
from last decade

—

+134%

+79%

+318%

Source: Sourcebook of Criminal Justice Statistics 2000.

The New Landscape of Imprisonment

7

2

Overall Growth in the Number of Prisons

Highlights: This analysis focuses on the 10 states that experienced the largest growth in the number
of prisons. These states account for 63 percent of the total increase in the number of prisons since
1979. The number of prisons in each of the 10 states more than doubled between 1979 and 2000.

Over the past two decades, the number of prisons increased in all jurisdictions; fifty states are
operating more prison facilities today than they were 20 years ago, even after controlling for
general population growth. The rise in the number of prisons has been extraordinary, as shown in
figure 4. During the last quarter of the 20th century, state prison systems grew from 592 prisons
to 1,023 prisons—an increase of 73 percent.

Figure 4. Number of state prisons, 1923–2000
# of state facilities
1,023
1,000

800

592

600

400

200

127
61

1923

1950

1974

2000

Sources: U.S. Department of Justice 1974; U.S. Department of Justice 1986; U.S. Department of Justice
2003.
Note: Because data for the federal system were not available for this time period, this figure only includes
state facilities.

In this report we focus on prison expansion for the 21-year period between 1979 and 2000, two
years in which BJS surveyed correctional facilities. In order to conduct more in-depth analyses at
the state and county levels, we focus on a set of 10 states. The 10 states were included because
they are home to the largest increases in the number of adult confinement facilities.6 Figure 5
shows the share of the national-level prison growth accounted for by these top 10 states. Sixtythree percent of the total growth in the number of prisons occurred in the 10 selected states
(referred to in this report as “the top 10 states”).

6

Another plausible measure is the percentage increase in the number of facilities. This yields misleading results, as
smaller states that grew from, for example, one prison to four prisons would be ranked the highest.
8

The New Landscape of Imprisonment

Figure 5. Percent of growth in number of prisons
nationwide, 1979–2000

Top ten states
63%

40 states
37%

Sources: Census of Correctional Facilities, 1979 data from the Inter-University
Consortium for Political and Social Research; U.S. Department of Justice,
2001.

Figure 6 lists the top 10 states ranked from the highest growth to the lowest growth. They are
Texas, Florida, California, New York, Michigan, Georgia, Illinois, Ohio, Colorado, and
Missouri. The magnitude of prison growth in these 10 states is remarkable. Between 1979 and
2000, the number of additional prisons ranged from 19 prisons in Missouri to 120 prisons in
Texas. The growth in Texas equates to an extraordinary average annual increase of 5.7 additional
prisons per year over the 21-year period. As a group, the 10 states were operating more than
three times as many prisons in 2000 as in 1979—increasing from 195 facilities to 604 facilities.
Figure 6 shows the relative growth in each state in addition to the absolute growth. In all 10
states, the number of prisons increased by more than 100 percent over the two decades. States
with the lowest relative growth are Florida, which grew by 115 percent, and New York, which
grew by 117 percent. Texas is again the clear leader growing by 706 percent over the 21-year
period. Indeed, Texas is in a league of its own, as it added the most prisons (120), currently has
the largest number of prisons in operation (137), and experienced the largest percentage increase
(706 percent).

The New Landscape of Imprisonment

9

Figure 6. States with highest growth in number of prisons
Number of prisons in 2000 and percent change in number of prisons between 1979 and 2000
TX

137 prisons in 2000

FL

84

CA

83

NY

+ 177%

65

MI

+ 117%
+ 140%

60

GA

42

IL

+ 133%
+ 233%

40

OH

+ 250%

35

CO

32

MO

+ 357%

26
-

+ 706% between
1979 and 2000

+ 115%

+ 271%
20

40

60

80

100

120

140

Number of prisons

Sources: Bureau of Justice Statistics 1997; U.S. Department of Justice 2001; and state-specific research by Urban
Institute staff.

10

The New Landscape of Imprisonment

3

Changes in the General Population

Highlights: The share of counties in the top 10 states classified as non-metro decreased from 72 to
41 percent between 1979 and 2000. The share of residents in the top 10 states living in non-metro
counties decreased from 16 to 5 percent during this same period.

Before examining the extent to which prisons were sited in non-metro versus metro counties, it is
useful to understand changes in the nation’s general population over the same period of time. As
previously described, the Census Bureau periodically updates its list of counties defined as
“metro” counties, reflecting decennial census data. As the overall U.S. population rose over the
past 20 years, the number of non-metro counties decreased. In 1980, more than three-fourths of
U.S. counties (76 percent) were classified as non-metro; this represents 2,400 of the nation’s
3,141 counties. In 2000, by contrast, the share of non-metro counties decreased to about twothirds of all counties (65 percent or 2,052 counties). Among the top 10 states analyzed in this
report, the decline was even more pronounced. In 1979, 72 percent of counties were classified as
non-metro, and that figure declined to 41 percent in 2000.
The shares of counties that are non-metro for the 50 states and for the top 10 states are shown in
figure 7. This chart illustrates two important points. First, it reflects the decline in non-metro
counties. Between 1979 and 2000 the share of non-metro counties declined for both the 50-state
average (76 to 65 percent) and the 10-state average (72 to 41 percent). Second, the figure
suggests that the 10 states have experienced a greater level of “urbanization,” as the decline in
non-metro counties was higher than for the entire country (a decrease of 31 percentage points for
the 10 states compared with a decrease of 11 percentage points for the 50 states). These trends
are important when we consider the degree to which prison facilities are located in non-metro
counties versus metro counties. If no prisons had been added to correctional systems between
1979 and 2000, the number and share of prisons in non-metro counties still would have declined,
simply as a result of fewer counties being classified as non-metro.

The New Landscape of Imprisonment

11

Figure 7. Percent of counties that are non-metro counties
80%

76%
72%

70%

65%

60%

50%
41%
40%

30%

20%

10%

0%
1979

2000
U.S. Average

Ten-State Average

Sources: U.S. Office of Management and Budget 2003; U.S. Census Bureau 1983a,
1983b.

Just as the mix of metro and non-metro counties changed over the past two decades, so did the
share of residents living in metro and non-metro counties. In the top 10 states examined in this
report, the share of the resident population who lived in non-metro counties decreased. In 1979,
16 percent of residents in the top 10 states lived in non-metro counties; by 2000, only 5 percent
of residents lived in non-metro counties. (Appendix B provides the share of residents and the
share of prisoners who lived in non-metro counties by state in 1979 and 2000.) Translating
percents into absolutes, the number of people living in non-metro counties in the top 10 states
declined from just under 17 million people in 1979 to approximately 7 million people in 2000.
The top 10 states varied in terms of the mix of metro versus non-metro counties. The share of
counties that were non-metro ranged from a low of 18 percent in New York State to a high of 61
percent in Colorado in 2000. In terms of the share of residents, several states had less than 3
percent of residents in non-metro counties: California (2.8 percent), Florida (2.3 percent), and
New York (2.3 percent). Among the top 10 states, the highest share of residents living in nonmetro counties in 2000 was Missouri at 14 percent.

12

The New Landscape of Imprisonment

4

Prison Growth by State

Highlights: Prison facilities spread out geographically during the 1980s and 1990s. By the year
2000, 31 percent of all counties in the top 10 states were home to at least one prison. Significant
numbers of prisoners were added to both non-metro and metro counties. A prison was more likely
to be situated in a metro county in 2000 than in 1979—50 percent of all prisons were located in
metro counties in 1979 compared with 74 percent in 2000. Prisoners disproportionately resided (i.e.,
were incarcerated) in non-metro counties for at least the last two decades; the share of prisoners
who resided in non-metro counties was greater than the share of the general population who
resided in non-metro counties.
PRISON DISTRIBUTION WITHIN A STATE

Now that we have identified states with the largest growth in the number of prisons and
understand metro/non-metro county classification, we will look at prisons within each of these
states. To what extent are prisons spread out across a state?
One measure of geographic distribution of prisons is the percent of counties that is home to at
least one prison. Because all of the top 10 states have experienced prison growth over the last
two decades, we would expect more counties to have prisons today than 20 years ago. In 2000,
almost one-third of all counties (31 percent) across the 10 states had a state or federal prison
within its boundaries compared with 13 percent of all counties in 1979. Stated another way, in
these states 197 more counties were home to prisons in 2000 than in 1979.
Figure 8 documents the spread of prisons across counties at the state level by showing that the
share of counties with a prison increased in each of the top 10 states between 1979 and 2000.
The first column in figure 8 shows the percentage of counties in 1979 with at least one prison;
the second column shows the same measure for the year 2000. For example, in California 34
percent of counties were home to one or more state or federal prison facilities in 1979, and this
grew to 59 percent by 2000. During the 1980s and 1990s, the locations of prisons spread out in
all 10 states, as the percentage of counties with prisons increased in all 10 states. The state of
Texas is, again, the leader in growth. Seven of the 254 Texas counties (3 percent) had a prison
located within its borders in 1979. By 2000, that number had increased ten-fold to a total of 70
Texas counties—or 28 percent of all counties. Figure 8 also shows that states varied in the level
of geographic dispersion of prisons. In 2000, prisons were located in only 16 percent of
Missouri’s counties and 18 percent of Georgia’s counties. At the same time, more than half of
California’s counties (59 percent) and New York’s counties (52 percent), and more than threefourths of Florida’s counties (78 percent) were home to at least one prison. In these three states,
more counties had at least one prison than counties that had no prisons.
All of the top 10 states had many more prisons in 2000 than in 1979, as absolute growth was the
criterion for selection. Figure 8 shows that, in addition, the 10 states grew in terms of the spatial
distribution of prisons. Prison facilities were opened for the first time in numerous counties
across the states. In 2000, 330 counties had prisons, up from 133 counties in 1979. This suggests
that communities have become more willing to host a new prison facility than in the past. This
The New Landscape of Imprisonment

13

has been described as the “Yes, in my backyard” phenomenon by Tracy Huling,7 and is a direct
contrast to the long-standing opposition to prisons, described as “Not in my backyard” (or
NIMBY).
Figure 8. Percent of counties with at least one prison by state
1979

2000

California

34%

59%

Colorado

8%

22%

Florida

45%

78%

Georgia

10%

18%

Illinois

7%

28%

Michigan

19%

36%

Missouri

3%

16%

New York

32%

52%

Ohio

10%

25%

Texas

3%

28%

Sources: Bureau of Justice Statistics 1997, 1998a; and state-specific research by Urban Institute staff.

7

Tracy Huling and Galloping Girls Productions, Inc. produced a documentary titled, “Yes, In My Backyard” in
1998. Ms. Huling has also written several pieces on the issue. (See Huling 1999 and 2000.)

14

The New Landscape of Imprisonment

PRISON LOCATION BY COUNTY CLASSIFICATION

Now that we know that in the last two decades prisons spread out, we will look at the extent to
which expansion occurred in non-metro counties versus metro counties. The same
measurement—counties with at least one prison—is used to compare growth in non-metro and
metro counties.
Prisons were added to large numbers of both non-metro and metro counties in the top 10 states.
Between 1979 and 2000, 197 counties in the top 10 states gained at least one prison. Figure 9
compares prison growth in non-metro and metro counties for this period.8 The number of nonmetro counties with at least one prison increased from 69 counties in 1979 to 112 counties in
2000, an increase of 43 counties. During that same period, the number of metro counties with at
least one prison grew from 64 counties to 218 counties, an increase of 154 metro counties. The
increase in non-metro counties is particularly notable because the total number of non-metro
counties decreased during this 21-year period from 757 counties to 431 counties. In other words,
non-metro counties with prisons increased despite a decrease in the total number of non-metro
counties. Similarly, some of the growth in metro counties with a prison can be attributed to the
fact that there were more metro counties in 2000 than there were in 1979. If the number of
prisons had not changed during the 21-year period, the number of metro counties with at least
one prison still would have increased as a result of county reclassification based on decennial
census data. Figure 9 also shows that the percent increase in counties with prisons was
significantly higher for metro counties (241 percent, 64 to 218 counties) than for non-metro
counties (62 percent, 69 to 112 counties).

Figure 9. Number of counties with at least one prison, 10-state total
Number of counties
250

218

200
150
112
100

69

64

50
0
1979
Non-metro counties

2000
Metro counties

Sources: U.S. Office of Management and Budget 2003; U.S. Census Bureau 1983a, 1983b; Bureau
of Justice Statistics 1997, 1998a; and state-specific research by Urban Institute staff.

8

Again, this is based on the June 6, 2003 classification of metro counties.
The New Landscape of Imprisonment

15

As another comparison of prison expansion by county type we use prison facilities rather than
counties as the unit of analysis. What percentage of all prison facilities are located in non-metro
counties versus metro counties? Prisons are not new to non-metro counties. One half of all
prisons in the top 10 states were located in non-metro counties in 1979 (figure 10). Of the 195
prisons in the top 10 states in 1979, 97 were situated in non-metro counties and 98 in metro
counties. This is notable because it challenges the conventional wisdom that, in the past, the
majority of prisons had been operating in or near metro areas and that locating prisons in nonmetro areas is a relatively new phenomenon. Figure 10 demonstrates that a significant share of
prisons have been operating in non-metro areas for several decades.
Figure 10 also shows that over the last two decades, the percentage of prisons located in metro
counties has increased. In 2000, approximately three out of four prisons (74 percent) were in
metro counties, up from 50 percent in 1979, and one out of four prisons (26 percent) were in
non-metro counties. This finding also challenges the notion that the prison-building boom took
place primarily in non-metro areas.

Figure 10. Percent of prisons by type of county, 10-state total
1979

50%

2000

50%

26%
74%

Sources: U.S. Office of Management and Budget 2003; U.S. Census Bureau 1983a, 1983b; Bureau of Justice Statistics
1997, 1998a; and state-specific research by Urban Institute staff.

PRISONER LOCATION BY COUNTY CLASSIFICATION

In addition to changes in the locations of prison facilities, changes in the locations of prisoners
are also of interest. This is especially true in light of the potential financial and political
implications of the Census policy of counting prisoners where they are incarcerated. In this
section, the degree to which prisoners are incarcerated in non-metro versus metro counties is
examined and compared with the general population.
A disproportionate share of prisoners are imprisoned in non-metro counties relative to the
general population. In the top 10 states in 2000, 23 percent of state and federal prisoners lived
16

The New Landscape of Imprisonment

(i.e., were incarcerated) in non-metro counties, while only 5 percent of the general population in
the top 10 states lived in non-metro counties (see figure 11). Put another way, 1 out of every 20
people in the general population resided in non-metro counties, while 1 out of every 5 prisoners
was incarcerated in non-metro counties. Data from 1979 indicate that this imbalance has existed
for some time. In 1979, 52 percent of the prisoner population and 16 percent of the general
population in the top 10 states lived in non-metro counties. Even before the prison construction
boom of the 1980s and 1990s, prisoners were more likely to reside in non-metro counties than
the general population.
Another observation from figure 11 is that a smaller share of prisoners in the top 10 states is
located in non-metro counties today compared with 21 years ago. More than half of prisoners (52
percent) were located in non-metro counties in 1979, and that fraction declined to less than onequarter of prisoners (23 percent) in 2000. Although a decline in non-metro residence is expected
as the number of non-metro counties decreased, the decline in prisoners and residents in nonmetro counties is not entirely attributable to the reclassification of counties.9

Figure 11. Percent of population in non-metro counties, 10-state total
60%
52%
50%
40%
30%
20%

23%
16%

10%

5%

0%
1979
General population

2000
Prisoner population

Sources: U.S. Office of Management and Budget 2003; U.S. Census Bureau 1983a, 1983b; decennial census data
1980 and 2000; Bureau of Justice Statistics 1997.

The growth in the prison population during the 1980s and 1990s was enormous, increasing by
over 1 million prisoners nationwide. This has resulted in a larger number of prisoners being
incarcerated in non-metro counties today compared with the number incarcerated in 1979,
despite the fact that there are fewer non-metro counties. The number of prisoners in non-metro
counties in the 10 states more than doubled between 1979 and 2000. As shown in figure 12, non-

9

This assumes that there have been no significant changes in the average population of a prison facility.
The New Landscape of Imprisonment

17

metro prisoners increased from approximately 72,000 to approximately 156,000 during this time
period.
The growth of prisoners in non-metro counties, however, pales in comparison to the growth of
prisoners in metro counties. Prisoners living in metro counties increased from 67,637 to 531,652
prisoners (+464,015) between 1979 and 2000. This increase of 686 percent is far greater than the
doubling of the prisoners in non-metro counties. The larger number of prisoners in metro
counties has been fueled by two factors: the increase in the number of metro counties and the
increase in the total prison population.

Figure 12. Number and percentage of prisoners by county type, 10-state total
1979

2000

1979 to 2000

# prisoners in non-metro
counties

72,256

156,302

+ 84,046

# prisoners in metro counties

67,637

531,652

+ 464,015

Percentage of prisoners in nonmetro counties

52

23

- 29 percentage points

Percentage of prisoners in
metro counties

48

77

+ 29 percentage points

Sources: U.S. Office of Management and Budget 2003; U.S. Census Bureau 1983a, 1983b; U.S. Census Bureau 2000;
Bureau of Justice Statistics 1997, 2003.

The expansion of prisons and the growth in prisoners in the top 10 states were presented in
section 4. In addition, the extent to which expansion occurred in non-metro versus metro
counties was examined. Prisons spread out geographically during the 1980s and 1990s. Many
more counties are home to a prison today than was the case 20 years ago. This expansion
affected both non-metro and metro counties in the top 10 states, as significant numbers of
prisoners were introduced in both non-metro and metro counties. Analyses in this section also
showed that, in 2000, prisoners were disproportionately housed (incarcerated) in non-metro
counties relative to the general population, and this had been the case for at least the last two
decades.

18

The New Landscape of Imprisonment

5

Mapping County-Level Changes in the Top 10 States

Highlights: Ten state-level maps showing the number of additional prisons over the 21-year period
are presented. They suggest that in some states, such as California, Florida, and New York, prison
expansion has been geographically clustered, while in other states, such as Missouri and Ohio,
prisons have opened throughout the state. The maps add further support to the notion that growth
has been larger in metro counties than in non-metro counties over the last two decades.

Thus far we have looked at prison expansion from a national perspective and in terms of totals
for the top 10 states. This section goes one step further and looks at changes within states at the
county level, which allows us to address the following questions: Do certain counties within a
state account for a disproportionate share of the prison expansion? How many prisons have been
added to each county? What are the similarities and differences across these states in terms of
prison expansion?
A series of maps is used to address these and other questions. One map for each of the top 10
states is included (figures 13 through 22) and reflects the 2000 county classifications; non-metro
counties are mapped as white, and metro counties are shaded. The circles and stars shown on the
maps represent the number of additional prisons in a county for the 1979 to 2000 period. A
smaller circle represents the addition of one or two prisons, a larger circle represents three or
four prisons, and a star represents the addition of five or more prisons to a single county. For
example, in figure 13, the larger-sized circle shown in San Diego County means that the county
had three or four prison facilities in 2000 than it did in 1979. Note that these maps have been
designed to show only the county-level changes in prisons. Therefore, if a county operated two
prisons in 1979 and two prisons in 2000, then that county would not contain circles or stars on
the map.

The New Landscape of Imprisonment

19

PRISON EXPANSION IN CALIFORNIA

California is very much a “metro state” with 36 of 58 counties being classified as metro in 2000
and 97 percent of its residents living in metro counties. Metro counties are generally located
along the coast and include the cities of San Francisco, Los Angeles, and San Diego. The less
populous parts of the state are north of the San Francisco Bay area and inland in central
California, commonly referred to as the “Valley.” As shown in figure 13, prison expansion was
widespread in California; all areas of the state were recipients of additional prison facilities
between 1979 and 2000 (e.g., north/south, coastal/inland, non-metro/metro).
The majority of new facilities are located in metro areas—of the 53 prisons added between 1979
and 2000, 42 were in 16 metro counties and 11 were in 11 non-metro counties.10 All of the
counties with the largest growth are metro counties. Seven counties are home to three or more
new prisons and all of them are metro counties.11 Los Angeles County has experienced the
largest increase in the number of prisons, increasing from zero prisons in 1979 to eight prisons in
2000.

Figure 13. Prison expansion in California by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific research by
Urban Institute staff.

10

This includes conservation camps as well as institutions. See Appendix A for more details.
They are Los Angeles County, Kern County, Riverside County, San Diego County, Fresno County, Alameda
County (which includes the City of Oakland), and Santa Barbara County.

11

20

The New Landscape of Imprisonment

PRISON EXPANSION IN COLORADO

Colorado is the smallest of the top 10 states. It has the smallest general population (about 4.3
million residents in 2000) and the smallest prisoner population (just over 16,000 prisoners).
Compared with most of the other top 10 states, Colorado is “less metro.” Only 39 percent of
Colorado’s counties (25 of 64 counties based on Census 2000 data) are classified as metro
counties compared with the 10-state average of 60 percent of counties.
Colorado experienced some of the largest relative growth in terms of prisons during the 1980s
and 1990s. In 1979, only seven state or federal prisons were operating in the state. By the year
2000, the number of prisons had grown to 32—a 357 percent increase. Comparatively, the
average growth in the number of prisons in the top 10 states was 210 percent.
Colorado differs dramatically from the other states in that its prison expansion was concentrated
in a single county: 10 of the 25 additional prisons are located in Fremont County (see figure 14).
With a resident population of approximately 46,000, prisoners who are housed in Fremont
County account for a significant share of the total population—18 percent. (Prisoners as a share
of all residents are discussed further in section 6). Fremont County aside, the remaining prison
expansion in Colorado was almost equally split between non-metro and metro counties.

Figure 14. Prison expansion in Colorado by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific research by
Urban Institute staff.

The New Landscape of Imprisonment

21

PRISON EXPANSION IN FLORIDA

Florida has the highest level of prison dispersion relative to the other nine states. Of Florida’s 67
counties, 52 (or 78 percent) had at least one prison in 2000. In other words, only one in four
counties did not have a prison situated within its boundaries in 2000. California is a distant
second in terms of geographic dispersion, with 59 percent of counties having at least one prison.
Almost all of the non-metro counties in Florida have at least one prison. In fact, only two nonmetro counties in Florida had no prisons within their boundaries.12 Prisons were introduced for
the first time to many counties during the 1980s and 1990s. Twenty-four counties in Florida
received their first prison during this period.
Figure 15 shows that non-metro and metro counties in Florida are geographically segregated.
Non-metro counties are primarily located in the northern part of the state, or the “pan handle.”
Figure 15 also shows that prisons added during the last two decades are likely to be in the
northern counties, regardless of the non-metro/metro classification. This suggests that Florida’s
prison expansion is more a function of geography than a function of population density.
Miami-Dade County is notable in a few respects. First, it experienced the largest growth in
Florida in the number of prisons, increasing by five prisons between 1979 and 2000. Second,
Miami-Dade County is the only county in the southernmost part of the state to experience a net
gain in prisons. None of the six counties surrounding Miami-Dade County have added a prison
since 1979.
Figure 15. Prison expansion in Florida by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific
research by Urban Institute staff.

12

22

They are Franklin County and Suwannee County.
The New Landscape of Imprisonment

PRISON EXPANSION IN GEORGIA

Georgia experienced significant population growth during the 1980s and 1990s. The number of
residents rose from about 5.4 million in 1979 to approximately 8.2 million in 2000. This
population growth is reflected in the classification of counties, because many of Georgia’s
counties were reclassified from non-metro to metro. In 1979, more than three-fourths of
Georgia’s counties (76 percent or 121 of 159 counties) were non-metro. Based on Census 2000
data, only 38 percent of Georgia’s counties (61 counties) are classified as non-metro. Stated
differently, 60 counties changed from non-metro to metro status between 1979 and 2000.
The number of prisons in Georgia more than doubled between 1979 and 2000, increasing from
18 to 42 prisons. Prison expansion in Georgia has occurred throughout most of the state. As
figure 16 shows, there is no discernable pattern in terms of where new prisons have been located.
The additional 24 prisons were located in both non-metro and metro counties—9 non-metro
counties and 11 metro counties gained at least one prison. Fulton County saw the largest addition
of prisons, with three prisons opening between 1979 and 2000.
Prisoners are more likely to reside in non-metro areas than the general resident population in all
of the top 10 states (see section 4.3), but the imbalance is more pronounced in Georgia. In the
year 2000, approximately 1 in 10 residents lived in non-metro counties, compare to 4 in 10
prisoners.

Figure 16. Prison expansion in Georgia by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific research by
Urban Institute staff.

The New Landscape of Imprisonment

23

PRISON EXPANSION IN ILLINOIS

Population growth in Illinois was smaller than most of the other top-10 states, increasing by 9
percent between 1979 and 2000. However, the number of prisoners increased four-fold during
this period. To accommodate the increase in prisoners, the number of prisons in Illinois rose
from 12 to 40 during the 21-year period.
The expansion of prisons in Illinois occurred in many counties across the state. No single county
in Illinois gained more than two prisons, as shown by the presence of only smaller-sized circles
in figure 17. Of the 102 counties in Illinois, 25 gained at least one prison during the 1980s and
1990s. Eighteen of the 25 counties (72 percent) that gained prisons are metro counties, compared
with two-thirds of all counties in Illinois (64 percent) being classified as metro counties.
Figure 17 shows that very few prison facilities were added to the northern part of the state. This
is worth noting because Cook County, a northern county and home to the City of Chicago, is a
significant source of prisoners in Illinois: 60 percent of prisoners come from Cook County
according to the Illinois Department of Corrections (2000).

Figure 17. Prison expansion in Illinois by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific
research by Urban Institute staff.

24

The New Landscape of Imprisonment

PRISON EXPANSION IN MICHIGAN

During the 1980s and 1990s, the number of prisons in Michigan grew from 25 to 60, and by the
year 2000, 36 percent of Michigan counties were home to at least one prison, up from 19 percent
in 1979. Prison expansion occurred in both non-metro and metro counties as shown in figure 18,
but the majority of prisons were opened in metro counties—25 of 35 additional prisons in
Michigan were in metro counties. Despite the fact that much of the prison expansion occurred in
metro counties, prisoners were still disproportionately located in non-metro counties. In 2000, 14
percent of the prisoner population was imprisoned in non-metro counties compared with 8
percent of the general population.
Figure 18 also shows that with a few exceptions, such as Chippewa County, new prisons in
Michigan were located in counties in the southern half of the state. All of the Michigan counties
that gained three or more prisons over the last two decades were metro counties. Chippewa
County and Wayne County, which includes the City of Detroit, increased by four prisons
between 1979 and 2000. Branch County, Gratiot County, and Iona County increased by three
prisons.

Figure 18. Prison expansion in Michigan by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific research by
Urban Institute staff.

The New Landscape of Imprisonment

25

PRISON EXPANSION IN MISSOURI

Missouri is a relatively small state with a total population of about 5.6 million in 2000. The
prisoner population in Missouri increased from approximately 5,000 to approximately 22,000
between 1979 and 2000. To accommodate this growth, the number of prisons in Missouri rose
from seven prisons in 1979 to 26 in 2000. In relative terms, prison expansion in Missouri was
very high—the number of prisons increased by 271 percent compared with the 10-state average
of 210 percent.
Prison expansion in Missouri has also been widespread. No single county gained more than two
prisons and 15 different counties gained at least one prison. In addition, prison expansion in
Missouri mostly occurred in metro counties. Twenty of 26 prisons in Missouri were in metro
counties in 2000.

Figure 19. Prison expansion in Missouri by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific research
by Urban Institute staff.

26

The New Landscape of Imprisonment

PRISON EXPANSION IN NEW YORK

Although the number of state and federal prisons in New York State more than doubled during
the 1980s and 1990s (from 30 to 65 facilities), its relative growth was low compared with the
other top 10 states. The number of prisons in New York State increased by 117 percent,
compared with the 10-state average of 210 percent.
New York State is similar to Florida in that the growth of prisons appears to be largely a function
of geography. A handful of northern counties accounts for a disproportionate share of the
growth. Franklin County gained four new facilities, St. Lawrence County gained three new
facilities, and Essex County and Jefferson County each gained two new facilities. The fourcounty total of 12 prisons in 2000 is up from one prison in 1979.
Despite this concentrated growth, prisons were added to almost all parts of the state. Figure 20
shows that counties in the western part of the state, in the central part of the state, and counties
outside of the New York City metropolitan area were recipients of new prison facilities between
1979 and 2000.

Figure 20. Prison expansion in New York by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific
research by Urban Institute staff.

The New Landscape of Imprisonment

27

PRISON EXPANSION IN OHIO

The number of prisons in Ohio grew significantly during the 1980s and 1990s. Prisons in Ohio
more than tripled over the last two decades, increasing from 10 in 1979 to 35 in 2000. Most of
the expansion took place in metro counties. Almost all of the state and federal prisons in Ohio
were located in metro counties—34 of 35 prisons were in metro counties in 2000. This is, in part,
due to the fact that more than three-fourths of Ohio’s counties (78 percent) are classified as
metro.
As shown in figure 21, prisons were added throughout the state of Ohio. Nineteen counties
gained at least one prison between 1979 and 2000. Allen County saw the largest increase in
prisons, with an increase of three.

Figure 21. Prison expansion in Ohio by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific
research by Urban Institute staff.

28

The New Landscape of Imprisonment

PRISON EXPANSION IN TEXAS

Texas ranks number one among the top 10 states along several of the measures discussed thus
far. It had the highest growth in prisons (120 additional prisons, up 706 percent). It had the
highest growth in counties that were home to at least one prison, increasing from 3 percent of all
counties to 28 percent. Texas is also a leader in terms of the number of counties that gained
several prisons. In 12 Texas counties, three or more facilities opened during the 1980s and 1990s
(shown as stars and larger circles in figure 22). California is a distant second with seven counties
gaining three or more prisons. Coryell County and Jefferson County each had five more prisons
in 1979 than in 2000. Anderson, Bowie, El Paso, and Liberty counties all had four additional
prisons.
Texas is similar to other states in that prison expansion occurred both in metro and non-metro
counties. Of the 120 additional facilities, 92 were in metro counties and 28 were in non-metro
counties.

Figure 22. Prison expansion in Texas by county, 1979–2000

Sources: U.S. Office of Management and Budget 2003; Bureau of Justice Statistics 1998a; state-specific
research by Urban Institute staff.

The New Landscape of Imprisonment

29

SUMMARY OF PRISON EXPANSION IN THE TOP 10 STATES

The 10 states included in this analysis were chosen because of the large numbers of prisons
introduced during the 1980s and 1990s. As stated earlier, the number of prisons in each state
more than doubled during this time period. Figures 13 to 22 use mapping to document this
growth at the county level. The maps show that in most of the states prison expansion occurred
in both non-metro and metro counties and that prison expansion was geographically dispersed.
Prisons opened across the states in numerous counties, and this was particularly true in Georgia,
Illinois, Missouri, and Ohio. A relatively small number of counties gained several prisons during
the 1980s and 1990s. Only 33 counties out of a total of 1,052 counties across the 10 states grew
by three or more prisons, and 12 of these 33 counties were in Texas.
There are some notable exceptions to this geographic dispersion. Colorado saw the greatest
concentration, as Fremont County gained 10 prisons between 1979 and 2000. The next largest
growth in Colorado was in counties that gained only two prisons. Prisons in Florida were located
in the northern part of the state and relatively few prisons were opened in central and southern
Florida. In New York, although prisons were located across the state, a disproportionate number
of new prisons were in the northern counties that border Canada.

30

The New Landscape of Imprisonment

6

Effects on Small Communities

Highlights: All 10 states had counties where a notable share of the resident population was in
prison. They all had at least five counties where 5 percent or more of the resident population was
incarcerated. There are 13 counties where 20 percent or more of the population was imprisoned.
However, five of the 10 states had no counties with 20 percent or more of incarcerated residents.

The previous section showed that prison expansion during the 1980s and 1990s occurred in both
metro and non-metro counties. In terms of population, the effects of additional prisons would be
negligible in most metro counties. In other words, the addition of prisoners to a county’s
population is unlikely to significantly affect the total number of residents when a county is
already populous. However, significant numbers of state and federal prisoners added to smaller,
non-metro counties may have considerable impacts on the total population. In this section we
address the question: are there counties in which the share of residents who are in prison is not
negligible?
This section looks at counties in the top 10 states in terms of the percentage of the population in
state or federal prisons using data from 2000. (Recall that appendix A includes the types of
facilities that are included by state.) Figure 23 summarizes this analysis and presents the number
of counties in each of the top 10 states that had different levels of residents incarcerated in 2000.
The levels presented in figure 23 are 30 percent or more of the county’s total population
incarcerated, 20 percent or more, 10 percent or more, 5 percent or more, and 1 percent or more.
For example, seven California counties had 5 percent or more of the population incarcerated, and
eight Georgia counties had 10 percent or more in prison in 2000.

Figure 23. Number of counties by percent of population in prison, 2000
30% or
more in
prison

20% or
more

10% or
more

5% or
more

1% or
more

Total # of
counties

CA

--

1

3

7

14

58

CO

--

--

3

6

7

64

FL

1

1

8

18

28

67

GA

--

--

8

16

27

159

IL

--

2

2

12

26

102

MI

--

--

2

6

17

83

MO

--

1

2

6

13

115

NY

--

--

1

5

18

62

OH

--

--

2

5

15

88

TX

1

8

16

33

49

254

Ten-state total

2

13

47

114

214

1,052

Source: U.S. Census 2000.

The New Landscape of Imprisonment

31

Figure 23 shows that prison systems in the top 10 states are, for the most part, geographically
spread out. Approximately one in every five counties in the top 10 states had 1 percent or more
of the population imprisoned in state or federal prisons in 2000 (214 of 1,052 counties). Figure
23 also reveals several other interesting points. First, counties with a notable share of residents in
state or federal prisons were found in all 10 states. All of the top 10 states had at least one county
in which 10 percent or more of the county’s population were imprisoned in state or federal
prisons in the year 2000. Across the 10 states, 47 of 1,052 counties met the 10 percent threshold.
In addition, all 10 states had at least five counties where 5 percent or more of the population
were in prison. Second, counties with higher levels of imprisoned residents were
disproportionately located in Texas. Thirteen counties across the 10 states had 20 percent or
more of the residents in prison. More than half of these counties (8 of 13) were located in Texas.
Third, only two of the 1,052 counties in the top 10 states surpassed the 30 percent threshold. The
county with the largest share incarcerated was Concho County in Texas. Concho County, whose
population was just under 4,000 in the year 2000, had 33 percent of its total population living in
prison. The second highest percentage was Union County in Florida, where 30 percent of its
13,400 residents were imprisoned.
We noted earlier that Florida had the highest percentage of counties with a prison. Seventy-eight
percent of Florida’s counties were home to at least one prison. Along those same lines, Florida
had the highest share of counties with at least 1 percent of the population in prison in 2000. More
than 40 percent of Florida’s counties passed the 1 percent threshold of incarcerated residents. In
other words, in 28 of Florida’s 67 counties, at least 1 percent of residents were in state or federal
prison. Furthermore, more than one-quarter of Florida’s counties (27 percent) have at least 5
percent of the population in prison.
Not surprisingly, most of the counties with a notable share of residents imprisoned are non-metro
counties. Nine of the 13 counties that exceed the 20 percent threshold are classified as non-metro
counties. However, many of the counties that exceed the 10 percent threshold are in fact metro
counties. Of the 47 counties meeting the 10 percent or more threshold, 14 of them (30 percent)
are metro counties. The most populous county to pass the 10 percent threshold is Kings County,
California, where 13 percent of its approximately 130,000 residents were in prison in 2000.
The analyses presented thus far show that prison expansion in the top 10 states was
geographically widespread and affected large numbers of non-metro and metro counties. This
section shows that in light of this dispersion, expanding prison systems resulted in numerous
counties having substantial shares of the total resident population incarcerated.

32

The New Landscape of Imprisonment

7

Mapping County of Sentencing versus County of Incarceration

Highlights: A set of five maps compares sentencing counties with the counties of imprisonment in
five states. Generally, there is a notable contrast in the counties from which prisoners are sentenced
and counties in which prisoners are housed. Georgia, Ohio, and Texas have little overlap between
the counties of incarceration and sentencing. In California and Florida, overlap between the two
groups is moderate.

Another issue related to prison expansion of the 1980s and 1990s is the disparity between where
prisoners come from (“home counties”) and where prisoners serve their sentences (“prison
counties”). Many believe that the prison construction boom of the last 20 years happened in areas
that were located far away from prisoners’ homes. This has been an area of concern because
greater distances between a prisoner’s home and where he or she is incarcerated can negatively
impact a prisoner and his or her family members. Being incarcerated far away from home makes
it more challenging to maintain familial relationships and parent/child relationships in particular.
In addition, challenges related to reintegrating into the community increase when a prisoner is
housed far away from home. For example, steps that may facilitate prisoner reentry, such as
finding a job and a place to live, are more difficult when a prisoner is imprisoned a long distance
from the place to which he or she will return after release.
In this section, we compare where prisoners are from with where they are imprisoned in 5 of the
top 10 states: California, Florida, Georgia, Ohio, and Texas.13 These five states were selected
based on availability of data on the sentencing counties of state prisoners. Sentencing county
data serve as a proxy for mapping prisoners’ home counties. The data do not allow us to compare
home counties and prison counties at an individual level; rather, they allow us to make countylevel aggregate comparisons of where prisoners come from with where they are incarcerated.
The following series of maps presents a comparison of home counties and prison counties.
Figures 24 through 28 include two maps for each of the five states. The first map for each state
shows the state prison population by sentencing county, and the second map shows the state
prison population by the county of incarceration. For example, figure 24 shows that in
California, 6.6 percent of state prisoners were sentenced and 7.5 percent of state prisoners were
imprisoned in San Bernadino County in 2000. We are interested in the extent to which the
population distribution in the first map matches the population distribution in the second map.
CALIFORNIA

In 2000, California’s prisoners were likely to come from one of five southern counties, but were
much less likely to be incarcerated in one of these five counties (figure 24). San Bernadino,
Riverside, San Diego, Orange, and Los Angeles counties accounted for the majority of
sentencing (59.3 percent). At the same time, only 23.2 percent of prisoners were imprisoned in
these five counties. Counties of imprisonment were more spread out across the state relative to

13

The analyses in this section include only state prisoners, and federal prisoners are excluded.
The New Landscape of Imprisonment

33

sentencing counties. For example, Lassen County in the north accounted for 5.7 percent of
incarceration and 0.1 percent of sentencing, Monterey County along the coast accounted for 7.6
percent of incarceration and 0.9 percent of sentencing, and Madera County in the central valley
accounted for 4.6 percent of incarceration and 0.5 percent of sentencing.
Looking at individual counties reveals that several California counties were very much out of
balance in terms of sentencing locations versus imprisonment locations. Most notably, Los
Angeles County accounted for a remarkable 33.8 percent of sentencing in California in 2000 and
only 3.2 percent of imprisonment. San Diego was the second largest source of prisoners,
representing 8.2 percent of sentencing, while only 3.2 percent of prisoners were serving their
time in San Diego County. Another example is Orange County, which accounted for 5.2 percent
of sentencing and 0.0 percent of imprisonment.
Despite these mismatches, figure 24 shows that, in some select counties, notable shares of
prisoners are both sentenced and incarcerated. That is, prisoners’ home counties overlapped with
prisoners’ prison counties. For example, Riverside County accounted for 5.5 percent of
sentencing and 9.4 percent of incarceration and San Bernadino County accounted for 6.6 percent
and 7.5 percent of sentencing and incarceration, respectively. However, the discrepancy between
sentenced and incarcerated prisoners might be larger than these numbers suggest, as the prisoners
sentenced in a given county may not be incarcerated in that county. For example, the 6.6 percent
of prisoners sentenced in San Bernadino are not necessarily included in the 7.5 percent of
prisoners incarcerated there, and many may have been sent to another county.
FLORIDA

In Florida, the disparity between where prisoners come from and where they serve their sentence
was significant, as illustrated in figure 25. Florida’s prisoners were more likely to come from the
southern part of the state than they were to be imprisoned there. A significant share of prisoners
came from three southern counties (28.5 percent from Miami-Dade County, Broward County,
and Palm Beach County.) At the same time, a smaller share of prisoners were imprisoned in
these three counties (11.1 percent).
Figure 25 shows that Florida’s prisoners were more likely to be incarcerated in counties located
in the northern part of the state. Jackson County, Union County, and Bradford County combined
accounted for only 0.7 percent of sentencing in 2000, but 15.5 percent of Florida’s prisoners
were imprisoned in these three counties. This finding should not be surprising as figure 15
previously documented that prison expansion in Florida occurred for the most part in the
northern part of the state.
In 2000, several of Florida’s counties were home counties and prison counties for large numbers
of prisoners. Miami-Dade County was the highest ranked county both in terms of share of
sentencing (12.7 percent) and share of incarceration (6.6 percent). Other examples of counties
with overlap across the two maps include Polk County (4.0 percent of sentencing and 3.9 percent
of incarceration), Orange County (6.1 and 3.2 percent), and Duval County (6.6 and 2.9 percent).

34

The New Landscape of Imprisonment

GEORGIA

In Georgia, the locations of incarceration were very different from prisoners’ home counties.
There was almost no overlap between the highest-ranked home counties and prison counties in
2000 as shown in figure 26. No county was both a significant source of prisoners and a
significant location for incarceration. The six counties in Georgia that accounted for the most
sentencing were Fulton (8.4 percent), DeKalb (6.6 percent), Chatham (5.1 percent), Cobb (4.7
percent), Richmond (4.1 percent), and Clayton (4.0 percent). (The city of Atlanta is split between
Fulton and DeKalb counties.) These six counties accounted for one-third of all sentencing (32.9
percent) of state prisoners in 2000, yet only 6.4 percent of prisoners were incarcerated there.
The top counties for incarceration were mutually exclusive from the top sentencing counties in
2000. Counties with 4 percent or more of all prisoners included Baldwin (13.0 percent), Tattnall
(9.9 percent), Butts (4.8 percent), and Mitchell (4.0 percent), which totalled 31.8 percent of all
state prisoners. At the same time, these four counties represented 2.1 percent of sentencing
counties.
Figure 26 illustrates that the home counties of prisoners are concentrated in and around the city
of Atlanta. Prison counties were more geographically dispersed relative to home counties.
Prisoners were more likely to be imprisoned in many counties in the central and southern parts of
the state.
OHIO

The majority of prisoners in Ohio came from a handful of counties (figure 27). Cuyahoga County
(which includes the city of Cleveland), Franklin County (Columbus), Hamilton County
(Cincinnati), Montgomery County (Dayton), Summit County (Akron), and Lucas County
(Toledo) accounted for 58.7 percent of the sentencing of state prisoners in 2000, yet only 4.6
percent of prisoners were incarcerated there.
In terms of location of incarceration, 10 counties each held 4 percent or more of state prisoners:
Richland, Ross, Marion, Madison, Lorain, Pickaway, Warren, Allen, Belmont, and Noble. More
than three-fourths of all of Ohio’s prisoners (78.3 percent) were imprisoned in these 10 counties
in 2000, while only 8.3 percent of prisoners would consider these counties to be home counties.
In Ohio, both the sources of prisoners and the locations of prisoners were geographically spread
out across the state. Prisoners were imprisoned throughout the state, and the southeast was the
only part of the state that was not a major source of prisoners. Despite this spatial distribution,
little overlap between sources of prisoners and locations of prisoners existed. None of the
counties listed above was included in both maps as one of the top-ranked counties.
TEXAS

In 2000, approximately half of Texas prisoners (50.7 percent) were sentenced in four counties.
These counties are, not surprisingly, the counties of some of the state’s largest cities, including
The New Landscape of Imprisonment

35

Harris County (which includes the city of Houston), Dallas County (city of Dallas), Tarrant
County (Fort Worth), and Bexar County (San Antonio). Prisoners were housed in counties other
than their home counties, as only 3.7 percent of Texas prisoners were incarcerated in these four
counties.
Prisoners were more widely dispersed across the state relative to where they were sentenced
(figure 28). The highest shares of Texas prisoners were located in six counties: Walker,
Anderson, Brazoria, Coryell, Bee, and Jefferson counties. These six counties accounted for 40.8
percent of all prisoners in 2000.
SUMMARY OF SENTENCING LOCATIONS VERSUS INCARCERATION LOCATIONS

The maps presented in figures 24 through 28 reveal a few interesting points about prisoners in
the five states. Most importantly, where state prisoners come from and where state prisoners are
incarcerated are generally different locations. In three of the five states that were examined, not
one county was among the highest ranked for sentencing and the highest ranked for
incarceration. In Florida, only one county accounted for a significant source of prisoners as well
as a significant share of imprisonment—Miami/Dade County. Using this measure, California had
the most overlap between home counties and prison counties, as three counties fell into the top
group in both maps. Los Angeles County, Riverside County, and San Bernadino County had 4
percent or more of sentencing and imprisonment in 2000. In short, only 4 out of a total of 626
counties in these five states appeared in the top groups of home counties and prison counties.
Another observation from this mapping exercise is that the locations of home counties were more
concentrated compared with the locations of where prisoners were housed. The largest sources of
prisoners were very much aligned with the major cities in a state, which is what we would
expect. At the same time, the counties of imprisonment were more widely dispersed across the
states. This may also be expected in light of the prior discussions on prison expansion in terms of
geographic dispersion.

36

The New Landscape of Imprisonment

Figure 24. California prisoners by counties of sentencing and imprisonment, 2000

Sources: U.S. Department of Commerce 2000; California Department of Corrections 2001.

Figure 25. Florida prisoners by counties of sentencing and imprisonment, 2000

Sources: U.S. Department of Commerce 2000; Florida Department of Corrections 2000.

Figure 26. Georgia prisoners by counties of sentencing and imprisonment, 2000

Sources: U.S. Department of Commerce 2000; Georgia Department of Corrections 2001.

Figure 27. Ohio prisoners by counties of sentencing and imprisonment, 2000

Sources: U.S. Department of Commerce 2000; Ohio Department of Rehabilitation and Corrections 2001.

Figure 28. Texas prisoners by counties of sentencing and imprisonment, 2000

Sources: U.S. Department of Commerce 2000; Texas Department of Criminal Justice 2001.

8

Conclusion

During the 1980s and 1990s, the growth in the number of prisons across the country was
remarkable. In fact, it was higher than at any other time in history. However, because of limited
research and analysis on the nationwide prison expansion, we know little about what the
expansion looks like and about its subsequent impacts, both intended and unintended. We hope
that the analyses presented in this report will help ground the debate about prison expansion with
empirical data and provide the basis for the next generation of inquiry into the locations of
prisons.
A number of interesting questions related to prison expansion remain. For example, how much
money has been transferred out of communities that have large numbers of residents incarcerated
in other places? What are the most important factors when deciding the location of a new prison?
Has political power shifted as a result of the construction of new prisons? What are the economic
impacts, both positive and negative, on smaller communities that have gained a prison? For these
questions to be answered, it is important to understand the contours of prisons expansion.
Building a foundation of knowledge on where and to what extent prisons have been located is a
necessary first step.
When we look at all of the report findings together, several themes emerge. The first theme is the
pervasiveness of prison growth. The prison construction boom was not concentrated in a few,
key states or in certain regions of the country. Prison systems expanded significantly in states
across the country. Prison systems also expanded within states. The share of counties in the top
10 states that were home to at least one prison increased from 13 to 31 percent between 1979 and
2000. State level maps (figures 13 to 22) illustrate that new prisons were geographically
dispersed throughout the states. New prisons were generally not spatially concentrated, as few
counties gained three or more prisons. Finally, prisons expanded into different types of counties;
prisons increased significantly in both non-metro counties and metro counties.
The pervasiveness of the prison expansion challenges some commonly held beliefs. For example,
prison expansion has not been a primarily non-metro phenomenon. In fact, metro counties
experienced the largest increases in the numbers of prisons and prisoners. Furthermore, our
analysis suggests that the removal of prisoners from metro counties to imprisonment in nonmetro counties was not systemic. In some states, prison expansion was in fact accompanied by
significant movement of prisoners from metro to non-metro counties. In other states, however,
prisons were located primarily in metro counties and relatively few prisoners were placed in nonmetro counties.
A second theme emerged from this report: prison expansion has significantly impacted a select
number of communities, most of which are non-metro. There are numerous counties in which a
substantial share of the population—as much as 30 percent—is in prison. The analyses presented
here should prove useful to policymakers by starting to identify areas that may have experienced
significant impacts as a result of a new prison. Some of the states examined here have relatively
few counties where the share of incarcerated residents is cause for concern. Yet other top-10
states have numerous counties in which an alarmingly high share of residents is imprisoned.

42 The New Landscape of Imprisonment

A third theme of this report is the mismatch between the counties where prisoners come from
and the counties where prisoners serve their sentences. The maps in section 7 show large
differences between sentencing counties and counties of imprisonment. Prisoners serving time in
places far from their homes could have a variety of consequences, but their examination is
outside the scope of this report. For example, are family relationships weakened if prison visits
are infrequent? What are the fiscal consequences for communities from which significant
numbers of prisoners are removed? Why are prisons located in counties that are different from
prisoners’ home counties?
This last question is an important one, and it is also a difficult one. Another way to phrase it is to
ask: what factors determine the location of a new prison? Population density is believed by some
to be an important factor. We’ve shown that significant numbers of prisons were opened in both
non-metro and metro areas. So, although density may be a factor, it may not be one of the most
important, or it may be a proxy for something else. Many factors probably contribute to the
determination of a prison’s location, such as influence of local politicians, natural and human
resources, community acceptance, and the price of land, among others. This report does not
attempt to analyze the driving forces behind siting prisons in this report. What is presented are
data supporting the notion that there are consistent differences between prisoners’ homes and
where they are incarcerated. Spatial mismatch between prisoners and their homes not only
impacts the communities that host prisons, but it also impacts family members and friends of
prisoners. This is an important issue whose consequences also warrant further examination.
Clearly, issues surrounding the prison expansion of the 1980s and 1990s are numerous and
complex. Many of the findings presented in this report suggest that deciding the location of a
prison and the subsequent impacts from that decision are not clear-cut. We hope that after
reading this report the reader will have a good understanding of the geographic distribution of
the network of prisons. We also hope that it prompts re-examination of some commonly held
ideas about prison growth and prompts some first examinations of some of the highlighted
issues. In closing, it is our hope that the analyses presented in this report will improve future
debates and studies on prison expansion by providing an empirically based foundation of
information.

The New Landscape of Imprisonment

43

Bibliography
Beale, Calvin. 1993. “Prisons, Population, and Jobs in Non-metro America,” Rural Development
Perspectives. 8(3).
———. 1996. “Rural Prisons: An Update.” Rural Development Perspectives. 11(2).
———. 1997. “Forces Behind the Non-metropolitan Population Growth Rebound of the 1990s.”
From the proceeding of the 31st Annual Pacific Northwest Regional Economic Conference,
Spokane, WA, April 24-27.
———. 2001. “Population Change in Rural and Small Town America: 1970-2000.” Rural
Voices. 6(3): 2-5.
Besser, Terry. 2003. “The Development of Last Resort.” Paper presented at the Rural
Sociological Society Meeting, August.
California Department of Corrections. 2001. “California Prisoners and Parolees, 2001.”
http://www.cdc.state.ca.us. [Accessed March 19, 2003.]
———. 2003. “Facts and Figures, 1st Quarter 2003.” http://www.cdc.state.ca.us. [Accessed
March 19, 2003.]
Chuang, Su Chong. 1998. “The Distribution of Texas State Prisons: Economic Impact Analysis
of State Prison Sitting on Local Communities.” Ph.D. diss., University of Texas, Arlington.
Colorado Department of Corrections. 2001. “Fiscal Year 2000 Statistical Report.”
http://www.doc.state.co.us. [Accessed March 19, 2003.]
Duggan, Molly. 2000. “Census Dollars Bring Bounty to Prison Towns.” The Chicago Reporter.
July/August.
Duke, Lynne. 2000. “Building a Boom Behind Bars: Prisons Revive Small Towns, but Costs Are
Emerging.” The Washington Post, September 8.
Florida Department of Corrections. 2000. “Florida Department of Corrections 1999 – 2000
Annual Report.” http://www.dc.state.fl.us. [Accessed February 24, 2003; April 10, 2003.]
———. 2003. Electronic mail communication with Public Information Office, April 7, 2003.
———. 2003. “Statistics in Brief, January 2003.” http://www.dc.state.fl.us. [Accessed April 10,
2003.]
Florida Department of Corrections, Bureau of Research and Data Analysis. 2001. “Trends in
Prison Populations by Facility.” Tallahassee: Florida Department of Corrections.
Georgia Department of Corrections. 2001. “Georgia Department of Corrections Annual Report,
2000.” Atlanta, GA: Georgia Department of Corrections.
Georgia Department of Corrections, Office of Planning and Analysis. 2003. Electronic mail
communication with Public Information Office, March 21.
Haberman, Clyde. 2000. “Census Time: Will the Felon Please Rise?” The New York Times, April
4, section B, page 1.

44 The New Landscape of Imprisonment

Huling, Tracy. 1999. “Prisons as a Growth Industry in Rural America: An Exploratory
Discussion of the Effects on Young African-American Men in the Inner Cities.” Paper
presented for the U.S. Commission on Civil Rights, Washington, D.C., April 15.
———. 2000. “Prisoners of the Census.” MojoWire, May 10.
Huling, Tracy, and M. Mauer. 2000. “Locked up in the Census Count.” The Chicago Tribune.
March 29.
Illinois Department of Corrections. 2000. “IDOC Human Services Plan: Fiscal Years 1998 –
2000.” Chicago, IL: Illinois Department of Corrections.
———. 2003. Electronic mail communication with IDOC Communications Office, March 28.
———. 2003. http://www.idoc.state.il.us. [Accessed March 2003; April 2003; July 2003.]
Kilborn, Peter. 2001. “Rural Towns Turn to Prisons to Reignite Their Economies.” The New
York Times, August 1.
King, Ryan, Marc Mauer, and Tracy Huling. 2003. “Big Prisons, Small Towns: Prison
Economics in Rural America.” Washington, DC: The Sentencing Project.
Michigan Department of Corrections. 2001. “2000 Statistical Report.” Lansing, MI: Michigan
Department of Corrections.
———. 2003. http://www.michigan.gov/corrections. [Accessed February 24, 2003; April 3,
2003; May 27, 2003.]
Missouri Department of Corrections. 2003. http://www.corrections.state.mo.us. [Accessed March
25, 2003; April 2, 2003; May 27, 2003.]
National Association of Counties. 2003. http://www.naco.org. [Accessed May - December
2003.]
New York State Department of Correctional Services. 2003. Information from website
http://www.docs.state.ny.us. [Accessed February 24, 2003.]
Ohio Department of Rehabilitation and Correction. 2001. “Monthly Progress Report for October
2001.”http://www.drc.state.oh.us. [Accessed February 24, 2003.]
——. 2003. “February 2003 Facts” http://www.drc.state.oh.us. [Accessed February 24, 2003.]
Ohio Department of Rehabilitation and Correction, Bureau of Research/Office of Policy. 2001.
“Institutional Census, July 1, 2001.” Columbus, OH: Ohio Department of Rehabilitation and
Correction
Schlosser, Eric. 1998. “The Prison-Industrial Complex.” The Atlantic Monthly. December.
Schuerman, Matthew. 2004. “The Big Idea: Down for the (Re)count.” City Limits Monthly.
January.
Texas Department of Criminal Justice. 2001. “Texas Department of Criminal Justice Statistical
Report, Fiscal Year 2000.” Huntsville, TX: Texas Department of Criminal Justice.
———. 2003. Electronic mail communication with Frank Anzian, Executive Services, March
28.
———. 2003. “Texas Department of Criminal Justice Statistical Report Fiscal Year 2002.”
Huntsville, TX: Texas Department of Criminal Justice.
The New Landscape of Imprisonment

45

Uggen, Christopher, and Jeff Manza. 2000. “The Political Consequences of Felon
Disfranchisement Laws in the United States.” Paper presented at the American Sociological
Association, Washington, DC, August 16.
U.S. Department of Agriculture. 2003. Data on county-level population change from 1990 to
2000. http://www.ers.usda.gov. [Accessed May 27, 2003.]
———. 2003. “Measuring Rurality: New Definitions in 2003.” http://www.ers.usda.gov.
[Accessed December, 1 2003.]
———. 2003. “United States State Fact Sheet.” http://www.ers.usda.gov. [Accessed December
1, 2003.]
U.S. Department of Agriculture, Economic Research Service. 2003. “Rural America at a
Glance.” Rural Development Research Report. 97-1(September).
U.S. Department of Commerce, U.S. Census Bureau. 1983. “Historical Metropolitan Area
Definitions.” http://www.census.gov/population/www/estimates/pastmetro.html.
———. 1983. “Metropolitan areas and Components, 1983 with FIPS Codes.”
http://www.census.gov/population/estimates/metro-city/83mfips.txt.
———. 2000. “Facts About Census 2000 Residence Rules.” http://www.census.gov.
———. 2001. “Data Collection and Processing Procedures for Census 2000.” Washington, DC:
U.S. Census Bureau.
———. 2001. “Population Change and Distribution: 1990 to 2000.” Census 2000 Brief.
Washington, DC: US Department of Commerce.
———. 2004. Census Source Tabulation Matrices for 1980, 1990, 2000.
http://www.census.gov.
———. 2004. Decennial Census data, 1980, 1990, and 2000. http://www.census.gov.
U.S. Department of Education, National Center for Education Statistics. “Digest of Education
Statistics, 2001.” http://www.nces.ed.gov. [Accessed on April 29, 2003.]
U.S. Department of Justice, Bureau of Justice Statistics. 1975. “Census of Correctional Facilities:
1974.” Washington, DC: Bureau of Justice Statistics.
———. 1986. “Historical Corrections Statistics in the United States: 1850–1984.” Washington,
DC: Bureau of Justice Statistics.
———. 1997. “Census of State Adult Correctional Facilities, 1979 [Computer file].” Conducted
by U.S. Dept. of Commerce, Bureau of the Census. 2nd ICPSR ed. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research.
———. 1998a. “Census of State Adult Correctional Facilities, 1995 [Computer file].”
Conducted by U.S. Dept. of Commerce, Bureau of the Census. ICPSR ed. Ann Arbor, MI: Interuniversity Consortium for Political and Social Research.
———. 1998b. “Sourcebook of Criminal Justice Statistics.” http://www.albany.edu/sourcebook/
[Accessed December 2003.]
———. 2000. “Sourcebook of Criminal Justice Statistics.” http://www.albany.edu/sourcebook/
[Accessed December 2003.]

46 The New Landscape of Imprisonment

———. 2001. “Prison and Jail Inmates at Midyear 2001.” Washington, DC: Bureau of Justice
Statistics.
———. 2001. “Prisoners in 2000.” Washington, DC: Bureau of Justice Statistics.
———. 2002. “Prison and Jail Inmates at Midyear 2001.” Washington, DC: Bureau of Justice
Statistics.
———. 2003. “Census of State and Federal Correctional Facilities, 2000.” Washington, DC:
Bureau of Justice Statistics.
U.S. Department of Justice, Federal Bureau of Prisons. 1991. “State of the Bureau 1991.”
http://www.bop.gov. [Accessed March 20,2003.]
———. 1994. “State of the Bureau 1994.” http://www.bop.gov. [Accessed March 20,2003.].
———. 1995. “State of the Bureau 1995.” http://www.bop.gov. [Accessed March 20,2003.]
———. 1996. “State of the Bureau 1996.” http://www.bop.gov. [Accessed March 20,2003.]
———. 1997. “State of the Bureau 1997.” http://www.bop.gov. [Accessed March 20,2003.]
———. 1998. “State of the Bureau 1998.” http://www.bop.gov. [Accessed March 20,2003.]
———. 1999. “State of the Bureau 1998.” http://www.bop.gov. [Accessed March 20,2003.]
U.S. General Accounting Office. 2003. “Formula Grants: 2000 Census Redistributes Federal
Funding among States.” Washington, DC: U.S. General Accounting Office.
U.S. Office of Management and Budget. 2000. “Standards for Defining Metropolitan and
Micropolitan Statistical Areas.” Federal Register. 65 (249): 82228
U.S. Office of Management and Budget. 2003: “OMB Bulletin No. 03-04”
http://www.whitehouse.gov/omb/bulletins/b03-04.html.
U.S. Representative Mark Green. 1999. “Green Introduces Bill. New Proposal to Count State
Prisoners Will Help Ensure Wisconsin Gets Fair Share of Federal Funds.” Press release,
Washington, DC, April 29.
Wagner, Peter. 2002. “Detaining for Dollars: Federal Aid Follows Inner-City Prisoners to Rural
Town Coffers.” Springfield, MA: Prison Policy Initiative.
———. 2002. “Importing Constituents: Prisoners and Political Clout in New York.” Springfield,
MA: Prison Policy Initiative.

The New Landscape of Imprisonment

47

Appendices

48 The New Landscape of Imprisonment

Appendix A. Types of facilities in state and federal correctional
systems

California:

State prisons, mother/infant programs, and conservation camps are included.
Community corrections centers are not included.

Colorado:

Correctional facilities and centers, private prisons, and diagnostic units are
included.

Federal:

Correctional complexes, correctional institutions, medical centers, transfer centers,
and U.S. penitentiaries are included. Community corrections management centers,
detention centers, transfer centers, metropolitan correctional centers, and
metropolitan detention centers are not included.

Florida:

Correctional institutions, work camps, forestry camps, road prisons, confinement
drug treatment facilities, and one mental health facility are included. Work release
centers and community corrections centers are not included.

Georgia:

State correctional institutions are included. Transitional centers, county camps, and
prison boot camps are not included.

Illinois:

Correctional centers, incarceration programs, minimal security units, and work
camps are included. Community corrections centers or Adult Transition Centers
(ATCs) are not included.

Michigan:

Correctional facilities, technical rule violator centers, and boot camps are included.
Correctional centers and community residential programs are not included.

Missouri:

Correctional centers, diagnostic centers, therapeutic community centers, and
treatment centers are included. Community release centers are not included.

New York: Correctional facilities, shock incarceration facility, camps, and Alcohol and
Substance Abuse Counseling Treatment Centers (ASACTC) are included.

Ohio:

Correctional facilities and institutions, medical centers, and prerelease centers are
included.

Texas:

Prison prerelease facilities, Substance Abuse Felony Punishment Facility (SAFPF)
state jails, private prisons, and medical, psychiatric, and transfer facilities are
included. Intermediate sanction facilities are not included.

The New Landscape of Imprisonment

49

Appendix B. Share of counties and share of residents in non-metro
counties by state

Percent of
counties

Percent of
residents

1979

2000

1979

2000

California

47

38

5

2.8

Colorado

84

61

19

9

Florida

54

27

10

2.3

Georgia

76

38

38

10

Illinois

75

36

18

5

Michigan

73

41

19

8

Missouri

85

50

34

14

New York

44

18

10

2.3

Ohio

59

22

21

4

Texas

81

52

21

7

Average

72

41

16

5

50 The New Landscape of Imprisonment

Appendix C. State-level data profiles
California

1979

2000

Change

Metro counties

# of prisons:

30

83

177%

Georgia

1979

2000

Change

In non-metro counties

12

23

92%

# of prisons:

18

42

133%

In metro counties

18

60

233%

In non-metro counties

10

14

40%

In metro counties

8

28

250%

% of prisons:

31

49

In non-metro counties

40%

28%

% of prisons:

In metro counties

60%

72%

In non-metro counties

56%

33%

% of counties w/ 1+
prisons:
Non-metro counties

34%

59%

In metro counties

44%

67%

26%

64%

10%

18%

Metro counties

42%

56%

% of counties w/ 1+
prisons:
Non-metro counties

7%

18%

% of residents in nonmetro
% of prisoners in nonmetro
Total # counties:

5%

3%

Metro counties

21%

18%

34%

18%

38%

10%

58

58

56%

39%

Non-metro counties

27

22

% of residents in nonmetro
% of prisoners in nonmetro
Total # counties:

159

159

Metro counties

31

Non-metro counties

121

Metro counties

38

36

61
98

Colorado

1979

2000

Change

# of prisons:

7

32

357%

Illinois

1979

2000

Change

In non-metro counties

6

11

83%

# of prisons:

12

40

233%

In metro counties

1

21

2000%

In non-metro counties

8

12

50%

In metro counties

4

28

600%

% of prisons:
In non-metro counties

86%

34%

% of prisons:

In metro counties

14%

66%

In non-metro counties

67%

30%

% of counties w/ 1+ prisons:

8%

22%

In metro counties

33%

70%

% of counties w/ 1+
prisons:
Non-metro counties

7%

28%

8%

24%

Metro counties

4%

31%

Non-metro counties

7%

21%

Metro counties

10%

24%

% of residents in non-metro

19%

9%

% of prisoners in non-metro

97%

27%

Total # counties:

64

64

Non-metro counties

54

39

Metro counties

10

25

Florida

1979

2000

Change

# of prisons:

39

84

115%

In non-metro counties

18

28

56%

In metro counties

21

56

167%

% of prisons:
In non-metro counties

46%

33%

In metro counties

54%

67%

% of counties w/ 1+ prisons:

45%

78%

Non-metro counties

39%

89%

Metro counties

52%

73%

% of residents in non-metro

10%

2.3%

% of prisoners in non-metro

50%

35%

Total # counties:

67

67

Non-metro counties

36

18

% of residents in non-metro

18%

5%

% of prisoners in non-metro

69%

33%

Total # counties:

102

102

Non-metro counties

76

Metro counties

26

65

Michigan

1979

2000

Change

# of prisons:

25

60

140%

In non-metro counties

13

16

23%

In metro counties

12

44

267%

37

% of prisons:
In non-metro counties

52%

27%

In metro counties

48%

73%

% of counties w/ 1+
prisons:
Non-metro counties

19%

36%

16%

38%

Metro counties

27%

35%

% of residents in non-metro

19%

8%

% of prisoners in non-metro

41%

14%

The New Landscape of Imprisonment

51

Metro counties

Total # counties:

83

83

Non-metro counties

61

34

Metro counties

22

49

Missouri

1979

2000

Change

# of prisons:

7

26

271%

In non-metro counties

7

6

-14%

In metro counties

0

35

69

Texas

1979

2000

Change

# of prisons:

17

137

706%

In non-metro counties

8

36

350%

20

In metro counties

9

101

1022%

23%

In non-metro counties

47%

26%

In metro counties

53%

74%

In metro counties

100
%
0%

77%

28%

3%

16%

Non-metro counties

3%

9%

% of counties w/ 1+
prisons:
Non-metro counties

3%

% of counties w/ 1+ prisons:

2%

22%

Metro counties

6%

22%

Metro counties

6%

34%

% of residents in non-metro

34%

14%

% of residents in non-metro

21%

7%

% of prisoners in non-metro

24%

% of prisoners in non-metro

67%

24%

Total # counties:

254

254

Total # counties:

100
%
115

115

Non-metro counties

205

Non-metro counties

98

57

Metro counties

49

121

Metro counties

17

58
Ten-state totals

1979

2000

Change

New York

1979

2000

Change

# of prisons:

195

604

210%

# of prisons:

30

65

117%

In non-metro counties

97

158

63%

In non-metro counties

12

11

-8%

In metro counties

98

446

355%

In metro counties

18

54

200%

% of prisons:
In non-metro counties

50%

26%

In non-metro counties

40%

17%

In metro counties

50%

74%

In metro counties

60%

83%

31%

32%

52%

% of counties w/ 1+
prisons:
Non-metro counties

13%

% of counties w/ 1+
prisons:
Non-metro counties

9%

26%

37%

55%

Metro counties

22%

35%

Metro counties

29%

51%

% of residents in non-metro

16%

5%

% of residents in non-metro

10%

2.3%

% of prisoners in non-metro

52%

23%

% of prisoners in non-metro

41%

15%

Total # counties:

1,052

1,052

Total # counties:

62

62

Non-metro counties

758

431

Metro counties

294

621

% of prisons:

% of prisons:
In non-metro counties

% of prisons:

Non-metro counties

27

11

Metro counties

35

51

Ohio

1979

2000

Change

# of prisons:

10

35

250%

In non-metro counties

3

1

-67%

In metro counties

7

34

386%

% of prisons:
In non-metro counties

30%

3%

In metro counties

70%

97%

% of counties w/ 1+
prisons:
Non-metro counties

10%

25%

6%

5%

Metro counties

17%

30%

% of residents in non-metro

21%

4%

% of prisoners in non-metro

37%

6%

Total # counties:

88

88

Non-metro counties

53

19

52 The New Landscape of Imprisonment

133